diabetes management research papers

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Pharmacological Reviews

New Aspects of Diabetes Research and Therapeutic Development

Both type 1 and type 2 diabetes mellitus are advancing at exponential rates, placing significant burdens on health care networks worldwide. Although traditional pharmacologic therapies such as insulin and oral antidiabetic stalwarts like metformin and the sulfonylureas continue to be used, newer drugs are now on the market targeting novel blood glucose–lowering pathways. Furthermore, exciting new developments in the understanding of beta cell and islet biology are driving the potential for treatments targeting incretin action, islet transplantation with new methods for immunologic protection, and the generation of functional beta cells from stem cells. Here we discuss the mechanistic details underlying past, present, and future diabetes therapies and evaluate their potential to treat and possibly reverse type 1 and 2 diabetes in humans.

Significance Statement

Diabetes mellitus has reached epidemic proportions in the developed and developing world alike. As the last several years have seen many new developments in the field, a new and up to date review of these advances and their careful evaluation will help both clinical and research diabetologists to better understand where the field is currently heading.

I. Introduction

Diabetes mellitus, a metabolic disease defined by elevated fasting blood glucose levels due to insufficient insulin production, has reached epidemic proportions worldwide (World Health Organization, 2020 ). Type 1 and type 2 diabetes (T1D and T2D, respectively) make up the majority of diabetes cases with T1D characterized by autoimmune destruction of the insulin-producing pancreatic beta cells. The much more prevalent T2D arises in conjunction with peripheral tissue insulin resistance and beta cell failure and is estimated to increase to 21%–33% of the US population by the year 2050 (Boyle et al., 2010 ). To combat this growing health threat and its cardiac, renal, and neurologic comorbidities, new and more effective diabetes drugs and treatments are essential. As the last several years have seen many new developments in the field of diabetes pharmacology and therapy, we determined that a new and up to date review of these advances was in order. Our aim is to provide a careful evaluation of both old and new therapies ( Fig. 1 ) in a manner that we hope will be of interest to both clinical and bench diabetologists. Instead of the usual encyclopedic approach to this topic, we provide here a targeted and selective consideration of the underlying issues, promising new treatments, and a re-examination of more traditional approaches. Thus, we do not discuss less frequently used diabetes agents, such as alpha-glucosidase inhibitors; these were discussed in other recent reviews (Hedrington and Davis, 2019 ; Lebovitz, 2019 ).

Pharmacologic targeting of numerous organ systems for the treatment of diabetes. Treatment of diabetes involves targeting of various organ systems, including the kidney by SGLT2 inhibitors; the liver, gut, and adipose tissue by metformin; and direct actions upon the pancreatic beta cell. Beta cell compounds aim to increase secretion or mass and/or to protect from autoimmunity destruction. Ultimately, insulin therapy remains the final line of diabetes treatment with new technologies under development to more tightly regulate blood glucose levels similar to healthy beta cells. hESC, human embryonic stem cell.

II. Diabetes Therapies

A. metformin.

Metformin is a biguanide originally based on the natural product galegine, which was extracted from the French lilac (Bailey, 1992 ; Rojas and Gomes, 2013 ; Witters, 2001 ). A closely related biguanide, phenformin, was also used initially for its hypoglycemic actions. Based on its successful track record as a safe, effective, and inexpensive oral medication, metformin has become the most widely prescribed oral agent in the world in treating T2D (Rojas and Gomes, 2013 ; He and Wondisford, 2015 ; Witters, 2001 ), whereas phenformin has been largely bypassed due to its unacceptably high association with lactic acidosis (Misbin, 2004 ). Unlike sulfonylureas, metformin lowers blood glucose without provoking hypoglycemia and improves insulin sensitivity (Bailey, 1992 ). Despite these well known beneficial metabolic actions, metformin’s mechanism of action and even its main target organ remain controversial. In fact, metformin has multiple mechanisms of action at the organ as well as the cellular level, which has hindered our understanding of its most important molecular effects on glucose metabolism (Witters, 2001 ). Adding to this, a specific receptor for metformin has never been identified. Metformin has actions on several tissues, although the primary foci of most studies have been the liver, skeletal muscle, and the intestine (Foretz et al., 2014 ; Rena et al., 2017 ). Metformin and phenformin clearly suppress hepatic glucose production and gluconeogenesis, and they improve insulin sensitivity in the liver and elsewhere (Bailey, 1992 ). The hepatic actions of metformin have been the most exhaustively studied to date, and there is little doubt that these actions are of some importance. However, several of the studies remain highly controversial, and there are still open questions.

One of the first reported specific molecular targets of metformin was mitochondrial complex I of the electron transport chain. Inhibition of this complex results in reduced oxidative phosphorylation and consequently decreased hepatic ATP production (El-Mir et al., 2008 ; Evans et al., 2005 ; Owen et al., 2000 ). As is the case in many other studies of metformin, however, high concentrations of the drug were found to be necessary to depress metabolism at this site (El-Mir et al., 2000 ; He and Wondisford, 2015 ; Owen et al., 2000 ). Also controversial is whether metformin works by activating 5′ AMP-activated protein kinase (AMPK), a molecular energy sensor that is known to be a major metabolic sensor in cells, or if not AMPK directly, then one of its upstream regulators such as liver kinase B2 (Zhou et al., 2001 ). Although metformin was shown to activate AMPK in several excellent studies, other studies directly contradicted the AMPK hypothesis. Most dramatic were studies showing that metformin’s actions to suppress hepatic gluconeogenesis persisted despite genetic deletion of the AMPK’s catalytic domain (Foretz et al., 2010 ). More recent studies identified additional or alternative targets, such as cAMP signaling in the liver (Miller et al., 2013 ) or glycogen synthase kinase-3 (Link, 2003 ). Other work showed that the phosphorylation of acetyl-CoA carboxylase and acetyl-CoA carboxylase 2 are involved in regulating lipid homeostasis and improving insulin sensitivity after exposure to metformin (Fullerton et al., 2013 ).

Although there are strong data to support each of these pathways, it is not entirely clear which signaling pathway(s) is most essential to the actions of metformin in hepatocytes. Metformin clearly inhibits complex I and concomitantly decreases ATP and increases AMP. The latter results in AMPK activation, reduced fatty acid synthesis, and improved insulin receptor activation, and increased AMP has been shown to inhibit adenylate cyclase to reduce cAMP and thus protein kinase A activation. Downstream, this reduces the expression of phosphoenolpyruvate carboxykinase and glucose 6-phosphatase via decreased cAMP response element-binding protein, the cAMP-sensitive transcription factor. Decreased PKA also promotes ATP-dependent 6-phosphofructokinase, liver type activity via fructose 2,6-bisphosphate and reduces gluconeogenesis, as fructose-bisphosphatase 1 is inhibited by fructose 2,6-bisphosphate, along with other mechanisms (Rena et al., 2017 ; Pernicova and Korbonits, 2014 ).

More recent work has shown that metformin at pharmacological rather than suprapharmacological doses increases mitochondrial respiration and complex 1 activity and also increases mitochondrial fission, now thought to be critical for maintaining proper mitochondrial density in hepatocytes and other cells. This improvement in respiratory activity occurs via AMPK activation (Wang et al., 2019 ).

Although the liver has historically been the major suspected site of metformin action, recent studies have suggested that the gut instead of the liver is a major target, a concept supported by the increased efficacy of extended-release formulations of metformin that reside for a longer duration in the gut after their administration (Buse et al., 2016 ). An older, but in our view an important observation, is that the intravenous administration of metformin has little or no effect on blood glucose, whereas, in contrast, orally administered metformin is much more effective (Bonora et al., 1984 ). Recent imaging studies using labeled glucose have shown directly that metformin stimulates glucose uptake by the gut in patients with T2D to reduce plasma glucose concentrations (Koffert et al., 2017 ; Massollo et al., 2013 ). Additionally, it is possible that metformin may exert its effect in the gut by inducing intestinal glucagon-like peptide-1 (GLP-1) release (Mulherin et al., 2011 ; Preiss et al., 2017) to potentiate beta cell insulin secretion and by stimulating the central nervous system (CNS) to exert control over both blood glucose and liver function. Indeed, CNS effects produced by metformin have been proposed to occur via the local release of GLP-1 to activate intestinal nerve endings of ascending nerve pathways that are involved in CNS glucose regulation (Duca et al., 2015 ). Lastly, several papers have now implicated that metformin may act by altering the gut microbiome, suggesting that changes in gut flora may be critical for metformin’s actions (McCreight et al., 2016 ; Wu et al., 2017 ; Devaraj et al., 2016 ). A new study proposed that activation of the intestinal farnesoid X receptor may be the means by which microbiota alter hyperglycemia (Sun et al., 2018 ). However, these studies will require more mechanistic detail and confirmation before they can be fully accepted by the field. In addition to the action of metformin on gut flora, the production of imidazole propionate by gut microbes in turn has been shown to interfere with metformin action through a p38-dependent mechanism and AMPK inhibition. Levels of imidazole propionate are especially higher in patients with T2D who are treated with metformin (Koh et al., 2020 ).

In summary, the combined contribution of these various effects of metformin on multiple cellular targets residing in many tissues may be key to the benefits of metformin treatment on lowering blood glucose in patients with type 2 diabetes (Foretz et al., 2019 ). In contrast, exciting new work showing metformin leads to weight loss by increasing circulating levels of the peptide hormone growth differentiation factor 15 and activation of brainstem glial cell-derived neurotropic factor family receptor alpha like receptors to reduce food intake and energy expenditure works independently of metformin’s glucose-lowering effect (Coll et al., 2020 ).

B. Sulfonylureas and Beta Cell Burnout

The class of compounds known as sulfonylureas includes one of the oldest oral antidiabetic drugs in the pharmacopoeia: tolbutamide. Tolbutamide is a “first generation” oral sulfonylurea secretagogue whose clinical usefulness is due to its prompt stimulation of insulin release from pancreatic beta cells. “Second generation” sulfonylureas include drugs such as glyburide, gliclazide, and glipizide. Sulfonylureas act by binding to a high affinity sulfonylurea binding site, the sulfonylurea receptor 1 subunit of the K(ATP) channel, which closes the channel. These drugs mimic the physiologic effects of glucose, which closes the K(ATP) channel by raising cytosolic ATP/ADP. This in turn provokes beta cell depolarization, resulting in increased Ca 2+ influx into the beta cell (Ozanne et al., 1995 ; Ashcroft and Rorsman, 1989 ; Nichols, 2006 ). Importantly, sulfonylureas, and all drugs that directly increase insulin secretion, are associated with hypoglycemia, which can be severe, and which limits their widespread use in the clinic (Yu et al., 2018 ). Meglitinides are another class of oral insulin secretagogues that, like the sulfonylureas, bind to sulfonylurea receptor 1 and inhibit K(ATP) channel activity (although at a different site of action). The rapid kinetics of the meglitinides enable them to effectively blunt the postprandial glycemic excursions that are a hallmark (along with elevated fasting glucose) of T2D (Rosenstock et al., 2004). However, the need for their frequent dosing (e.g., administration before each meal) has limited their appeal to patients.

The efficacy of sulfonylureas is known to decrease over time, leading to failure of the class for effective long-term treatment of T2D (Harrower, 1991 ). More broadly, it is now widely accepted that the number of functional beta cells in humans declines during the progression of T2D. Thus, one would expect that due to this decline, all manner of oral agents intended to target the beta cell and increase its cell function (and especially insulin secretion) will fail over time (RISE Consortium, 2019 ), a process referred to as “beta cell failure” (Prentki and Nolan, 2006 ). Currently, treatments that can expand beta cell mass or improve beta cell function or survival over time are not yet available for use in the clinic. As a result, treatments that may be able to help patients cope with beta cell burnout such as islet cell transplantation, insulin pumps, or stem cell therapy are alternatives that will be discussed below.

C. Ca 2+ Channel Blockers and Type 1 Diabetes

Strategies to treat and prevent T1D have historically focused on ameliorating the toxic consequences of immune dysregulation resulting in autoimmune destruction of pancreatic beta cells. More recently, a concerted focus on alleviating the intrinsic beta cell defects (Sims et al., 2020 ; Soleimanpour and Stoffers, 2013 ) that also contribute to T1D pathogenesis have been gaining traction at both the bench and the bedside. Several recent preclinical studies suggest that Ca 2+ -induced metabolic overload induces beta cell failure (Osipovich et al., 2020 ; Stancill et al., 2017 ; Xu et al., 2012 ), with the potential that excitotoxicity contributes to beta cell demise in both T1D and T2D, similar to the well known connection between excitotoxicity and, concomitantly, increased Ca 2+ loading of the cells and neuronal dysfunction. Indeed, the use of the phenylalkylamine Ca 2+ channel blocker verapamil has been successful in ameliorating beta cell dysfunction in preclinical models of both T1D and T2D (Stancill et al., 2017 ; Xu et al., 2012 ). Verapamil is a well known blocker of L-type Ca 2+ channels, and, in normally activated beta cells, it limits Ca 2+ entry into the beta cell (Ohnishi and Endo, 1981 ; Vasseur et al., 1987 ). This would be expected to, in turn, alter the expression of many Ca 2+ influx–dependent beta cell genes (Stancill et al., 2017 ), and the evidence to date suggests it is likely that verapamil preserves beta cell function in diabetes models by repressing thioredoxin-interacting protein (TXNIP) expression and thus protecting the beta cell. This is somewhat surprising given the physiologic role of Ca 2+ is to acutely trigger insulin secretion; this process would be expected to be inhibited by L-type Ca 2+ channel blockers (Ashcroft and Rorsman, 1989 ; Satin et al., 1995 ).

Hyperglycemia is a well known inducer of TXNIP expression, and a lack of TXNIP has been shown to protect against beta cell apoptosis after inflammatory stress (Chen et al., 2008a ; Shalev et al., 2002 ; Chen et al., 2008b ). Excitingly, the use of verapamil in patients with recent-onset T1D improved beta cell function and improved glycemic control for up to 12 months after the initiation of therapy, suggesting there is indeed promise for targeting calcium and TXNIP activation in T1D. Use of verapamil for a repurposed indication in the preservation of beta cell function in T1D is attractive due its well known safety profile as well as its cardiac benefits (Chen et al., 2009 ). Although the long-term efficacy of verapamil to maintain beta cell function in vivo is unclear, a recently described TXNIP inhibitor may also show promise in suppressing the hyperglucagonemia that also contributes to glucose intolerance in T2D (Thielen et al., 2020 ). As there is a clear need for increased Ca 2+ influx into the beta cell to trigger and maintain glucose-dependent insulin secretion (Ashcroft and Rorsman, 1990 ; Satin et al., 1995 ), it remains to be seen how well regulated insulin secretion is preserved in the presence of L-type Ca 2+ channel blockers like verapamil in the system. One might speculate that reducing but not fully eliminating beta cell Ca 2+ influx might reduce TXNIP levels while preserving enough influx to maintain glucose-stimulated insulin release. Alternatively, these two phenomena may operate on entirely different time scales. At present, these issues clearly will require further investigation.

D. GLP-1 and the Incretins

Studies dating back to the 1960s revealed that administering glucose in equal amounts via the peripheral circulation versus the gastrointestinal tract led to dramatically different amounts of glucose-induced insulin secretion (Elrick et al., 1964 ; McIntyre et al., 1964 ; Perley and Kipnis, 1967 ). Gastrointestinal glucose administration greatly increased insulin secretion versus intravenous glucose, and this came to be known as the “incretin effect” (Nauck et al., 1986a ; Nauck et al., 1986b ). Subsequent work showed that release of the gut hormone GLP-1 mediated this effect such that food ingestion induced intestinal cell hormone secretion. GLP-1 so released would then circulate to the pancreas via the blood to prime beta cells to secrete more insulin when glucose became elevated because these hormones stimulated beta cell cAMP formation (Drucker et al., 1987 ). The discovery that a natural peptide corresponding to GLP-1 could be found in the saliva of the Gila monster, a desert lizard, hastened progress in the field, and ample in vitro studies subsequently confirmed that GLP-1 potentiated insulin secretion in a glucose-dependent manner. GLP-1 has little or no significant action on insulin secretion in the absence of elevated glucose (such as might typically correspond to the postprandial case or during fasting), thus minimizing the likelihood of hypoglycemia provoked by GLP-1 in treated patients (Kreymann et al., 1987 ). Although not completely understood, the glucose dependence of GLP-1 likely reflects the requirement for adenine nucleotides to close glucose-inhibited K(ATP) channels and thus subsequently activate Ca 2+ influx–dependent insulin exocytosis. Besides potentiating GSIS at the level of the beta cell, glucagon-like peptide-1 receptor (GLP-1R) agonists also decrease glucagon secretion from pancreatic islet alpha cells, reduce gastric emptying, and may also increase beta cell proliferation, among other cellular actions (reviewed in Drucker, 2018 ; Muller et al., 2019).

Intense interest in the incretins by basic scientists, clinicians, and the pharma community led to the rapid development of new drugs for treating primarily T2D. These drugs include a range of GLP-1R agonists and inhibitors of the incretin hormone degrading enzyme dipeptidyl peptidase 4 (DPP4), whose targeting increases the half-lives of GLP-1 and gastric inhibitory polypeptide (GIP) and thereby increases protein hormone levels in plasma. GLP-1R agonists have been associated with not only a lowering of plasma glucose but also weight loss, decreased appetite, reduced risk of cardiovascular events, and other favorable outcomes (Gerstein et al., 2019; Hernandez et al., 2018; Husain et al., 2019; Marso et al., 2016a; Marso et al., 2016b ; Buse et al., 2004). Regarding their untoward actions, although hypoglycemia is not a major concern, there have been reports of pancreatitis and pancreatic cancer from use of GLP-1R agonists. However, a recent meta-analysis covering four large-scale clinical trials and over 33,000 participants noted no significantly increased risk for pancreatitis/pancreatic cancer in patients using GLP-1R agonists (Bethel et al., 2018).

Ongoing and future developments in the use of proglucagon-derived peptides such as GLP-1 and glucagon include the use of combined GLP-1/GIP, glucagon/GLP-1, and agents targeting all three peptides in combination (reviewed in Alexiadou and Tan, 2020 ). Although short-term infusions of GLP-1 with GIP failed to yield metabolic benefits beyond those seen with GLP-1 alone (Bergmann et al., 2019 ), several GLP-1/GIP dual agonists are currently in development and have shown promising metabolic results in clinical trials (Frias et al., 2017 ; Frias et al., 2020 ; Frias et al., 2018 ). At the level of the pancreatic islet, beneficial effects of dual GLP-1/GIP agonists may be related to imbalanced and biased preferences of these agonists for the gastric inhibitory polypeptide receptor over the GLP-1R (Willard et al., 2020 ) and possibly were not simply to dual hormone agonism in parallel. Dual glucagon/GLP-1 agonist therapy has also been shown to have promising metabolic effects in humans (Ambery et al., 2018 ; Tillner et al., 2019 ). Oxyntomodulin is a natural dual glucagon/GLP-1 receptor agonist and proglucagon cleavage product that is also secreted from intestinal enteroendocrine cells, which has beneficial effects on insulin secretion, appetite regulation, and body weight in both humans and rodents (Cohen et al., 2003 ; Dakin et al., 2001 ; Dakin et al., 2002 ; Shankar et al., 2018 ; Wynne et al., 2005 ). Interestingly, alpha cell crosstalk to beta cells through the combined effects of glucagon and GLP-1 is necessary to obtain optimal glycemic control, suggesting a potential pathway for therapeutic dual glucagon/GLP-1 agonism within the islets of patients with T2D (Capozzi et al., 2019a ; Capozzi et al., 2019b ). Although the early results appear promising, more studies will be necessary to better understand the mechanistic and clinical impacts of these multiagonist agents.

E. DPP4 Inhibitors

Inhibition of DPP4, the incretin hormone degrading enzyme, is one of the most common T2D treatments to increase GLP-1 and GIP plasma hormone levels. These DPP4 inhibitors or “gliptins” are generally used in conjunction with other T2D drugs such as metformin or sulfonylureas to obtain the positive benefits discussed above (Lambeir et al., 2008 ). DPP4 is a primarily membrane-bound peptidase belonging to the serine peptidase/prolyl oligopeptidase gene family, which cleaves a large number of substrates in addition to the incretin hormones (Makrilakis, 2019 ). DPP4 inhibitors provide glucose-lowering benefits while being generally well tolerated, and the variety of available drugs (including sitagliptin, saxagliptin, vildagliptin, alogliptin, and linagliptin) with slightly different dosing frequency, half-life, and mode of excretion/metabolism allows for use in multiple patient populations (Makrilakis, 2019 ). This includes the elderly and individuals with renal or hepatic insufficiency (Makrilakis, 2019 ).

Although hypoglycemia is not a concern for DPP4 inhibitor use, other considerations should be made. DPP4 inhibitors tend to be more expensive than metformin or other second-line oral drugs in addition to having more modest glycemic effects than GLP-1R agonists (Munir and Lamos, 2017 ). Finally, meta-analysis of randomized and observational studies concluded that heart failure in patients with T2D was not associated with use of DPP4 inhibitors; however, this study was limited by the short follow-up and lack of high-quality data (Li et al., 2016 ). Thus, the US Food and Drug Administration (FDA) did recommend assessing risk of heart failure hospitalization in patients with pre-existing cardiovascular disease, prior heart failure, and chronic kidney disease when using saxagliptin and alogliptin (Munir and Lamos, 2017 ).

F. Sodium Glucose Cotransporter 2 Inhibitors

A recent development in the field of T2D drugs are sodium glucose cotransporter 2 (SGLT2) inhibitors, which have an interesting and very different mechanism of action. Within the proximal tubule of the nephron, SGLT2 transports ingested glucose into the lumen of the proximal tubule between the epithelial layers, thereby reclaiming glucose by this reabsorption process (reviewed in Vallon, 2015 ). SGLT2 inhibitors target this transporter and increase glucose in the tubular fluid and ultimately increase it in the urine. In patients with diabetes, SGLT2 inhibition results in a lowering of plasma glucose with urine glucose content rising substantially (Adachi et al., 2000 ; Vallon, 2015 ). These drugs, although they are relatively new, have become an area of great interest for not only patients with T2D (Grempler et al., 2012 ; Imamura et al., 2012 ; Meng et al., 2008 ; Nomura et al., 2010 ) but also for patients with T1D (Luippold et al., 2012 ; Mudaliar et al., 2012 ). Part of their appeal also rests on reports that their use can lead to a statistically significant decline in cardiac events that are known to occur secondarily to diabetes, possibly independently of plasma glucose regulation (reviewed in Kurosaki and Ogasawara, 2013 ). Although the long-term consequences of their clinical use cannot yet be determined, raising the glucose content of the urogenital tract leads to an increased risk of urinary tract infections and other related infections in some patients (Kurosaki and Ogasawara, 2013 ).

Another recent concern about the use of SGLT2 inhibitors has been the development of normoglycemic diabetic ketoacidosis (DKA). Despite the efficacy of SGLT2 inhibitors, observations of hyperglucagonemia in patients with euglycemic DKA has led to a number of recent studies focused on SGLT2 actions on pancreatic islets. Initial studies of isolated human islets treated with small interfering RNA directed against SGLT2 and/or SGLT2 inhibitors demonstrated increased glucagon release. These studies were complemented by the finding of elevations in glucagon release in mice that were administered SGLT2 inhibitors in vivo (Bonner et al., 2015 ). Insights into the possible mechanistic links between SGLT2 inhibition, DKA frequency, and glucagon secretion in humans may relate to the observation of heterogeneity in SGLT2 expression, as SGLT2 expression appears to have a high frequency of interdonor and intradonor variability (Saponaro et al., 2020 ). More recently, both insulin and GLP-1 have been demonstrated to modulate SGLT2-dependent glucagon release through effects on somatostatin release from delta cells (Vergari et al., 2019 ; Saponaro et al., 2019 ), suggesting potentially complex paracrine effects that may affect the efficacy of these compounds.

On the other hand, several recent studies question that the development of euglycemic DKA after SGLT2 inhibitor therapy may be through alpha cell–dependent mechanisms. Three recent studies found no effect of SGLT2 inhibitors to promote glucagon secretion in mouse and/or rat models and could not detect SGLT2 expression in human alpha cells (Chae et al., 2020 ; Kuhre et al., 2019 ; Suga et al., 2019 ). A fourth study demonstrated only a brief transient effect of SGLT2 inhibition to raise circulating glucagon concentrations in immunodeficient mice transplanted with human islets, which returned to baseline levels after longer exposures to SGLT2 inhibitors (Dai et al., 2020 ). Furthermore, SGLT2 protein levels were again undetectable in human islets (Dai et al., 2020 ). These results could suggest alternative islet-independent mechanisms by which patients develop DKA, including alterations in ketone generation and/or clearance, which underscore the additional need for further studies both in molecular models and at the bedside. Nevertheless, SGLT2 inhibitors continue to hold promise as a valuable therapy for T2D, especially in the large segment of patients who also have superimposed cardiovascular risk (McMurray et al., 2019; Wiviott et al., 2019; Zinman et al., 2015).

G. Thiazolidinediones

Once among the most commonly used oral agents in the armamentarium to treat T2D, thiazolidinediones (TZDs) were clinically popular in their utilization to act specifically as insulin sensitizers. TZDs improve peripheral insulin sensitivity through their action as peroxisome proliferator-activated receptor (PPAR) γ agonists, but their clinical use fell sharply after studies suggested a connection between cardiovascular toxicity with rosiglitazone and bladder cancer risk with pioglitazone (Lebovitz, 2019 ). Importantly, an FDA panel eventually removed restrictions related to cardiovascular risk with rosiglitazone in 2013 (Hiatt et al., 2013 ). Similarly, concerns regarding use of bladder cancer risk with pioglitazone were later abated after a series of large clinical studies found that pioglitazone did not increase bladder cancer (Lewis et al., 2015 ; Schwartz et al., 2015 ). However, usage of TZDs had already substantially decreased and has not since recovered.

Although concerns regarding edema, congestive heart failure, and fractures persist with TZD use, there have been several studies suggesting that TZDs protect beta cell function. In the ADOPT study, use of rosiglitazone monotherapy in patients newly diagnosed with T2D led to improved glycemic control compared with metformin or sulfonylureas (Kahn et al., 2006). Later analyses revealed that TZD-treated subjects had a slower deterioration of beta cell function than metformin- or sulfonylurea-treated subjects (Kahn et al., 2011). Furthermore, pioglitazone use improved beta cell function in the prevention of T2D in the ACT NOW study (Defronzo et al., 2013; Kahn et al., 2011). Mechanistically, it is unclear if TZDs lead to beneficial beta cell function through direct effects or through indirect effects of reduced beta cell demand due to enhanced peripheral insulin sensitivity. Indeed, a beta cell–specific knockout of PPAR γ did not impair glucose homeostasis, nor did it impair the antidiabetic effects of TZD use in mice (Rosen et al., 2003 ). However, other reports demonstrated PPAR-responsive elements within the promoters of both glucose transporter 2 and glucokinase that enhance beta cell glucose sensing and function, which could explain beta cell–specific benefits for TZDs (Kim et al., 2002 ; Kim et al., 2000 ). Furthermore, TZDs have been shown to improve beta cell function by upregulating cholesterol transport (Brunham et al., 2007 ; Sturek et al., 2010 ). Additionally, use of TZDs in the nonobese diabetic (NOD) mouse model of T1D augmented the beta cell unfolded protein response and prevented beta cell death, suggesting potential benefits for TZDs in both T1D and T2D (Evans-Molina et al., 2009 ; Maganti et al., 2016 ). With a now refined knowledge of demographics in which to avoid TZD treatment due to adverse effects, together with genetic approaches to identify candidates more likely to respond effectively to TZD therapy (Hu et al., 2019 ; Soccio et al., 2015 ), it remains to be seen if TZD therapy will return to more prominent use in the treatment of diabetes.

H. Insulin and Beyond: The Use of “Smart” Insulin and Closed Loop Systems in Diabetes Treatment

Due to recombinant DNA technology, numerous insulin analogs are now available in various forms ranging from fast acting crystalline insulin to insulin glargine; all of these analogs exhibit equally effective insulin receptor binding. Most are generated by altering amino acids in the B26–B30 region of the molecule (Kurtzhals et al., 2000 ). The American Diabetes Association delineates these insulins by their 1) onset or time before insulin reaches the blood stream, 2) peak time or duration of maximum blood glucose–lowering efficacy, and 3) the duration of blood glucose–lowering time. Insulin administration is independent of the residuum of surviving and/or functioning beta cells in the patient and remains the principal pharmacological treatment of both T1D and T2D. The availability of multiple types of delivery methods, i.e., insulin pens, syringes, pumps, and inhalants, provides clinicians with a solid and varied tool kit with which to treat diabetes. The downsides, however, are that 1) hypoglycemia is a constant threat, 2) proper insulin doses are not trivial to calculate, 3) compliance can vary especially in children and young adults, and 4) there can be side effects of a variety of types. Nonetheless, insulin therapy remains a mainstay treatment of diabetes.

To eliminate the downsides of insulin therapy, research in the past several decades has worked toward generating glucose-sensitive or “smart” insulin molecules. These molecules change insulin bioavailability and become active only upon high blood glucose using glucose-binding proteins such as concanavalin A, glucose oxidase to alter pH sensitivity, and phenylboronic acid (PBA), which forms reversible ester linkages with diol-containing molecules including glucose itself (reviewed in Rege et al., 2017 ). Indeed, promising recent studies included various PBA moieties covalently bonded to an acylated insulin analog (insulin detemir, which contains myristic acid coupled to Lys B29 ). The detemir allows for binding to serum albumin to prolong insulin’s half-life in the circulation, and PBA provided reversible glucose binding (Chou et al., 2015 ). The most promising of the PBA-modified conjugates showed higher potency and responsiveness in lowering blood glucose levels compared with native insulin in diabetic mouse models and decreased hypoglycemia in healthy mice, although the molecular mechanisms have not yet been determined (Chou et al., 2015 ).

An additional active area of research includes structurally defining the interaction between insulin and the insulin receptor ectodomain. Importantly, a major conformational change was discovered that may be exploited to impair insulin receptor binding under hypoglycemic conditions (Menting et al., 2013 ; Rege et al., 2017 ). Challenges in the design, testing, and execution of glucose-responsive insulins may be overcome by the adaptation of novel modeling approaches (Yang et al., 2020 ), which may allow for more rapid screening of candidate compounds.

Technologies have also progressed in the field of artificial pancreas design and development. Currently two “closed loop” systems are now available: Minimed 670G from Medtronic and Control-IQ from Tandem Diabetes Care. Both systems use a continuous glucose monitor, insulin pump, and computer algorithm to predict correct insulin doses and administer them in real time. Such algorithm systems also take into account insulin potency, the rate of blood glucose increase, and the patient’s heart rate and temperature to adjust insulin delivery levels during exercise and after a meal. In addition, so-called “artificial pancreas” systems have also been clinically tested, which use both insulin and glucagon and as such result in fewer reports of hypoglycemic episodes (El-Khatib et al., 2017 ). These types of systems will continue to become more popular as the development of room temperature–stable glucagon analogs continue, such as GVOKE by Xeris Pharmaceuticals (currently available in an injectable syringe) and Baqsimi, a nasally administered glucagon from Eli Lilly.

I. Present and Future Therapies: Beta Cell Transplantation, Replication, and Immune Protection

1. islet transplantation.

The idea to use pancreatic allo/xenografts to treat diabetes remarkably dates back to the late 1800s (Minkowski, 1892 ; Pybus, 1924 ; Williams, 1894 ). Before proceeding to the discovery of insulin (together with Best, MacLeod, and Collip), Frederick Banting also postulated the potential for transplantation of pancreatic tissue emulsions to treat diabetes in dog models in a notebook entry in 1921 (Bliss, 1982 ). Decades later, Paul Lacy, David Scharp, and colleagues successfully isolated intact functional pancreatic islets and transplanted them into rodent models (Kemp et al., 1973 ). These studies led to the initial proof of concept studies for humans, with the first successful islet transplant in a patient with T1D occurring in 1977 (Sutherland et al., 1978 ). A rapid expansion of islet transplantation, inspired by these original studies led to key observations of successfully prolonged islet engraftment by the “Edmonton protocol” whereby corticosteroid-sparing immunosuppression was applied, and islets from at least two allogeneic donors were used to achieve insulin independence (Shapiro et al., 2000 ). More recent work has focused on improving upon the efficiency and long-term engraftment of allogeneic transplants leading to more prolonged graft function (to the 5-year mark) and successful transplantation from a single islet donor (Hering et al., 2016; Hering et al., 2005 ; Rickels et al., 2013 ). Critical to these efforts to improve the success rate was the recognition that the earlier generation of immunosuppressive agents to counter tissue rejection was toxic to islets (Delaunay et al., 1997 ; Paty et al., 2002 ; Soleimanpour et al., 2010 ) and that more appropriate and less toxic agents were needed (Hirshberg et al., 2003 ; Soleimanpour et al., 2012 ).

Certainly, islet transplantation as a therapeutic approach for patients with T1D has been scrutinized due to several challenges, including (but not limited to) the lack of available donor supply to contend with demand, limited long-term functional efficacy of islet allografts, the potential for re-emergence of autoimmune islet destruction and/or metabolic overload-induced islet failure, and significant adverse effects of prolonged immunosuppression (Harlan, 2016 ). Furthermore, although islet transplantation is not currently available for individuals with T2D, simultaneous pancreas-kidney transplantation in T2D had similar favorable outcomes to simultaneous pancreas-kidney transplantation in T1D; therefore, islet-kidney transplantation may eventually be a feasible option to treat T2D, as patients will already be on immunosuppressors (Sampaio et al., 2011 ; Westerman et al., 1983 ). An additional significant obstacle is the tremendous expense associated with islet transplantation therapy. Indeed, the maintenance, operation, and utilization of an FDA-approved and Good Manufacturing Practice–compliant islet laboratory can lead to operating costs at nearly $150,000 per islet transplant, which is not cost effective for the vast majority of patients with T1D (Naftanel and Harlan, 2004 ; Wallner et al., 2016 ). At present, the focus has been to obtain FDA approval for islet allo-transplantation as a therapy for T1D to allow for insurance compensation (Hering et al., 2016; Rickels and Robertson, 2019 ). In the interim, the islet biology, stem cell, immunology, and bioengineering communities have continued the development of cell-based therapies for T1D by other approaches to overcome the challenges identified during the islet transplantation boom of the 1990s and 2000s.

2. Pharmacologic Induction of Beta Cell Replication

Besides transplantation, progress in islet cell biology and especially in developmental biology of beta cells over several decades raised the additional possibility that beta cell mass reduction in diabetes might be countered by increasing beta cell number through mitogenic means. A key method to expand pancreatic beta cell mass is through the enhancement of beta cell replication. Although the study of pancreatic beta cell replication has been an area of intense focus in the beta cell biology field for several decades, only recently has this seemed truly feasible. Seminal studies identified that human beta cells are essentially postmitotic, with a rapid phase of growth occurring in the prenatal period that dramatically tapers off shortly thereafter (Gregg et al., 2012 ; Meier et al., 2008 ). The plasticity of rodent beta cells is considerably higher than that of human beta cells (Dai et al., 2016 ), which has led to a renewed focus on validation of pharmacologic agents to enhance rodent beta cell replication using isolated and/or engrafted human islets (Bernal-Mizrachi et al., 2014 ; Kulkarni et al., 2012 ; Stewart et al., 2015 ). Indeed, a large percentage of agents that were successful when applied to rodent systems were largely unsuccessful at inducing replication in human beta cells (Bernal-Mizrachi et al., 2014 ; Kulkarni et al., 2012 ; Stewart et al., 2015 ). However, several recent studies have begun to make significant progress on successfully pushing human beta cells to replicate.

Several groups have reported successful human beta cell proliferation, both in vitro and in vivo, in response to inhibitors of the dual specificity tyrosine phosphorylation-regulated kinase 1A (DYRK1A). These inhibitors include harmine, INDY, GNF4877, 5-iodotubericidin, leucettine-42, TG003, AZ191, CC-401, and more specific, recently developed DYRK1A inhibitors (Ackeifi et al., 2020 ). Although DYRK1A is conclusively established as the important mediator of human beta cell proliferation, comprehensively determining other cellular targets and if additional gene inhibition amplifies the proliferative response is still in process. New evidence from Wang and Stewart shows dual specificity tyrosine phosphorylation-regulated kinase 1B to be an additional mitogenic target and also describes variability in the range of activated kinases within cells and/or levels of inhibition for the many DYRK1A inhibitors listed above (Ackeifi et al., 2020 ). Interestingly, opposite to these human studies, earlier mouse studies from the Scharfmann group demonstrated that Dyrk1a haploinsufficiency leads to decreased proliferation and loss of beta cell mass (Rachdi et al., 2014b ). In addition, overexpression of Dyrk1a in mice led to beta cell mass expansion with increased glucose tolerance (Rachdi et al., 2014a ).

Although important differences in beta cell proliferative capacity have been shown between human and rodent species, there are also significant differences in the mitogenic capacity of beta cells from juvenile, adult, and pregnant individuals. This demonstrates that proliferative stimuli appear to act within the complex islet, pancreas, and whole-body environments unique to each time point. For example, the administration of the hormones platelet-derived growth factor alpha or GLP-1 result in enhanced proliferation in juvenile human beta cells yet are ineffective in adult human beta cells (Chen et al., 2011 ; Dai et al., 2017 ). This has been shown to be due to a loss of platelet-derived growth factor alpha receptor expression as beta cells age but appears to be unrelated to GLP-1 receptor expression levels (Chen et al., 2011 ). Indeed, the GLP-1 receptor is highly expressed in adult beta cells, and GLP-1 secretion increases insulin secretion, as detailed previously; however, the induction of proliferative factors such as nuclear factor of activated T cells, cytoplasmic 1; forkhead box protein 1; and cyclin A1 is only seen in juvenile islets (Dai et al., 2017 ). Human studies using cadaveric pancreata from pregnant donors also showed increased beta cell mass, yet lactogenic hormones from the pituitary or placenta (prolactin, placental lactogen, or growth hormone) are unable to stimulate proliferation in human beta cells despite their ability to produce robust proliferation in mouse beta cells (reviewed in Baeyens et al., 2016 ). Experiments overexpressing mouse versus human signal transducer and activator of transcription 5, the final signaling factor inducing beta cell adaptation, in human beta cells allows for prolactin-mediated proliferation revealing fundamental differences in prolactin pathway competency in human (Chen et al., 2015 ). Overcoming the barrier of recapitulating human pregnancy’s effect on beta cells through isolating placental cells or blood serum during pregnancy may result in the discovery of a factor(s) that facilitates the increase in beta cell mass observed during human pregnancy.

Mechanisms that stimulate beta cell proliferation have also been discovered from studying genetic mutations that result in insulinomas, spontaneous insulin-producing beta cell adenomas. The most common hereditary mutation occurs in the multiple endocrine neoplasia type 1 (MEN1) gene. Indeed, administration of a MEN1 inhibitor in addition to a GLP-1 agonist (which cannot induce proliferation alone) is able to increase beta cell proliferation in isolated human islets through synergistic activation of KRAS proto-oncogene, GTPase downstream signals (Chamberlain et al., 2014 ). Interestingly, MEN1 mutations are uncommon in sporadic insulinomas, yet assaying genomic and epigenetic changes in a large cohort of non-MEN1 insulinomas found alterations in trithorax and polycomb chromatin modifying genes that were functionally related to MEN1 (Wang et al., 2017 ). Stewart and colleagues hypothesized that changes in histone 3 lysine 27 and histone 3 lysine 4 methylation status led to increased enhancer of zeste homolog 2 and lysine demethylase 6A, decreased cyclin-dependent kinase inhibitor 1C, and thereby increased beta cell proliferation, among other phenotypes. They also proposed that these findings help to explain why increased proliferation always occurs despite broad heterogeneity of mutations found between individual insulinomas (Wang et al., 2017 ).

Although factors that induce proliferation are continuing to be discovered, there are drawbacks that still limit their clinical application. Harmine and other DYRK1A inhibitors are not beta cell specific, nor have all their cellular targets been determined (Ackeifi et al., 2020 ). Targeting other pathways to induce human beta cell proliferation such as modulation of prostaglandin E2 receptors (i.e., inhibition of prostaglandin E receptor 3 alone or in combination with prostaglandin E receptor 4 activation) showed promising increases in proliferative rate yet suffers from the same lack of specificity (Carboneau et al., 2017 ). Induction of proliferation may also come at the expense of glucose sensing as in insulinomas, which have an increased expression of “disallowed genes” and alterations in glucose transporter and hexokinase expression (Wang et al., 2017 ). A further untoward consequence that must be avoided is the production of cancerous cells through unchecked proliferation. Finally, increasing beta cell mass through low rates of proliferation may increase the pool of functional insulin-secreting cells in T2D, but without additional measures, these beta cells will still ultimately be targeted for immune cell destruction in T1D.

3. Beta Cell Stress Relieving Therapies

Metabolic, inflammatory, and endoplasmic reticulum (ER) stress contribute to beta cell dysfunction and failure in both T1D and T2D. Although reduction of metabolic overload of beta cells by early exogenous insulin therapy or insulin sensitizers can temporarily reduce loss of beta cell mass/function early in diabetes, a focus on relieving ER and inflammatory stress is also of interest to preserve beta cell health.

ER stress is a well known contributor to beta cell demise both in T1D and T2D (Laybutt et al., 2007 ; Marchetti et al., 2007 ; Marhfour et al., 2012 ; Tersey et al., 2012 ) and a target of interest in the prevention of beta cell loss in both diseases. Preclinical studies suggest that the use of chemical chaperones, including 4-phenylbutyric acid and tauroursodeoxycholic acid (TUDCA), to alleviate ER stress improves beta cell function and insulin sensitivity in mouse models of T2D (Cnop et al., 2017 ; Ozcan et al., 2006 ). Furthermore, TUDCA has been shown to preserve beta cell mass and reduce ER stress in mouse models of T1D (Engin et al., 2013 ). Interestingly, TUDCA has shown promise at improving insulin action in obese nondiabetic human subjects, yet beta cell function and insulin secretion were not assessed (Kars et al., 2010 ). A clinical trial regarding the use of TUDCA for humans with new-onset T1D is also ongoing ( {"type":"clinical-trial","attrs":{"text":"NCT02218619","term_id":"NCT02218619"}} NCT02218619 ). However, a note of caution regarding use of ER chaperones is that they may prevent low level ER stress necessary to potentiate beta cell replication during states of increased insulin demand (Sharma et al., 2015 ), suggesting that the broad use of ER chaperone therapies should be carefully considered.

The blockade of inflammatory stress has long been an area of interest for treatments of both T1D and T2D (Donath et al., 2019 ; Eguchi and Nagai, 2017 ). Indeed, use of nonsteroidal anti-inflammatory drugs (NSAIDs), which block cyclooxygenase, have been observed to improve metabolic control in patients with diabetes since the turn of the 20th century (Williamson, 1901 ). Salicylates have been shown to improve insulin secretion and beta cell function in both obese human subjects and those with T2D (Fernandez-Real et al., 2008; Giugliano et al., 1985 ). However, another NSAID, salsalate, has not been shown to improve beta cell function while improving other metabolic outcomes (Kim et al., 2014 ; Penesova et al., 2015 ), possibly suggesting distinct mechanisms of action for anti-inflammatory compounds. The regular use of NSAIDs to enhance metabolic outcomes is also often limited to the tolerability of long-term use of these agents due to adverse effects. Recently, golilumab, a monoclonal antibody against the proinflammatory cytokine tumor necrosis factor alpha, was demonstrated to improve beta cell function in new-onset T1D, suggesting that targeting the underlying inflammatory milieu may have benefits to preserve beta cell mass and function in T1D (Quattrin et al., 2020). Taken together, both new and old approaches to target beta cell stressors still remain of long-term interest to improve beta cell viability and function in both T1D and T2D.

3. New Players to Induce Islet Immune Protection

Countless researchers have expended intense industry to determine T1D disease etiology and treatments focused on immunotherapy and tolerogenic methods. Multiple, highly comprehensive reviews are available describing these efforts (Goudy and Tisch, 2005 ; Rewers and Gottlieb, 2009 ; Stojanovic et al., 2017 ). Here we will focus on the protection of beta cells through programmed cell death protein-1 ligand (PD-L1) overexpression, major histocompatibility complex class I, A, B, C (HLA-A,B,C) mutated human embryonic stem cell–derived beta cells, and islet encapsulation methods.

Cancer immunotherapies that block immune checkpoints are beneficial for treating advanced stage cancers, yet induction of autoimmune diseases, including T1D, remains a potential side effect (Stamatouli et al., 2018 ; Perdigoto et al., 2019 ). A subset of these drugs target either the programmed cell death-1 protein on the surface of activated T lymphocytes or its receptor PD-L1 (Stamatouli et al., 2018 ; Perdigoto et al., 2019 ). PD-L1 expression was found in insulin-positive beta cells from T1D but not insulin-negative islets or nondiabetic islets, leading to the hypothesis that PD-L1 is upregulated in an attempt to drive immune cell attenuation (Osum et al., 2018 ; Colli et al., 2018 ). Adenoviral overexpression of PD-L1 specifically in beta cells rescued hyperglycemia in the NOD mouse model of T1D, but these animals eventually succumbed to diabetes by the study’s termination (El Khatib et al., 2015 ). A more promising report from Ben Nasr et al. ( 2017 ) demonstrated that pharmacologically or genetically induced overexpression of PD-L1 in hematopoietic stem and progenitor cells inhibited beta cell autoimmunity in the NOD mouse as well as in vitro using human hematopoietic stem and progenitor cells from patients with T1D.

As mentioned above, islet transplantation to treat T1D is limited by islet availability, cost, and the requirement for continuous immunosuppression. Islet cells generated by differentiating embryonic or induced pluripotent stem (iPS) cells could circumvent these limitations. Ideally, iPS-derived beta cells could be manipulated to eliminate the expression of polymorphic HLA-A,B,C molecules, which were found to be upregulated in T1D beta cells (Bottazzo et al., 1985 ; Richardson et al., 2016 ). These molecules allow peptide presentation to CD8+ T cells or cytotoxic T lymphocytes and may lead to beta cell removal. Interestingly, remaining insulin-positive cells in T1D donor pancreas are not HLA-A,B,C positive (Nejentsev et al., 2007; Rodriguez-Calvo et al., 2015 ). However, current differentiation protocols are still limited in their ability to produce fully glucose-responsive beta cells without transplantation into animal models to induce mature characteristics. Additionally, use of iPS-derived beta cells will still lead to concerns regarding DNA mutagenesis resulting from the methods used to obtain pluripotency or teratoma formation from cells that have escaped differentiation.

Encapsulation devices would protect islets or stem cells from immune cell infiltration while allowing for the proper exchange of nutrients and hormones. Macroencapsulation uses removable devices that would help assuage fears surrounding mutation or tumor formation; indeed, the first human trial using encapsulated hESC-derived beta cells will be completed in January 2021 ( {"type":"clinical-trial","attrs":{"text":"NCT02239354","term_id":"NCT02239354"}} NCT02239354 ). Macroencapsulation of islets prior to transplantation using various alginate-based hydrogels has historically been impeded by a strong in vivo foreign body immune response (Desai and Shea, 2017 ; Doloff et al., 2017 ; Pueyo et al., 1993 ). More recently, chemically modified forms of alginate that avoid macrophage recognition and fibrous deposition have been successfully used in rodents and for up to 6 months in nonhuman primates (Vegas et al., 2016 ). Indeed, Bochenek et al. ( 2018 ) successfully transplanted alginate protected islets for 4 months without immunosuppression in the bursa omentalis of nonhuman primates demonstrating the feasibility for this approach to be extended to humans. It remains to be seen if these devices will be successful for long-term use, perhaps decades, in patients with diabetes.

III. Summary

Although existing drug therapies using classic oral antidiabetic drugs like sulfonylureas and metformin or injected insulin remain mainstays of diabetes treatment, newer drugs based on incretin hormone actions or SGLT2 inhibitors have increased the pharmacological armamentarium available to diabetologists ( Fig. 1 ). However, the explosion of progress in beta cell biology has identified potential avenues that can increase beta cell mass in sophisticated ways by employing stem cell differentiation or enhancement of beta cell proliferation. Taken together, there should be optimism that the increased incidence of both T1D and T2D is being matched by the creativity and hard work of the diabetes research community.

Abbreviations

Authorship contributions.

Wrote and contributed to the writing of the manuscript: Satin, Soleimanpour, Walker

This work was supported by the National Institutes of Health National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) [Grant R01-DK46409] (to L.S.S.), [Grant R01-DK108921] (to S.A.S.), and [Grant P30-DK020572 pilot and feasibility grant] (to S.A.S.), the Juvenile Diabetes Research Foundation (JDRF) [Grant CDA-2016-189] (to L.S.S. and S.A.S.), [Grant SRA-2018-539] (to S.A.S.), and [Grant COE-2019-861] (to S.A.S.), and the US Department of Veterans Affairs [Grant I01 BX004444] (to S.A.S.). The JDRF Career Development Award to S.A.S. is partly supported by the Danish Diabetes Academy and the Novo Nordisk Foundation.

https://doi.org/10.1124/pharmrev.120.000160

• Ackeifi C, Swartz E, Kumar K, Liu H, Chalada S, Karakose E, Scott DK, Garcia-Ocaña A, Sanchez R, DeVita RJ, et al. (2020) Pharmacologic and genetic approaches define human pancreatic β cell mitogenic targets of DYRK1A inhibitors . JCI Insight 5 :e132594. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Adachi T, Yasuda K, Okamoto Y, Shihara N, Oku A, Ueta K, Kitamura K, Saito A, Iwakura I, Yamada Y, et al. (2000) T-1095, a renal Na+-glucose transporter inhibitor, improves hyperglycemia in streptozotocin-induced diabetic rats . Metabolism 49 :990–995. [ PubMed ] [ Google Scholar ]
• Alexiadou K, Tan TM (2020) Gastrointestinal peptides as therapeutic targets to mitigate obesity and metabolic syndrome . Curr Diab Rep 20 :26. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Ambery P, Parker VE, Stumvoll M, Posch MG, Heise T, Plum-Moerschel L, Tsai LF, Robertson D, Jain M, Petrone M, et al. (2018) MEDI0382, a GLP-1 and glucagon receptor dual agonist, in obese or overweight patients with type 2 diabetes: a randomised, controlled, double-blind, ascending dose and phase 2a study . Lancet 391 :2607–2618. [ PubMed ] [ Google Scholar ]
• Ashcroft FM, Rorsman P (1989) Electrophysiology of the pancreatic beta-cell . Prog Biophys Mol Biol 54 :87–143. [ PubMed ] [ Google Scholar ]
• Ashcroft FM, Rorsman P (1990) ATP-sensitive K+ channels: a link between B-cell metabolism and insulin secretion . Biochem Soc Trans 18 :109–111. [ PubMed ] [ Google Scholar ]
• Baeyens L, Hindi S, Sorenson RL, German MS (2016) β-Cell adaptation in pregnancy . Diabetes Obes Metab 18 ( Suppl 1 ):63–70. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Bailey CJ (1992) Biguanides and NIDDM . Diabetes Care 15 :755–772. [ PubMed ] [ Google Scholar ]
• Ben Nasr M, Tezza S, D’Addio F, Mameli C, Usuelli V, Maestroni A, Corradi D, Belletti S, Albarello L, Becchi G, et al. (2017) PD-L1 genetic overexpression or pharmacological restoration in hematopoietic stem and progenitor cells reverses autoimmune diabetes . Sci Transl Med 9 :eaam7543. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Bergmann NC, Lund A, Gasbjerg LS, Meessen ECE, Andersen MM, Bergmann S, Hartmann B, Holst JJ, Jessen L, Christensen MB, et al. (2019) Effects of combined GIP and GLP-1 infusion on energy intake, appetite and energy expenditure in overweight/obese individuals: a randomised, crossover study . Diabetologia 62 :665–675. [ PubMed ] [ Google Scholar ]
• Bernal-Mizrachi E, Kulkarni RN, Scott DK, Mauvais-Jarvis F, Stewart AF, Garcia-Ocaña A (2014) Human β-cell proliferation and intracellular signaling part 2: still driving in the dark without a road map . Diabetes 63 :819–831. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Bethel MA, Patel RA, Merrill P, Lokhnygina Y, Buse JB, Mentz RJ, Pagidipati NJ, Chan JC, Gustavson SM, Iqbal N, et al.; EXSCEL Study Group (2018) Cardiovascular outcomes with glucagon-like peptide-1 receptor agonists in patients with type 2 diabetes: a meta-analysis . Lancet Diabetes Endocrinol 6 :105–113. [ PubMed ] [ Google Scholar ]
• Bliss M (1982) Banting’s, Best’s, and Collip’s accounts of the discovery of insulin . Bull Hist Med 56 :554–568. [ PubMed ] [ Google Scholar ]
• Bochenek MA, Veiseh O, Vegas AJ, McGarrigle JJ, Qi M, Marchese E, Omami M, Doloff JC, Mendoza-Elias J, Nourmohammadzadeh M, et al. (2018) Alginate encapsulation as long-term immune protection of allogeneic pancreatic islet cells transplanted into the omental bursa of macaques . Nat Biomed Eng 2 :810–821. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Bonner C, Kerr-Conte J, Gmyr V, Queniat G, Moerman E, Thévenet J, Beaucamps C, Delalleau N, Popescu I, Malaisse WJ, et al. (2015) Inhibition of the glucose transporter SGLT2 with dapagliflozin in pancreatic alpha cells triggers glucagon secretion . Nat Med 21 :512–517. [ PubMed ] [ Google Scholar ]
• Bonora E, Cigolini M, Bosello O, Zancanaro C, Capretti L, Zavaroni I, Coscelli C, Butturini U (1984) Lack of effect of intravenous metformin on plasma concentrations of glucose, insulin, C-peptide, glucagon and growth hormone in non-diabetic subjects . Curr Med Res Opin 9 :47–51. [ PubMed ] [ Google Scholar ]
• Bottazzo GF, Dean BM, McNally JM, MacKay EH, Swift PG, Gamble DR (1985) In situ characterization of autoimmune phenomena and expression of HLA molecules in the pancreas in diabetic insulitis . N Engl J Med 313 :353–360. [ PubMed ] [ Google Scholar ]
• Boyle JP, Thompson TJ, Gregg EW, Barker LE, Williamson DF (2010) Projection of the year 2050 burden of diabetes in the US adult population: dynamic modeling of incidence, mortality, and prediabetes prevalence . Popul Health Metr 8 :29. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Brunham LR, Kruit JK, Pape TD, Timmins JM, Reuwer AQ, Vasanji Z, Marsh BJ, Rodrigues B, Johnson JD, Parks JS, et al. (2007) Beta-cell ABCA1 influences insulin secretion, glucose homeostasis and response to thiazolidinedione treatment . Nat Med 13 :340–347. [ PubMed ] [ Google Scholar ]
• Buse JB, DeFronzo RA, Rosenstock J, Kim T, Burns C, Skare S, Baron A, Fineman M (2016) The primary glucose-lowering effect of metformin resides in the gut, not the circulation: results from short-term pharmacokinetic and 12-week dose-ranging studies . Diabetes Care 39 :198–205. [ PubMed ] [ Google Scholar ]
• Buse JB, Henry RR, Han J, Kim DD, Fineman MS, Baron AD; Exenatide-113 Clinical Study Group (2004) Effects of exenatide (exendin-4) on glycemic control over 30 weeks in sulfonylurea-treated patients with type 2 diabetes . Diabetes Care 27 :2628–2635. [ PubMed ] [ Google Scholar ]
• Capozzi ME, Svendsen B, Encisco SE, Lewandowski SL, Martin MD, Lin H, Jaffe JL, Coch RW, Haldeman JM, MacDonald PE, et al. (2019a) β Cell tone is defined by proglucagon peptides through cAMP signaling . JCI Insight 4 :e126742. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Capozzi ME, Wait JB, Koech J, Gordon AN, Coch RW, Svendsen B, Finan B, D’Alessio DA, Campbell JE (2019b) Glucagon lowers glycemia when β-cells are active . JCI Insight 5 :e129954. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Carboneau BA, Allan JA, Townsend SE, Kimple ME, Breyer RM, Gannon M (2017) Opposing effects of prostaglandin E 2 receptors EP3 and EP4 on mouse and human β-cell survival and proliferation . Mol Metab 6 :548–559. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chae H, Augustin R, Gatineau E, Mayoux E, Bensellam M, Antoine N, Khattab F, Lai BK, Brusa D, Stierstorfer B, et al. (2020) SGLT2 is not expressed in pancreatic α- and β-cells, and its inhibition does not directly affect glucagon and insulin secretion in rodents and humans . Mol Metab 42 :101071. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chamberlain CE, Scheel DW, McGlynn K, Kim H, Miyatsuka T, Wang J, Nguyen V, Zhao S, Mavropoulos A, Abraham AG, et al. (2014) Menin determines K-RAS proliferative outputs in endocrine cells . J Clin Invest 124 :4093–4101. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chen H, Gu X, Liu Y, Wang J, Wirt SE, Bottino R, Schorle H, Sage J, Kim SK (2011) PDGF signalling controls age-dependent proliferation in pancreatic β-cells . Nature 478 :349–355. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chen H, Kleinberger JW, Takane KK, Salim F, Fiaschi-Taesch N, Pappas K, Parsons R, Jiang J, Zhang Y, Liu H, et al. (2015) Augmented Stat5 signaling bypasses multiple impediments to lactogen-mediated proliferation in human β-cells . Diabetes 64 :3784–3797. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chen J, Cha-Molstad H, Szabo A, Shalev A (2009) Diabetes induces and calcium channel blockers prevent cardiac expression of proapoptotic thioredoxin-interacting protein . Am J Physiol Endocrinol Metab 296 :E1133–E1139. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chen J, Hui ST, Couto FM, Mungrue IN, Davis DB, Attie AD, Lusis AJ, Davis RA, Shalev A (2008a) Thioredoxin-interacting protein deficiency induces Akt/Bcl-xL signaling and pancreatic beta-cell mass and protects against diabetes . FASEB J 22 :3581–3594. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chen J, Saxena G, Mungrue IN, Lusis AJ, Shalev A (2008b) Thioredoxin-interacting protein: a critical link between glucose toxicity and beta-cell apoptosis . Diabetes 57 :938–944. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Chou DH, Webber MJ, Tang BC, Lin AB, Thapa LS, Deng D, Truong JV, Cortinas AB, Langer R, Anderson DG (2015) Glucose-responsive insulin activity by covalent modification with aliphatic phenylboronic acid conjugates . Proc Natl Acad Sci USA 112 :2401–2406. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Cnop M, Toivonen S, Igoillo-Esteve M, Salpea P (2017) Endoplasmic reticulum stress and eIF2α phosphorylation: the Achilles heel of pancreatic β cells . Mol Metab 6 :1024–1039. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Cohen MA, Ellis SM, Le Roux CW, Batterham RL, Park A, Patterson M, Frost GS, Ghatei MA, Bloom SR (2003) Oxyntomodulin suppresses appetite and reduces food intake in humans . J Clin Endocrinol Metab 88 :4696–4701. [ PubMed ] [ Google Scholar ]
• Coll AP, Chen M, Taskar P, Rimmington D, Patel S, Tadross JA, Cimino I, Yang M, Welsh P, Virtue S, et al. (2020) GDF15 mediates the effects of metformin on body weight and energy balance . Nature 578 :444–448. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Colli ML, Hill JLE, Marroquí L, Chaffey J, Dos Santos RS, Leete P, Coomans de Brachène A, Paula FMM, Op de Beeck A, Castela A, et al. (2018) PDL1 is expressed in the islets of people with type 1 diabetes and is up-regulated by interferons-α and-γ via IRF1 induction . EBioMedicine 36 :367–375. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• RISE Consortium (2019) Lack of durable improvements in β-cell function following withdrawal of pharmacological interventions in adults with impaired glucose tolerance or recently diagnosed type 2 diabetes . Diabetes Care 42 :1742–1751. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Dai C, Hang Y, Shostak A, Poffenberger G, Hart N, Prasad N, Phillips N, Levy SE, Greiner DL, Shultz LD, et al. (2017) Age-dependent human β cell proliferation induced by glucagon-like peptide 1 and calcineurin signaling . J Clin Invest 127 :3835–3844. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Dai C, Kayton NS, Shostak A, Poffenberger G, Cyphert HA, Aramandla R, Thompson C, Papagiannis IG, Emfinger C, Shiota M, et al. (2016) Stress-impaired transcription factor expression and insulin secretion in transplanted human islets . J Clin Invest 126 :1857–1870. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Dai C, Walker JT, Shostak A, Bouchi Y, Poffenberger G, Hart NJ, Jacobson DA, Calcutt MW, Bottino R, Greiner DL, et al. (2020) Dapagliflozin does not directly affect human α or β cells . Endocrinology 161 :bqaa080. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Dakin CL, Gunn I, Small CJ, Edwards CM, Hay DL, Smith DM, Ghatei MA, Bloom SR (2001) Oxyntomodulin inhibits food intake in the rat . Endocrinology 142 :4244–4250. [ PubMed ] [ Google Scholar ]
• Dakin CL, Small CJ, Park AJ, Seth A, Ghatei MA, Bloom SR (2002) Repeated ICV administration of oxyntomodulin causes a greater reduction in body weight gain than in pair-fed rats . Am J Physiol Endocrinol Metab 283 :E1173–E1177. [ PubMed ] [ Google Scholar ]
• Defronzo RA, Tripathy D, Schwenke DC, Banerji M, Bray GA, Buchanan TA, Clement SC, Gastaldelli A, Henry RR, Kitabchi AE, et al.; ACT NOW Study (2013) Prevention of diabetes with pioglitazone in ACT NOW: physiologic correlates . Diabetes 62 :3920–3926. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Delaunay F, Khan A, Cintra A, Davani B, Ling ZC, Andersson A, Ostenson CG, Gustafsson J, Efendic S, Okret S (1997) Pancreatic beta cells are important targets for the diabetogenic effects of glucocorticoids . J Clin Invest 100 :2094–2098. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Desai T, Shea LD (2017) Advances in islet encapsulation technologies . Nat Rev Drug Discov 16 :338–350. [ PubMed ] [ Google Scholar ]
• Devaraj S, Venkatachalam A, Chen X (2016) Metformin and the gut microbiome in diabetes . Clin Chem 62 :1554–1555. [ PubMed ] [ Google Scholar ]
• Doloff JC, Veiseh O, Vegas AJ, Tam HH, Farah S, Ma M, Li J, Bader A, Chiu A, Sadraei A, et al. (2017) Colony stimulating factor-1 receptor is a central component of the foreign body response to biomaterial implants in rodents and non-human primates . Nat Mater 16 :671–680. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Donath MY, Dinarello CA, Mandrup-Poulsen T (2019) Targeting innate immune mediators in type 1 and type 2 diabetes . Nat Rev Immunol 19 :734–746. [ PubMed ] [ Google Scholar ]
• Drucker DJ (2018) Mechanisms of action and therapeutic application of glucagon-like peptide-1 . Cell Metab 27 :740–756. [ PubMed ] [ Google Scholar ]
• Drucker DJ, Philippe J, Mojsov S, Chick WL, Habener JF (1987) Glucagon-like peptide I stimulates insulin gene expression and increases cyclic AMP levels in a rat islet cell line . Proc Natl Acad Sci USA 84 :3434–3438. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Duca FA, Côté CD, Rasmussen BA, Zadeh-Tahmasebi M, Rutter GA, Filippi BM, Lam TK (2015) Metformin activates a duodenal Ampk-dependent pathway to lower hepatic glucose production in rats . Nat Med 21 :506–511. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Eguchi K, Nagai R (2017) Islet inflammation in type 2 diabetes and physiology . J Clin Invest 127 :14–23. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• El Khatib MM, Sakuma T, Tonne JM, Mohamed MS, Holditch SJ, Lu B, Kudva YC, Ikeda Y (2015) β-Cell-targeted blockage of PD1 and CTLA4 pathways prevents development of autoimmune diabetes and acute allogeneic islets rejection . Gene Ther 22 :430–438. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• El-Khatib FH, Balliro C, Hillard MA, Magyar KL, Ekhlaspour L, Sinha M, Mondesir D, Esmaeili A, Hartigan C, Thompson MJ, et al. (2017) Home use of a bihormonal bionic pancreas versus insulin pump therapy in adults with type 1 diabetes: a multicentre randomised crossover trial . Lancet 389 :369–380. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• El-Mir MY, Detaille D, R-Villanueva G, Delgado-Esteban M, Guigas B, Attia S, Fontaine E, Almeida A, Leverve X (2008) Neuroprotective role of antidiabetic drug metformin against apoptotic cell death in primary cortical neurons . J Mol Neurosci 34 :77–87. [ PubMed ] [ Google Scholar ]
• El-Mir MY, Nogueira V, Fontaine E, Avéret N, Rigoulet M, Leverve X (2000) Dimethylbiguanide inhibits cell respiration via an indirect effect targeted on the respiratory chain complex I . J Biol Chem 275 :223–228. [ PubMed ] [ Google Scholar ]
• Elrick H, Stimmler L, Hlad CJ Jr, Arai Y (1964) Plasma Insulin Response to Oral and Intravenous Glucose Administration . J Clin Endocrinol Metab 24 :1076–1082. [ PubMed ] [ Google Scholar ]
• Engin F, Yermalovich A, Nguyen T, Hummasti S, Fu W, Eizirik DL, Mathis D, Hotamisligil GS (2013) Restoration of the unfolded protein response in pancreatic β cells protects mice against type 1 diabetes [published correction appears in Sci Transl Med (2013) 5 :214er11] . Sci Transl Med 5 :211ra156. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR, Morris AD (2005) Metformin and reduced risk of cancer in diabetic patients . BMJ 330 :1304–1305. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Evans-Molina C, Robbins RD, Kono T, Tersey SA, Vestermark GL, Nunemaker CS, Garmey JC, Deering TG, Keller SR, Maier B, et al. (2009) Peroxisome proliferator-activated receptor gamma activation restores islet function in diabetic mice through reduction of endoplasmic reticulum stress and maintenance of euchromatin structure . Mol Cell Biol 29 :2053–2067. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Fernández-Real JM, López-Bermejo A, Ropero AB, Piquer S, Nadal A, Bassols J, Casamitjana R, Gomis R, Arnaiz E, Pérez I, et al. (2008) Salicylates increase insulin secretion in healthy obese subjects . J Clin Endocrinol Metab 93 :2523–2530. [ PubMed ] [ Google Scholar ]
• Foretz M, Guigas B, Bertrand L, Pollak M, Viollet B (2014) Metformin: from mechanisms of action to therapies . Cell Metab 20 :953–966. [ PubMed ] [ Google Scholar ]
• Foretz M, Guigas B, Viollet B (2019) Understanding the glucoregulatory mechanisms of metformin in type 2 diabetes mellitus . Nat Rev Endocrinol 15 :569–589. [ PubMed ] [ Google Scholar ]
• Foretz M, Hébrard S, Leclerc J, Zarrinpashneh E, Soty M, Mithieux G, Sakamoto K, Andreelli F, Viollet B (2010) Metformin inhibits hepatic gluconeogenesis in mice independently of the LKB1/AMPK pathway via a decrease in hepatic energy state . J Clin Invest 120 :2355–2369. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Frias JPBastyr EJ 3rd, Vignati L, Tschöp MH, Schmitt C, Owen K, Christensen RHDiMarchi RD (2017) The sustained effects of a dual GIP/GLP-1 receptor agonist, NNC0090-2746, in patients with type 2 diabetes . Cell Metab 26 :343–352.e2. [ PubMed ] [ Google Scholar ]
• Frias JP, Nauck MA, Van J, Benson C, Bray R, Cui X, Milicevic Z, Urva S, Haupt A, Robins DA (2020) Efficacy and tolerability of tirzepatide, a dual glucose-dependent insulinotropic peptide and glucagon-like peptide-1 receptor agonist in patients with type 2 diabetes: A 12-week, randomized, double-blind, placebo-controlled study to evaluate different dose-escalation regimens . Diabetes Obes Metab 22 :938–946. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Frias JP, Nauck MA, Van J, Kutner ME, Cui X, Benson C, Urva S, Gimeno RE, Milicevic Z, Robins D, et al. (2018) Efficacy and safety of LY3298176, a novel dual GIP and GLP-1 receptor agonist, in patients with type 2 diabetes: a randomised, placebo-controlled and active comparator-controlled phase 2 trial . Lancet 392 :2180–2193. [ PubMed ] [ Google Scholar ]
• Fullerton MD, Galic S, Marcinko K, Sikkema S, Pulinilkunnil T, Chen ZP, O’Neill HM, Ford RJ, Palanivel R, O’Brien M, et al. (2013) Single phosphorylation sites in Acc1 and Acc2 regulate lipid homeostasis and the insulin-sensitizing effects of metformin . Nat Med 19 :1649–1654. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Gerstein HC, Colhoun HM, Dagenais GR, Diaz R, Lakshmanan M, Pais P, Probstfield J, Riesmeyer JS, Riddle MC, Rydén L, et al.; REWIND Investigators (2019) Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial . Lancet 394 :121–130. [ PubMed ] [ Google Scholar ]
• Giugliano D, Ceriello A, Saccomanno F, Quatraro A, Paolisso G, D’Onofrio F (1985) Effects of salicylate, tolbutamide, and prostaglandin E2 on insulin responses to glucose in noninsulin-dependent diabetes mellitus . J Clin Endocrinol Metab 61 :160–166. [ PubMed ] [ Google Scholar ]
• Goudy KS, Tisch R (2005) Immunotherapy for the prevention and treatment of type 1 diabetes . Int Rev Immunol 24 :307–326. [ PubMed ] [ Google Scholar ]
• Gregg BE, Moore PC, Demozay D, Hall BA, Li M, Husain A, Wright AJ, Atkinson MA, Rhodes CJ (2012) Formation of a human β-cell population within pancreatic islets is set early in life . J Clin Endocrinol Metab 97 :3197–3206. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Grempler R, Thomas L, Eckhardt M, Himmelsbach F, Sauer A, Sharp DE, Bakker RA, Mark M, Klein T, Eickelmann P (2012) Empagliflozin, a novel selective sodium glucose cotransporter-2 (SGLT-2) inhibitor: characterisation and comparison with other SGLT-2 inhibitors . Diabetes Obes Metab 14 :83–90. [ PubMed ] [ Google Scholar ]
• Harlan DM (2016) Islet transplantation for hypoglycemia unawareness/severe hypoglycemia: caveat emptor . Diabetes Care 39 :1072–1074. [ PubMed ] [ Google Scholar ]
• Harrower AD (1991) Efficacy of gliclazide in comparison with other sulphonylureas in the treatment of NIDDM . Diabetes Res Clin Pract 14 ( Suppl 2 ):S65–S67. [ PubMed ] [ Google Scholar ]
• He L, Wondisford FE (2015) Metformin action: concentrations matter . Cell Metab 21 :159–162. [ PubMed ] [ Google Scholar ]
• Hedrington MS, Davis SN (2019) Considerations when using alpha-glucosidase inhibitors in the treatment of type 2 diabetes . Expert Opin Pharmacother 20 :2229–2235. [ PubMed ] [ Google Scholar ]
• Hering BJ, Clarke WR, Bridges ND, Eggerman TL, Alejandro R, Bellin MD, Chaloner K, Czarniecki CW, Goldstein JS, Hunsicker LG, et al.; Clinical Islet Transplantation Consortium (2016) Phase 3 trial of transplantation of human islets in type 1 diabetes complicated by severe hypoglycemia . Diabetes Care 39 :1230–1240. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hering BJ, Kandaswamy R, Ansite JD, Eckman PM, Nakano M, Sawada T, Matsumoto I, Ihm SH, Zhang HJ, Parkey J, et al. (2005) Single-donor, marginal-dose islet transplantation in patients with type 1 diabetes . JAMA 293 :830–835. [ PubMed ] [ Google Scholar ]
• Hernandez AFGreen JBJanmohamed SD’Agostino RB Sr , Granger CB, Jones NP, Leiter LA, Rosenberg AE, Sigmon KN, Somerville MCet al.; Harmony Outcomes committees and investigators (2018) Albiglutide and cardiovascular outcomes in patients with type 2 diabetes and cardiovascular disease (Harmony Outcomes): a double-blind, randomised placebo-controlled trial . Lancet 392 :1519–1529. [ PubMed ] [ Google Scholar ]
• Hiatt WR, Kaul S, Smith RJ (2013) The cardiovascular safety of diabetes drugs--insights from the rosiglitazone experience . N Engl J Med 369 :1285–1287. [ PubMed ] [ Google Scholar ]
• Hirshberg B, Preston EH, Xu H, Tal MG, Neeman Z, Bunnell D, Soleimanpour S, Hale DA, Kirk AD, Harlan DM (2003) Rabbit antithymocyte globulin induction and sirolimus monotherapy supports prolonged islet allograft function in a nonhuman primate islet transplantation model . Transplantation 76 :55–60. [ PubMed ] [ Google Scholar ]
• Hu W, Jiang C, Guan D, Dierickx P, Zhang R, Moscati A, Nadkarni GN, Steger DJ, Loos RJF, Hu C, et al. (2019) Patient adipose stem cell-derived adipocytes reveal genetic variation that predicts antidiabetic drug response . Cell Stem Cell 24 :299–308.e6. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Husain M, Birkenfeld AL, Donsmark M, Dungan K, Eliaschewitz FG, Franco DR, Jeppesen OK, Lingvay I, Mosenzon O, Pedersen SD, et al.; PIONEER 6 Investigators (2019) Oral semaglutide and cardiovascular outcomes in patients with type 2 diabetes . N Engl J Med 381 :841–851. [ PubMed ] [ Google Scholar ]
• Imamura M, Nakanishi K, Suzuki T, Ikegai K, Shiraki R, Ogiyama T, Murakami T, Kurosaki E, Noda A, Kobayashi Y, et al. (2012) Discovery of Ipragliflozin (ASP1941): a novel C-glucoside with benzothiophene structure as a potent and selective sodium glucose co-transporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes mellitus . Bioorg Med Chem 20 :3263–3279. [ PubMed ] [ Google Scholar ]
• Kahn SE, Haffner SM, Heise MA, Herman WH, Holman RR, Jones NP, Kravitz BG, Lachin JM, O’Neill MC, Zinman B, et al.; ADOPT Study Group (2006) Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy . N Engl J Med 355 :2427–2443. [ PubMed ] [ Google Scholar ]
• Kahn SE, Lachin JM, Zinman B, Haffner SM, Aftring RP, Paul G, Kravitz BG, Herman WH, Viberti G, Holman RR; ADOPT Study Group (2011) Effects of rosiglitazone, glyburide, and metformin on β-cell function and insulin sensitivity in ADOPT . Diabetes 60 :1552–1560. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kars M, Yang L, Gregor MF, Mohammed BS, Pietka TA, Finck BN, Patterson BW, Horton JD, Mittendorfer B, Hotamisligil GS, et al. (2010) Tauroursodeoxycholic acid may improve liver and muscle but not adipose tissue insulin sensitivity in obese men and women . Diabetes 59 :1899–1905. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kemp CB, Knight MJ, Scharp DW, Lacy PE, Ballinger WF (1973) Transplantation of isolated pancreatic islets into the portal vein of diabetic rats . Nature 244 :447. [ PubMed ] [ Google Scholar ]
• Kim HI, Cha JY, Kim SY, Kim JW, Roh KJ, Seong JK, Lee NT, Choi KY, Kim KS, Ahn YH (2002) Peroxisomal proliferator-activated receptor-gamma upregulates glucokinase gene expression in beta-cells . Diabetes 51 :676–685. [ PubMed ] [ Google Scholar ]
• Kim HI, Kim JW, Kim SH, Cha JY, Kim KS, Ahn YH (2000) Identification and functional characterization of the peroxisomal proliferator response element in rat GLUT2 promoter . Diabetes 49 :1517–1524. [ PubMed ] [ Google Scholar ]
• Kim SH, Liu A, Ariel D, Abbasi F, Lamendola C, Grove K, Tomasso V, Ochoa H, Reaven G (2014) Effect of salsalate on insulin action, secretion, and clearance in nondiabetic, insulin-resistant individuals: a randomized, placebo-controlled study . Diabetes Care 37 :1944–1950. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Koffert JP, Mikkola K, Virtanen KA, Andersson AD, Faxius L, Hällsten K, Heglind M, Guiducci L, Pham T, Silvola JMU, et al. (2017) Metformin treatment significantly enhances intestinal glucose uptake in patients with type 2 diabetes: Results from a randomized clinical trial . Diabetes Res Clin Pract 131 :208–216. [ PubMed ] [ Google Scholar ]
• Koh A, Mannerås-Holm L, Yunn NO, Nilsson PM, Ryu SH, Molinaro A, Perkins R, Smith JG, Bäckhed F (2020) Microbial imidazole propionate affects responses to metformin through p38γ-dependent inhibitory AMPK phosphorylation . Cell Metab 32 :643–653.e4. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kreymann B, Williams G, Ghatei MA, Bloom SR (1987) Glucagon-like peptide-1 7-36: a physiological incretin in man . Lancet 2 :1300–1304. [ PubMed ] [ Google Scholar ]
• Kuhre RE, Ghiasi SM, Adriaenssens AE, Wewer Albrechtsen NJ, Andersen DB, Aivazidis A, Chen L, Mandrup-Poulsen T, Ørskov C, Gribble FM, et al. (2019) No direct effect of SGLT2 activity on glucagon secretion . Diabetologia 62 :1011–1023. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kulkarni RN, Mizrachi EB, Ocana AG, Stewart AF (2012) Human β-cell proliferation and intracellular signaling: driving in the dark without a road map . Diabetes 61 :2205–2213. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kurosaki E, Ogasawara H (2013) Ipragliflozin and other sodium-glucose cotransporter-2 (SGLT2) inhibitors in the treatment of type 2 diabetes: preclinical and clinical data . Pharmacol Ther 139 :51–59. [ PubMed ] [ Google Scholar ]
• Kurtzhals P, Schäffer L, Sørensen A, Kristensen C, Jonassen I, Schmid C, Trüb T (2000) Correlations of receptor binding and metabolic and mitogenic potencies of insulin analogs designed for clinical use . Diabetes 49 :999–1005. [ PubMed ] [ Google Scholar ]
• Lambeir AM, Scharpé S, De Meester I (2008) DPP4 inhibitors for diabetes--what next? Biochem Pharmacol 76 :1637–1643. [ PubMed ] [ Google Scholar ]
• Laybutt DR, Preston AM, Akerfeldt MC, Kench JG, Busch AK, Biankin AV, Biden TJ (2007) Endoplasmic reticulum stress contributes to beta cell apoptosis in type 2 diabetes . Diabetologia 50 :752–763. [ PubMed ] [ Google Scholar ]
• Lebovitz HE (2019) Thiazolidinediones: the forgotten diabetes medications . Curr Diab Rep 19 :151. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Lewis JD, Habel LA, Quesenberry CP, Strom BL, Peng T, Hedderson MM, Ehrlich SF, Mamtani R, Bilker W, Vaughn DJ, et al. (2015) Pioglitazone use and risk of bladder cancer and other common cancers in persons with diabetes . JAMA 314 :265–277. [ PubMed ] [ Google Scholar ]
• Li L, Li S, Deng K, Liu J, Vandvik PO, Zhao P, Zhang L, Shen J, Bala MM, Sohani ZN, et al. (2016) Dipeptidyl peptidase-4 inhibitors and risk of heart failure in type 2 diabetes: systematic review and meta-analysis of randomised and observational studies . BMJ 352 :i610. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Link JT (2003) Pharmacological regulation of hepatic glucose production . Curr Opin Investig Drugs 4 :421–429. [ PubMed ] [ Google Scholar ]
• Luippold G, Klein T, Mark M, Grempler R (2012) Empagliflozin, a novel potent and selective SGLT-2 inhibitor, improves glycaemic control alone and in combination with insulin in streptozotocin-induced diabetic rats, a model of type 1 diabetes mellitus . Diabetes Obes Metab 14 :601–607. [ PubMed ] [ Google Scholar ]
• Maganti AV, Tersey SA, Syed F, Nelson JB, Colvin SC, Maier B, Mirmira RG (2016) Peroxisome proliferator-activated receptor-γ activation augments the β-cell unfolded protein response and rescues early glycemic deterioration and β cell death in non-obese diabetic mice . J Biol Chem 291 :22524–22533. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Makrilakis K (2019) The role of DPP-4 inhibitors in the treatment algorithm of type 2 diabetes mellitus: when to select, what to expect . Int J Environ Res Public Health 16 :2720. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Marchetti P, Bugliani M, Lupi R, Marselli L, Masini M, Boggi U, Filipponi F, Weir GC, Eizirik DL, Cnop M (2007) The endoplasmic reticulum in pancreatic beta cells of type 2 diabetes patients . Diabetologia 50 :2486–2494. [ PubMed ] [ Google Scholar ]
• Marhfour I, Lopez XM, Lefkaditis D, Salmon I, Allagnat F, Richardson SJ, Morgan NG, Eizirik DL (2012) Expression of endoplasmic reticulum stress markers in the islets of patients with type 1 diabetes . Diabetologia 55 :2417–2420. [ PubMed ] [ Google Scholar ]
• Marso SP, Bain SC, Consoli A, Eliaschewitz FG, Jódar E, Leiter LA, Lingvay I, Rosenstock J, Seufert J, Warren ML, et al.; SUSTAIN-6 Investigators (2016a) Semaglutide and cardiovascular outcomes in patients with type 2 diabetes . N Engl J Med 375 :1834–1844. [ PubMed ] [ Google Scholar ]
• Marso SP, Daniels GH, Brown-Frandsen K, Kristensen P, Mann JF, Nauck MA, Nissen SE, Pocock S, Poulter NR, Ravn LS, et al.; LEADER Steering Committee; LEADER Trial Investigators (2016b) Liraglutide and cardiovascular outcomes in type 2 diabetes . N Engl J Med 375 :311–322. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Massollo M, Marini C, Brignone M, Emionite L, Salani B, Riondato M, Capitanio S, Fiz F, Democrito A, Amaro A, et al. (2013) Metformin temporal and localized effects on gut glucose metabolism assessed using 18F-FDG PET in mice . J Nucl Med 54 :259–266. [ PubMed ] [ Google Scholar ]
• McCreight LJ, Bailey CJ, Pearson ER (2016) Metformin and the gastrointestinal tract . Diabetologia 59 :426–435. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• McIntyre N, Holdsworth CD, Turner DS (1964) New interpretation of oral glucose tolerance . Lancet 2 :20–21. [ PubMed ] [ Google Scholar ]
• McMurray JJV, Solomon SD, Inzucchi SE, Køber L, Kosiborod MN, Martinez FA, Ponikowski P, Sabatine MS, Anand IS, Bělohlávek J, et al.; DAPA-HF Trial Committees and Investigators (2019) Dapagliflozin in patients with heart failure and reduced ejection fraction . N Engl J Med 381 :1995–2008. [ PubMed ] [ Google Scholar ]
• Meier JJ, Butler AE, Saisho Y, Monchamp T, Galasso R, Bhushan A, Rizza RA, Butler PC (2008) Beta-cell replication is the primary mechanism subserving the postnatal expansion of beta-cell mass in humans . Diabetes 57 :1584–1594. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Meng W, Ellsworth BA, Nirschl AA, McCann PJ, Patel M, Girotra RN, Wu G, Sher PM, Morrison EP, Biller SA, et al. (2008) Discovery of dapagliflozin: a potent, selective renal sodium-dependent glucose cotransporter 2 (SGLT2) inhibitor for the treatment of type 2 diabetes . J Med Chem 51 :1145–1149. [ PubMed ] [ Google Scholar ]
• Menting JG, Whittaker J, Margetts MB, Whittaker LJ, Kong GK, Smith BJ, Watson CJ, Záková L, Kletvíková E, Jiráček J, et al. (2013) How insulin engages its primary binding site on the insulin receptor . Nature 493 :241–245. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Miller RA, Chu Q, Xie J, Foretz M, Viollet B, Birnbaum MJ (2013) Biguanides suppress hepatic glucagon signalling by decreasing production of cyclic AMP . Nature 494 :256–260. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Minkowski O (1892) Weitere Mitteilungen über den Diabetes mellitus nach Extirpation des Pankreas . Berliner Klinische Wochenschrift 29 :90–93. [ Google Scholar ]
• Misbin RI (2004) The phantom of lactic acidosis due to metformin in patients with diabetes . Diabetes Care 27 :1791–1793. [ PubMed ] [ Google Scholar ]
• Mudaliar S, Armstrong DA, Mavian AA, O’Connor-Semmes R, Mydlow PK, Ye J, Hussey EK, Nunez DJ, Henry RR, Dobbins RL (2012) Remogliflozin etabonate, a selective inhibitor of the sodium-glucose transporter 2, improves serum glucose profiles in type 1 diabetes . Diabetes Care 35 :2198–2200. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Mulherin AJ, Oh AH, Kim H, Grieco A, Lauffer LM, Brubaker PL (2011) Mechanisms underlying metformin-induced secretion of glucagon-like peptide-1 from the intestinal L cell . Endocrinology 152 :4610–4619. [ PubMed ] [ Google Scholar ]
• Müller TD, Finan B, Bloom SR, D’Alessio D, Drucker DJ, Flatt PR, Fritsche A, Gribble F, Grill HJ, Habener JF, et al. (2019) Glucagon-like peptide 1 (GLP-1) . Mol Metab 30 :72–130. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Munir KM, Lamos EM (2017) Diabetes type 2 management: what are the differences between DPP-4 inhibitors and how do you choose? Expert Opin Pharmacother 18 :839–841. [ PubMed ] [ Google Scholar ]
• Naftanel MA, Harlan DM (2004) Pancreatic islet transplantation . PLoS Med 1 :e58. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Nauck M, Stöckmann F, Ebert R, Creutzfeldt W (1986a) Reduced incretin effect in type 2 (non-insulin-dependent) diabetes . Diabetologia 29 :46–52. [ PubMed ] [ Google Scholar ]
• Nauck MA, Homberger E, Siegel EG, Allen RC, Eaton RP, Ebert R, Creutzfeldt W (1986b) Incretin effects of increasing glucose loads in man calculated from venous insulin and C-peptide responses . J Clin Endocrinol Metab 63 :492–498. [ PubMed ] [ Google Scholar ]
• Nejentsev S, Howson JM, Walker NM, Szeszko J, Field SF, Stevens HE, Reynolds P, Hardy M, King E, Masters J, et al.; Wellcome Trust Case Control Consortium (2007) Localization of type 1 diabetes susceptibility to the MHC class I genes HLA-B and HLA-A . Nature 450 :887–892. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Nichols CG (2006) KATP channels as molecular sensors of cellular metabolism . Nature 440 :470–476. [ PubMed ] [ Google Scholar ]
• Nomura S, Sakamaki S, Hongu M, Kawanishi E, Koga Y, Sakamoto T, Yamamoto Y, Ueta K, Kimata H, Nakayama K, et al. (2010) Discovery of canagliflozin, a novel C-glucoside with thiophene ring, as sodium-dependent glucose cotransporter 2 inhibitor for the treatment of type 2 diabetes mellitus . J Med Chem 53 :6355–6360. [ PubMed ] [ Google Scholar ]
• Ohnishi ST, Endo M, editors. (1981) The Mechanism of Gated Calcium Transport Across Biological Membranes , Academic Press, New York. [ Google Scholar ]
• Osipovich AB, Stancill JS, Cartailler JP, Dudek KD, Magnuson MA (2020) Excitotoxicity and overnutrition additively impair metabolic function and identity of pancreatic β-cells . Diabetes 69 :1476–1491. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Osum KC, Burrack AL, Martinov T, Sahli NL, Mitchell JS, Tucker CG, Pauken KE, Papas K, Appakalai B, Spanier JA, et al. (2018) Interferon-gamma drives programmed death-ligand 1 expression on islet β cells to limit T cell function during autoimmune diabetes . Sci Rep 8 :8295. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Owen MR, Doran E, Halestrap AP (2000) Evidence that metformin exerts its anti-diabetic effects through inhibition of complex 1 of the mitochondrial respiratory chain . Biochem J 348 :607–614. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Ozanne SE, Guest PC, Hutton JC, Hales CN (1995) Intracellular localization and molecular heterogeneity of the sulphonylurea receptor in insulin-secreting cells . Diabetologia 38 :277–282. [ PubMed ] [ Google Scholar ]
• Ozcan U, Yilmaz E, Ozcan L, Furuhashi M, Vaillancourt E, Smith RO, Görgün CZ, Hotamisligil GS (2006) Chemical chaperones reduce ER stress and restore glucose homeostasis in a mouse model of type 2 diabetes . Science 313 :1137–1140. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Paty BW, Harmon JS, Marsh CL, Robertson RP (2002) Inhibitory effects of immunosuppressive drugs on insulin secretion from HIT-T15 cells and Wistar rat islets . Transplantation 73 :353–357. [ PubMed ] [ Google Scholar ]
• Penesova A, Koska J, Ortega E, Bunt JC, Bogardus C, de Courten B (2015) Salsalate has no effect on insulin secretion but decreases insulin clearance: a randomized, placebo-controlled trial in subjects without diabetes . Diabetes Obes Metab 17 :608–612. [ PubMed ] [ Google Scholar ]
• Perdigoto AL, Quandt Z, Anderson M, Herold KC (2019) Checkpoint inhibitor-induced insulin-dependent diabetes: an emerging syndrome . Lancet Diabetes Endocrinol 7 :421–423. [ PubMed ] [ Google Scholar ]
• Perley MJ, Kipnis DM (1967) Plasma insulin responses to oral and intravenous glucose: studies in normal and diabetic subjects . J Clin Invest 46 :1954–1962. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Pernicova I, Korbonits M (2014) Metformin--mode of action and clinical implications for diabetes and cancer . Nat Rev Endocrinol 10 :143–156. [ PubMed ] [ Google Scholar ]
• Preiss D, Dawed A, Welsh P, Heggie A, Jones AG, Dekker J, Koivula R, Hansen TH, Stewart C, Holman RR, et al.; DIRECT consortium group (2017) Sustained influence of metformin therapy on circulating glucagon-like peptide-1 levels in individuals with and without type 2 diabetes . Diabetes Obes Metab 19 :356–363. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Prentki M, Nolan CJ (2006) Islet beta cell failure in type 2 diabetes . J Clin Invest 116 :1802–1812. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Pueyo ME, Darquy S, Capron F, Reach G (1993) In vitro activation of human macrophages by alginate-polylysine microcapsules . J Biomater Sci Polym Ed 5 :197–203. [ PubMed ] [ Google Scholar ]
• Pybus F (1924) Notes on suprarenal and pancreatic grafting . Lancet 204 :550–551. [ Google Scholar ]
• Quattrin T, Haller MJ, Steck AK, Felner EI, Li Y, Xia Y, Leu JH, Zoka R, Hedrick JA, Rigby MR, et al.; T1GER Study Investigators (2020) Golimumab and Beta-Cell Function in Youth with New-Onset Type 1 Diabetes . N Engl J Med 383 :2007–2017. [ PubMed ] [ Google Scholar ]
• Rachdi L, Kariyawasam D, Aïello V, Herault Y, Janel N, Delabar JM, Polak M, Scharfmann R (2014a) Dyrk1A induces pancreatic β cell mass expansion and improves glucose tolerance . Cell Cycle 13 :2221–2229. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rachdi L, Kariyawasam D, Guez F, Aïello V, Arbonés ML, Janel N, Delabar JM, Polak M, Scharfmann R (2014b) Dyrk1a haploinsufficiency induces diabetes in mice through decreased pancreatic beta cell mass . Diabetologia 57 :960–969. [ PubMed ] [ Google Scholar ]
• Rege NK, Phillips NFB, Weiss MA (2017) Development of glucose-responsive ‘smart’ insulin systems . Curr Opin Endocrinol Diabetes Obes 24 :267–278. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rena G, Hardie DG, Pearson ER (2017) The mechanisms of action of metformin . Diabetologia 60 :1577–1585. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rewers M, Gottlieb P (2009) Immunotherapy for the prevention and treatment of type 1 diabetes: human trials and a look into the future . Diabetes Care 32 :1769–1782. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Richardson SJ, Rodriguez-Calvo T, Gerling IC, Mathews CE, Kaddis JS, Russell MA, Zeissler M, Leete P, Krogvold L, Dahl-Jørgensen K, et al. (2016) Islet cell hyperexpression of HLA class I antigens: a defining feature in type 1 diabetes . Diabetologia 59 :2448–2458. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rickels MR, Liu C, Shlansky-Goldberg RD, Soleimanpour SA, Vivek K, Kamoun M, Min Z, Markmann E, Palangian M, Dalton-Bakes C, et al. (2013) Improvement in β-cell secretory capacity after human islet transplantation according to the c7 protocol . Diabetes 62 :2890–2897. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rickels MR, Robertson RP (2019) Pancreatic islet transplantation in humans: recent progress and future directions . Endocr Rev 40 :631–668. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rodriguez-Calvo T, Suwandi JS, Amirian N, Zapardiel-Gonzalo J, Anquetil F, Sabouri S, von Herrath MG (2015) Heterogeneity and lobularity of pancreatic pathology in type 1 diabetes during the prediabetic phase . J Histochem Cytochem 63 :626–636. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rojas LB, Gomes MB (2013) Metformin: an old but still the best treatment for type 2 diabetes . Diabetol Metab Syndr 5 :6. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rosen ED, Kulkarni RN, Sarraf P, Ozcan U, Okada T, Hsu CH, Eisenman D, Magnuson MA, Gonzalez FJ, Kahn CR, et al. (2003) Targeted elimination of peroxisome proliferator-activated receptor gamma in beta cells leads to abnormalities in islet mass without compromising glucose homeostasis . Mol Cell Biol 23 :7222–7229. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rosenstock J, Hassman DR, Madder RD, Brazinsky SA, Farrell J, Khutoryansky N, Hale PM; Repaglinide Versus Nateglinide Comparison Study Group (2004) Repaglinide versus nateglinide monotherapy: a randomized, multicenter study . Diabetes Care 27 :1265–1270. [ PubMed ] [ Google Scholar ]
• Sampaio MS, Kuo HT, Bunnapradist S (2011) Outcomes of simultaneous pancreas-kidney transplantation in type 2 diabetic recipients . Clin J Am Soc Nephrol 6 :1198–1206. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Saponaro C, Gmyr V, Thévenet J, Moerman E, Delalleau N, Pasquetti G, Coddeville A, Quenon A, Daoudi M, Hubert T, et al. (2019) The GLP1R agonist liraglutide reduces hyperglucagonemia induced by the SGLT2 inhibitor dapagliflozin via somatostatin release . Cell Rep 28 :1447–1454.e4. [ PubMed ] [ Google Scholar ]
• Saponaro C, Mühlemann M, Acosta-Montalvo A, Piron A, Gmyr V, Delalleau N, Moerman E, Thévenet J, Pasquetti G, Coddeville A, et al. (2020) Interindividual heterogeneity of SGLT2 expression and function in human pancreatic islets . Diabetes 69 :902–914. [ PubMed ] [ Google Scholar ]
• Satin LS, Tavalin SJ, Kinard TA, Teague J (1995) Contribution of L- and non-L-type calcium channels to voltage-gated calcium current and glucose-dependent insulin secretion in HIT-T15 cells . Endocrinology 136 :4589–4601. [ PubMed ] [ Google Scholar ]
• Schwartz AV, Chen H, Ambrosius WT, Sood A, Josse RG, Bonds DE, Schnall AM, Vittinghoff E, Bauer DC, Banerji MA, et al. (2015) Effects of TZD use and discontinuation on fracture rates in ACCORD Bone Study . J Clin Endocrinol Metab 100 :4059–4066. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Shalev A, Pise-Masison CA, Radonovich M, Hoffmann SC, Hirshberg B, Brady JN, Harlan DM (2002) Oligonucleotide microarray analysis of intact human pancreatic islets: identification of glucose-responsive genes and a highly regulated TGFbeta signaling pathway . Endocrinology 143 :3695–3698. [ PubMed ] [ Google Scholar ]
• Shankar SS, Shankar RR, Mixson LA, Miller DL, Pramanik B, O’Dowd AK, Williams DM, Frederick CB, Beals CR, Stoch SA, et al. (2018) Native oxyntomodulin has significant glucoregulatory effects independent of weight loss in obese humans with and without type 2 diabetes . Diabetes 67 :1105–1112. [ PubMed ] [ Google Scholar ]
• Shapiro AM, Lakey JR, Ryan EA, Korbutt GS, Toth E, Warnock GL, Kneteman NM, Rajotte RV (2000) Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen . N Engl J Med 343 :230–238. [ PubMed ] [ Google Scholar ]
• Sharma RB, O’Donnell AC, Stamateris RE, Ha B, McCloskey KM, Reynolds PR, Arvan P, Alonso LC (2015) Insulin demand regulates β cell number via the unfolded protein response . J Clin Invest 125 :3831–3846. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sims EK, Mirmira RG, Evans-Molina C (2020) The role of beta-cell dysfunction in early type 1 diabetes . Curr Opin Endocrinol Diabetes Obes 27 :215–224. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Soccio RE, Chen ER, Rajapurkar SR, Safabakhsh P, Marinis JM, Dispirito JR, Emmett MJ, Briggs ER, Fang B, Everett LJ, et al. (2015) Genetic variation determines PPARγ function and anti-diabetic drug response in vivo . Cell 162 :33–44. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Soleimanpour SA, Crutchlow MF, Ferrari AM, Raum JC, Groff DN, Rankin MM, Liu C, De León DD, Naji A, Kushner JA, et al. (2010) Calcineurin signaling regulates human islet beta-cell survival . J Biol Chem 285 :40050–40059. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Soleimanpour SA, Hirshberg B, Bunnell DJ, Sumner AE, Ader M, Remaley AT, Rother KI, Rickels MR, Harlan DM (2012) Metabolic function of a suboptimal transplanted islet mass in nonhuman primates on rapamycin monotherapy . Cell Transplant 21 :1297–1304. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Soleimanpour SA, Stoffers DA (2013) The pancreatic β cell and type 1 diabetes: innocent bystander or active participant? Trends Endocrinol Metab 24 :324–331. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Stamatouli AM, Quandt Z, Perdigoto AL, Clark PL, Kluger H, Weiss SA, Gettinger S, Sznol M, Young A, Rushakoff R, et al. (2018) Collateral damage: insulin-dependent diabetes induced with checkpoint inhibitors . Diabetes 67 :1471–1480. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Stancill JS, Cartailler JP, Clayton HW, O’Connor JT, Dickerson MT, Dadi PK, Osipovich AB, Jacobson DA, Magnuson MA (2017) Chronic β-cell depolarization impairs β-cell identity by disrupting a network of Ca 2+ -regulated genes . Diabetes 66 :2175–2187. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Stewart AF, Hussain MA, García-Ocaña A, Vasavada RC, Bhushan A, Bernal-Mizrachi E, Kulkarni RN (2015) Human β-cell proliferation and intracellular signaling: part 3 . Diabetes 64 :1872–1885. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Stojanovic I, Dimitrijevic M, Vives-Pi M, Mansilla MJ, Pujol-Autonell I, Rodríguez-Fernandez S, Palova-Jelínkova L, Funda DP, Gruden-Movsesijan A, Sofronic-Milosavljevic L, et al. (2017) Cell-based tolerogenic therapy, experience from animal models of multiple sclerosis, type 1 diabetes and rheumatoid arthritis . Curr Pharm Des 23 :2623–2643. [ PubMed ] [ Google Scholar ]
• Sturek JM, Castle JD, Trace AP, Page LC, Castle AM, Evans-Molina C, Parks JS, Mirmira RG, Hedrick CC (2010) An intracellular role for ABCG1-mediated cholesterol transport in the regulated secretory pathway of mouse pancreatic beta cells . J Clin Invest 120 :2575–2589. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Suga T, Kikuchi O, Kobayashi M, Matsui S, Yokota-Hashimoto H, Wada E, Kohno D, Sasaki T, Takeuchi K, Kakizaki S, et al. (2019) SGLT1 in pancreatic α cells regulates glucagon secretion in mice, possibly explaining the distinct effects of SGLT2 inhibitors on plasma glucagon levels . Mol Metab 19 :1–12. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sun L, Xie C, Wang G, Wu Y, Wu Q, Wang X, Liu J, Deng Y, Xia J, Chen B, et al. (2018) Gut microbiota and intestinal FXR mediate the clinical benefits of metformin . Nat Med 24 :1919–1929. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sutherland DE, Matas AJ, Najarian JS (1978) Pancreatic islet cell transplantation . Surg Clin North Am 58 :365–382. [ PubMed ] [ Google Scholar ]
• Tersey SA, Nishiki Y, Templin AT, Cabrera SM, Stull ND, Colvin SC, Evans-Molina C, Rickus JL, Maier B, Mirmira RG (2012) Islet β-cell endoplasmic reticulum stress precedes the onset of type 1 diabetes in the nonobese diabetic mouse model . Diabetes 61 :818–827. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Thielen LA, Chen J, Jing G, Moukha-Chafiq O, Xu G, Jo S, Grayson TB, Lu B, Li P, Augelli-Szafran CE, et al. (2020) Identification of an anti-diabetic, orally available small molecule that regulates TXNIP expression and glucagon action . Cell Metab 32 :353–365.e8. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Tillner J, Posch MG, Wagner F, Teichert L, Hijazi Y, Einig C, Keil S, Haack T, Wagner M, Bossart M, et al. (2019) A novel dual glucagon-like peptide and glucagon receptor agonist SAR425899: Results of randomized, placebo-controlled first-in-human and first-in-patient trials . Diabetes Obes Metab 21 :120–128. [ PubMed ] [ Google Scholar ]
• Vallon V (2015) The mechanisms and therapeutic potential of SGLT2 inhibitors in diabetes mellitus . Annu Rev Med 66 :255–270. [ PubMed ] [ Google Scholar ]
• Vasseur M, Debuyser A, Joffre M (1987) Sensitivity of pancreatic beta cell to calcium channel blockers. An electrophysiologic study of verapamil and nifedipine . Fundam Clin Pharmacol 1 :95–113. [ PubMed ] [ Google Scholar ]
• Vegas AJ, Veiseh O, Doloff JC, Ma M, Tam HH, Bratlie K, Li J, Bader AR, Langan E, Olejnik K, et al. (2016) Combinatorial hydrogel library enables identification of materials that mitigate the foreign body response in primates . Nat Biotechnol 34 :345–352. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Vergari E, Knudsen JG, Ramracheya R, Salehi A, Zhang Q, Adam J, Asterholm IW, Benrick A, Briant LJB, Chibalina MV, et al. (2019) Insulin inhibits glucagon release by SGLT2-induced stimulation of somatostatin secretion . Nat Commun 10 :139. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wallner K, Shapiro AM, Senior PA, McCabe C (2016) Cost effectiveness and value of information analyses of islet cell transplantation in the management of ‘unstable’ type 1 diabetes mellitus . BMC Endocr Disord 16 :17. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wang H, Bender A, Wang P, Karakose E, Inabnet WB, Libutti SK, Arnold A, Lambertini L, Stang M, Chen H, et al. (2017) Insights into beta cell regeneration for diabetes via integration of molecular landscapes in human insulinomas . Nat Commun 8 :767. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wang Y, An H, Liu T, Qin C, Sesaki H, Guo S, Radovick S, Hussain M, Maheshwari A, Wondisford FE, O’Rourke B, He L (2019) Metformin improves mitochondrial respiratory activity through activation of AMPK . Cell Rep 29 :1511–1523.e5. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Westerman J, Wirtz KW, Berkhout T, van Deenen LL, Radhakrishnan R, Khorana HG (1983) Identification of the lipid-binding site of phosphatidylcholine-transfer protein with phosphatidylcholine analogs containing photoactivable carbene precursors . Eur J Biochem 132 :441–449. [ PubMed ] [ Google Scholar ]
• World Health Organization (2020) World Health Organization Diabetes Fact Sheet . [ Google Scholar ]
• Willard FS, Douros JD, Gabe MB, Showalter AD, Wainscott DB, Suter TM, Capozzi ME, van der Velden WJ, Stutsman C, Cardona GR, et al. (2020) Tirzepatide is an imbalanced and biased dual GIP and GLP-1 receptor agonist . JCI Insight 5 :e140532. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Williams P (1894) Notes on diabetes treated with extract and by grafts of sheep’s pancreas . BMJ 2 :1303–1304. [ Google Scholar ]
• Williamson RT (1901) On the treatment of glycosuria and diabetes mellitus with sodium salicylate . BMJ 1 :760–762. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Witters LA (2001) The blooming of the French lilac . J Clin Invest 108 :1105–1107. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Wiviott SD, Raz I, Bonaca MP, Mosenzon O, Kato ET, Cahn A, Silverman MG, Zelniker TA, Kuder JF, Murphy SA, et al.; DECLARE–TIMI 58 Investigators (2019) Dapagliflozin and cardiovascular outcomes in type 2 diabetes . N Engl J Med 380 :347–357. [ PubMed ] [ Google Scholar ]
• Wu H, Esteve E, Tremaroli V, Khan MT, Caesar R, Mannerås-Holm L, Ståhlman M, Olsson LM, Serino M, Planas-Fèlix M, et al. (2017) Metformin alters the gut microbiome of individuals with treatment-naive type 2 diabetes, contributing to the therapeutic effects of the drug . Nat Med 23 :850–858. [ PubMed ] [ Google Scholar ]
• Wynne K, Park AJ, Small CJ, Patterson M, Ellis SM, Murphy KG, Wren AM, Frost GS, Meeran K, Ghatei MA, et al. (2005) Subcutaneous oxyntomodulin reduces body weight in overweight and obese subjects: a double-blind, randomized, controlled trial . Diabetes 54 :2390–2395. [ PubMed ] [ Google Scholar ]
• Xu G, Chen J, Jing G, Shalev A (2012) Preventing β-cell loss and diabetes with calcium channel blockers . Diabetes 61 :848–856. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Yang JF, Gong X, Bakh NA, Carr K, Phillips NFB, Ismail-Beigi F, Weiss MA, Strano MS (2020) Connecting rodent and human pharmacokinetic models for the design and translation of glucose-responsive insulin . Diabetes 69 :1815–1826. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Yu O, Azoulay L, Yin H, Filion KB, Suissa S (2018) Sulfonylureas as initial treatment for type 2 diabetes and the risk of severe hypoglycemia . Am J Med 131 :317.e11–317.e22. [ PubMed ] [ Google Scholar ]
• Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, Wu M, Ventre J, Doebber T, Fujii N, et al. (2001) Role of AMP-activated protein kinase in mechanism of metformin action . J Clin Invest 108 :1167–1174. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, Mattheus M, Devins T, Johansen OE, Woerle HJ, et al.; EMPA-REG OUTCOME Investigators (2015) Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes . N Engl J Med 373 :2117–2128. [ PubMed ] [ Google Scholar ]

• Methodology
• Open access
• Published: 18 January 2021

Diabetes management intervention studies: lessons learned from two studies

• Bettina Petersen 1 ,
• Iris Vesper 1 ,
• Bernhild Pachwald 1 ,
• Nicole Dagenbach 1 ,
• Sina Buck   ORCID: orcid.org/0000-0001-8428-1038 2 ,
• Delia Waldenmaier 2 &
• Lutz Heinemann 3  

Trials volume  22 , Article number:  61 ( 2021 ) Cite this article

4934 Accesses

4 Citations

1 Altmetric

Metrics details

Introduction

Several clinical studies investigated improvements of patient outcomes due to diabetes management interventions. However, chronic disease management is intricate with complex multifactorial behavior patterns. Such studies thus have to be well designed in order to allocate all observed effects to the defined intervention and to exclude effects of other confounders as well as possible.

This article aims to provide challenges in interpreting diabetes management intervention studies and suggests approaches for optimizing study implementation and for avoiding pitfalls based on current experiences.

Lessons from the STeP and ProValue studies demonstrated the difficulty in medical device studies that rely on behavioral changes in intervention group patients. To successfully engage patients, priority should be given to health care professionals being engaged, operational support in technical issues being available, and adherence being assessed in detail.

Another difficulty is to avoid contamination of the control group with the intervention; therefore, strict allocation concealment should be maintained. However, randomization and blinding are not always possible. A limited effect size due to improvements regarding clinical endpoints in the control group is often caused by the Hawthorne effect. Improvements in the control group can also be caused with increased attention paid to the subjects. In order to reduce improvements in the control group, it is essential to identify the specific reasons and adjust study procedures accordingly. A pilot phase is indispensable for this. Another option is to include a third study arm to control for enhanced standard of care and study effects. Furthermore, retrospective data collection could be a feasible option. Adaptive study designs might reduce the necessity of a separate pilot study and combine the exploratory and confirmatory stages of an investigation in one single study.

There are several aspects to consider in medical device studies when using interventions that rely on changes in behavior to achieve an effective implementation and significant study results. Improvements in the control group may reduce effect sizes and limit statistical significance; therefore, alternatives to the traditional randomized controlled trials may be considered.

Peer Review reports

Patients with diabetes require a life-long treatment that is not limited to a standardized intake of drugs, but requires a more complex disease management. In particular, type 1 diabetes management involves frequent treatment decisions like adjustment of insulin doses depending on the current glucose status, meal intake, and physical activity level. This requires the use of medical devices, adequate handling by the patient, and translation of the measurement results into appropriate therapeutic decisions. Health care professionals (HCPs) support patients with regular monitoring of markers of glucose control and adjustment of the treatment plan. These factors have a complicated interaction with one another to influence the achievement of a therapeutic goal. If individual components of diabetes management are investigated, e.g., in a clinical study, this interaction has to be taken into account. For example, frequent use of a CGM system and adequate interpretation of glucose values will more likely lead to improvements in diabetes management [ 1 ].

Therapeutic improvements that are observed as a result of device usage are not driven by the device itself, but by the behavioral changes the device enables. In clinical studies with medical devices for diabetes management, behavioral changes of study participants, not only those planned for the intervention group, but also unintended changes in control group participants, as well as those of the HCPs, should be taken into account and adequately considered in study design, implementation, and analysis.

Over the last several years, a number of studies have been published that investigated improvements in patient outcomes driven by interventions in their diabetes management with medical devices [ 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 ].

The Structured Testing Program (STeP) study and the two ProValue studies are examples of studies in which diabetes management interventions on clinical outcomes were investigated [ 17 , 18 ]. All of these studies were multicenter cluster randomized controlled trials (RCTs) with the primary outcome being improvements in glucose control (HbA1c reduction) in patients with type 2 diabetes.

In the STeP study, subjects in the intervention group performed structured self-monitoring of blood glucose (SMBG) and received enhanced care, while subjects in the control group received standard of care. In the ProValue studies, two levels of ambulatory care, namely diabetes specialized practices (first ProValue study) and general medical practices (second ProValue study), were examined. The ProValue studies also had an identical study design, where subjects in the intervention group used an integrated Personalized Diabetes Management (iPDM) system to support their diabetes therapy while subjects in the control group were to receive standard of care. The primary outcome was an improvement in glycemic control of participants using the iPDM tools compared to the subjects in the control group.

The STeP and the ProValue studies reported significant differences in the reductions in HbA1c between the intervention and control groups; however, considerable improvements in glucose control were observed in the control groups as well [ 3 , 19 , 20 ]. Improvements in the control group reduced the size of the effect of the intervention in these studies. Similar outcomes were observed in other studies, which report conservative or weakened between-group differences due to improvements in the control group [ 6 , 11 , 14 , 21 ]. Therefore, solely the recruitment to a clinical trial, independent of the interventions, resulted in improvements in HbA1c [ 22 ]. When designing an interventional study, such effects have to be taken into account; otherwise, the observed effects cannot be attributed to the defined intervention due to the profound effects of confounding factors. This is especially important when including behavioral interventions, as comprehensive support of HCPs is essential to utilize the full potential of the intervention.

This article aims to provide challenges in interpreting diabetes management intervention studies gained from the STeP and ProValue studies and suggests approaches for optimizing study implementation and for avoiding pitfalls for further studies with similar medical devices.

Study implementation

Engaging participants in the intervention group.

A major difficulty in intervention studies that rely on behavioral changes is to ensure that the intervention triggers the desired effect (Table  1 ).

While in drug studies an intervention is defined by a given medication scheme that the participants simply have to follow, behavioral changes cannot be triggered that straightforward [ 23 , 24 ].

There is no single intervention strategy to guarantee intervention adherence of all participants. Furthermore, a distinction between unintentional and intentional non-adherence should be taken into account. Whereas unintentional non-adherence depends on modifiable factors like poor understanding of the treatment or low health literacy or numeracy, in intentional non-adherence, participants rationally decide to not adhere by weighing the benefits and risk of following an intervention [ 25 ].

A set of key factors influencing attempts to improve participants’ unintentional non-adherence include a clear and effective communication between HCP and participant as well as realistic assessments of participants’ health literacy and knowledge. Health literacy is essential to the ability to adhere to study intervention as well as the ability to remember the details of the recommendations made to participants during visits influences the adherence. Thus, the explanation of the specific steps of the study regimen, review of the most important details, and written instructions may increase adherence. Furthermore, the risk of non-adherence is reduced with a more familiar relationship between HCP and participant. This enables HCPs to understand participants’ beliefs, attitudes, subjective norms, cultural context or social supports, and other relevant elements that are crucial for patient’s adherence [ 26 , 27 ].

Many different behavior change techniques are already included in the standard of care in diabetes therapy, such as providing feedback on HbA1c levels and setting personal goals [ 25 , 28 ].

In the ProValue studies, for example, the diabetes management intervention comprised a blood glucose meter and the corresponding software. Subjects in the intervention group had to perform structured SMBG with an individualized testing regimen selected by the treating physicians. This selection, as well as treatment adjustments, was supported by the diabetes management software. To ensure optimal engagement of the involved HCPs with a leading role in the implementation of the intervention, numerous local investigator meetings were organized. During these investigator meetings, HCPs of intervention group sites were trained by the coordinating investigator and the sponsor’s medical advisor in the use and concept of the diabetes management program. These meetings also enabled an exchange of experiences and individual patient cases could be discussed. HCPs could thus find a way to implement the iPDM into their daily practice and individual treatment patterns, and they were motivated to appropriately use the program. It was expected that this confidence and commitment of the HCPs would increase subjects’ compliance as well [ 29 ]. Inherently, the investigator meetings and the behavior changes of HCPs varied widely, partly because the initial knowledge and skills in the use of technology already differed between the study sites. Additional attempts for a standardization of handling iPDM would have increased comparability, but would have impaired the real-life aspects and competed with the need for actions individually tailored to each patient. As the desired behavioral changes are not expected to occur in one stroke but rather gradually in the course of the study and with the active use of the iPDM, repeated training sessions of study sites were supportive.

One instrument used in the STeP study was the “peer-to-peer review” that scheduled a review of intervention group patient data and subsequent therapy and advised by an uninvolved practitioner. The independent physician applied the same principles of the iPDM used in the regular study procedures and gave direct feedback to the treating physician on how to intensify or improve therapy adjustments. Even though the iPDM used in ProValue was constructed as an ongoing circle, most improvements were observed rapidly after study start [ 20 ]. This should be considered when the optimal timing of training activities has to be determined.

To be able to evaluate whether a behavioral intervention is successful, it is necessary to verify that recommendations have actually been followed. For the ProValue studies, this takes into account whether the HCP makes use of proposals made by the software and whether participants adhere to their HCP’s advices and instructions. While adherence to some instructions like the number of daily SMBG measurements can be assessed easily when device data are downloaded, other instructions like the consideration of SMBG results for insulin dosing decisions or dietary recommendations are difficult to be traced. Information on the intentional use of the intervention might be helpful in subgroup analysis to evaluate the effect of the intervention when completely implemented. In the ProValue studies, this was captured by questionnaires about the perception of the tools among HCPs and participants and by capturing detailed information about therapy adaptations and HCP recommendations. Data to identify and trace potentials and hurdles were available during each stage of the iPDM circle. In this regard, including an analysis of the “per protocol population” that fully adhered to the study intervention procedures is highly recommended in addition to results for the intention to treat population. Bartolo et al., for instance, could not show an advantage of a diabetes management intervention compared to standard care in their study [ 14 ]. However, they reported compliance to SMBG of less than approximately 50% in both groups and they reported larger improvements among patients that were more compliant. An inadequate use that did not lead to the intended behavior changes in the study might have been a reason for the limited effects.

However, the introduction of a new medical device is time and energy consuming for HCPs. For example, software like the one used in the ProValue studies is, at least in Germany, not as common as one would expect at the present time, especially among general practitioners. Support in all technical issues should be provided on an individual level by study sponsors, adjusted to the particular knowledge and requirements, on demand and on site. Moreover, to be able to support the study sites in an appropriate and targeted manner, technical trainers have to be familiar with study aims and procedures too. In addition to technical training for the HCPs, messages from such a digital technology-based intervention have to be conveyed to the patients in a way they can understand. One lesson learned from STeP and ProValue is that visualization is an important factor. In STeP, patients graphically documented their blood glucose levels in a paper tool with color grades, while in ProValue, downloaded data were reported in a traffic light scheme (Figs.  1 and 2 ). Such visual feedback provides a link between technology, HCPs, and patients and facilitates the implementation of the intervention.

Paper tool with color grades for BG level documentation used in the STeP study

Traffic light scheme used in the ProValue studies

Management of the control group

Another major difficulty in intervention studies that rely on behavioral change is to keep participants in the control group distant from the intervention, i.e., that behavioral changes desired in the intervention group do not occur in the control group as well (Table 1 ). While in drug studies finding an adequate control (e.g., a placebo) is mostly straightforward, control group design for behavioral studies is complex and the achievement of a truly “inactive” control group that strictly stays with standard care and does not change behavior is almost impossible. Usually, randomization and blinding are the preferred tools, but the implementation is not always feasible.

Adoption of behavioral changes requires an active involvement of both patients and HCPs; blinding is therefore not an option. If HCPs are also part of the intervention, like in STeP and ProValue studies, study personnel which gained knowledge from treating the intervention group subjects could transfer this to those in the control group, at least to a given extent. Thus, cluster-randomization, i.e., randomization of the study sites rather than the individual subjects, is necessary to avoid “contamination” of subjects in the control group [ 23 ]. This means cluster-randomization is suitable for interventions that are unlikely to be available for HCPs and patients in the trial [ 23 ].

Cluster-randomization, however, limits the possibility to control for differences between the sites, such as their implementation of standard of care. As the control sites are aware of the intervention done at the intervention sites, there is the risk that they tend to try to improve therapy in control group patients too and are more attentive to patient care than usual.

Statistical power of cluster-randomized studies is limited; they thus require larger sample sizes. Based on the variation between the clusters and the expected effect size, the optimal number of clusters, i.e., study sites, and subjects per cluster can be calculated [ 30 ]. However, scheduling the required numbers is often complicated by feasibility, ethical justifiability, and affordability, which may enforce the acceptance of compromises.

A limited effect size due to improvements also in the control group is an often-observed incident, caused by the so-called Hawthorne effect. This effect is attributable to subjects’ knowledge of being part of a study, i.e., being observed and having data collected. This study effect can improve the health status of a subject without any further intervention [ 31 , 32 ]. Asking questions, for instance, induces rethinking about the current behavior and might induce respective changes [ 33 ]. Another reason for improvements in the control group is increased attention paid to the subjects by their HCPs. In principle, this increase in attention should be kept to a minimum; however, in reality, it is difficult to avoid. In addition, an intensive data gathering approach as used in the ProValue studies induces a high engagement of the participating HCPs (and also of the patients in both study groups) leading to improvements in the control group as well [ 20 ].

Also, the monitoring effort of clinical research associates (CRAs) regarding study implementation by HCPs, which is an absolutely necessary study procedure, has an impact on study implementation.

Limiting information about the intervention might be a possibility to reduce control group effect. This is not in strict compliance with the guideline for Good Clinical Practice. However, as long as the safety of study participants is paramount, a degree of concealment is accepted by research ethics committees for behavioral intervention studies [ 34 ]. According to this, all subjects and HCPs regardless of the study group have to be fully informed about the background and procedures of the study prior to the start. As such, in the ProValue studies, participants in the control group and HCPs were fully aware of the hypothesis that an iPDM and structured SMBG were expected to improve glycemic control. Participants randomized to the control group might therefore, whether or not intentionally, have sought for a comparable treatment or intensified their therapy on their own [ 31 , 35 ]. Because of the detailed assessment of therapy adjustments and recommendations of HCPs in the ProValue studies, the main triggers for behavior changes that were identified for the subjects in the intervention group could also be detected among those in the control group [ 20 ]. Control group patients typically receive “standard of care” or “treatment as usual,” but these conditions are often less defined and monitored than the interventional treatment [ 29 ].

Standard care differs across countries, hospitals, and over time, depending on the respective health care provisions and updated guidelines and technologies that are introduced at variable rates. Especially in multicenter studies, the actual implementation might vary considerably between study sites and this cannot be controlled in cluster-randomized studies [ 36 ]. A clear definition of what is regarded as “standard” is essential for the validity of a study and should receive as much attention as the definition of the intervention. Mostly, patient care within a study is rather an enhanced standard of care for all the reasons discussed above. A meta-analysis of randomized control trials (RCTs) that investigated standard care conditions in control groups of behavior change studies in patients with diabetes showed that those control group patients that received a higher quality of standard care also showed larger improvements in study outcomes, thus reducing the effect size of the intervention [ 36 ].

Nevertheless, chronic disease management like diabetes therapy is complex and, like the intervention, standard of care cannot be fully standardized but has to be adapted to the individual patient and their compliance.

Alternative study designs

Due to recruitment and “contamination” problems in interventional trials requiring behavioral changes, the realization of standard RCTs may be difficult [ 23 ]. Alternatives to the standard RCT when designing a medical device studies that rely on behavioral changes may be considered. While the above-described aspects concern the detailed implementation of a study, some variations in the general design might be considered with regard to the effect size, which was often observed to be lower than expected.

If control group effects are expected, it is essential to identify the particular reasons or triggers for behavior changes that may occur. Once identified, study procedures can be adjusted to avoid them or to even include them into the intervention. A pilot phase or study is indispensable to identify such factors and should therefore be included, especially if a large trial is planned. Therefore, more and more studies consider the additional effort of a pilot trial [ 7 , 8 , 11 , 37 , 38 ].

As the Hawthorne effect is described to be temporary and of relatively short duration [ 39 ], one approach towards a reduction of influencing the behavior of the subjects in the control group is to add one or several pretest periods to the study design [ 40 , 41 ]. This means additional data collection before and after the pretest period, without an interventional treatment in any of the groups. Randomization and initiation of the intervention starts after this period using data obtained after the pretest as baseline data. Because it is expected that the majority of improvements induced by study effects occur between the first and the second data collection, the data used for the assessment of study outcomes will not be impaired, or at least less. However, the inclusion of a pretest period is cost and time expensive and might require a pilot study to determine an adequate duration. For STeP and the ProValue studies, a 3-month pretest period would have been sufficient, as the results indicate the strongest control group effects within the first 3 months of the study. Nevertheless, because these studies were accompanied by a lot of preparations for intervention group sites, such as training sessions, a postponement of randomization procedures would have interrupted the whole study flow.

One possibility for control conditions in RCTs is using a waitlist control [ 42 ]. Subjects of the control group that are on a waiting list, i.e., they expect to receive the active intervention at a later time point, have been shown to improve less than patients that receive only standard of care throughout the whole study [ 29 ]. A waiting control group could therefore be a more efficient way to influence the effect size than an inactive control group.

Other options include a third study arm to control for enhanced standard of care and study effects. Schwartz et al. proposed a design in which one arm receives the intervention, while the control condition has two arms, each with a crossover between a waiting list with standard of care and receiving the intervention [ 43 ]. Several data collection points are required for such a design. The crossover design reduces the heterogeneity within a group due to individually tailored implementation of the intervention, increasing statistical efficiency. Nevertheless, feasibility of a crossover depends on the kind of intervention and the expected long-term effects. Additionally, inclusion of further study arms reduces statistical power, and accordingly, it requires the inclusion of more subjects which also increases financial cost and study duration [ 41 ].

In this regard, retrospective data collection could be a feasible option, but only if required data are limited to standard assessments during usual patient visits, as expected when standard of care is claimed for control group subjects.

Use of historical controls, i.e., data assessed in other independent studies that already were conducted, is another promising option if study effects shall be reduced [ 44 , 45 ]. In addition, with the use of historical controls, more resources become available for the intervention arm (which could be used for a larger sample size and therefore an increased power). Identification of a suitable control data set for the respective objective, however, is challenging, as well as the correct use of these data. In addition, the progress in treatment standards, assessment technologies, and other factors over time have to be considered.

A better separation of intervention and control group might be reached by using two separate protocols for the two groups. Consequently, all other participants such as CRAs should be exclusively assigned to one of the groups. The ProValue studies already worked with two protocols, but those were divided by the type of practice of the study sites rather than by study groups. A separate control group protocol would on the one hand enable a clear definition of “standard of care” and on the other hand allow a reduction of procedures in the control group to an absolute minimum. This applies not only to contacts between subjects and HCPs, but also between HCPs and further study staff. In addition, suitably designed informed consent forms should avoid inclusion of interventional aspects.

To prevent patients from consenting to therapy forms they may not get, a two-stage randomization could be another option. Accordingly, all patients give consent for follow-up first. An additional consent for study intervention is only provided to a randomly selected sample. Thus, patients randomized to the control group do not feel disadvantaged not receiving the intervention [ 46 ]. However, ethical concerns remain because there is only a personal consent to patient’s treatment and no full consent to the project from all patients [ 23 , 46 ].

Adaptive study designs are becoming more and more common, however, not yet in medical device studies, but rather in drug studies, as adaptive designs are in particular effective in investigating dose-response relationships. Nevertheless, some of the several different approaches might also be used for medical device studies. Adaptive design means that procedures or conditions of a study are modified during the ongoing study based on results from interim analyses. However, these changes have to be planned and defined in advance [ 47 , 48 ]. Adaptations include, e.g., randomization based on baseline data or sample size re-estimation to ensure the desired power. Implementation of adequate adaptations in studies including behavioral change has yet to be investigated. Nonetheless, such an approach could reduce the necessity of a separate pilot study and combine the exploratory and confirmatory stages of an investigation in one single study [ 38 , 49 ]. Performance of an underpowered trial may furthermore be prevented [ 47 ]. In addition, it might be a better reflection of clinical practice if those patients that prove to be compliant and susceptible for an intervention are selected. Considerations about whether or not introducing new therapeutic options (might they be behavioral changes and/or diagnostic/treatment options) are usually made by HCPs based on their experiences with the respective patients.

Based on experiences from the STeP and ProValue studies, several crucial aspects have to be considered in medical device studies when using interventions that rely on changes in behavior of study participants and their HCPs to achieve an effective implementation and significant results.

The article summarizes experiences gained from the three studies and provides suggestions for the implementation of other studies with similar medical devices.

In particular, definition of control group conditions and an integrative support of the intervention group have to be included. Improvements in the control group may reduce effect sizes and limit statistical significance; therefore, alternatives to the traditional RCT, like pretest periods or separate study protocols, are worth to be considered. As there is no ideal design for such studies, integration of experiences from other studies is essential to achieve the best possible study outcome.

Availability of data and materials

Not applicable

Abbreviations

Continuous glucose monitoring

Clinical research associates

Health care professionals

Integrated Personalized Diabetes Management

• Randomized controlled trials

Self-monitoring of blood glucose

Structured Testing Program

The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977–86.

Article   Google Scholar  

Goyal S, Lewis G, Yu C, Rotondi M, Seto E, Cafazzo JA. Evaluation of a behavioral mobile phone app intervention for the self-management of type 2 diabetes: randomized controlled trial protocol. JMIR Res Protoc. 2016;5:e174.

Article   PubMed   PubMed Central   Google Scholar  

Polonsky WH, Fisher L, Schikman CH, et al. A structured self-monitoring of blood glucose approach in type 2 diabetes encourages more frequent, intensive, and effective physician interventions: results from the STeP study. Diabetes Technol Ther. 2011;13:797–802.

Article   PubMed   Google Scholar  

Scavini M, Bosi E, Ceriello A, et al. Prospective, randomized trial on intensive SMBG management added value in non-insulin-treated T2DM patients (PRISMA): a study to determine the effect of a structured SMBG intervention. Acta Diabetol. 2013;50:663–72.

Article   CAS   PubMed   Google Scholar  

Davies MJ, D'Alessio DA, Fradkin J, et al. Management of hyperglycemia in type 2 diabetes, 2018. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2018;41:2669–701.

Nishimura A, Harashima SI, Fujita Y, et al. Effects of structured testing versus routine testing of blood glucose in diabetes self-management: a randomized controlled trial. J Diabetes Complicat. 2017;31:228–33.

Taggart L, Truesdale M, Carey ME, et al. Pilot feasibility study examining a structured self-management diabetes education programme, DESMOND-ID, targeting HbA1c in adults with intellectual disabilities. Diabet Med. 2018;35:137–46.

Silva DD, Bosco AA. An educational program for insulin self-adjustment associated with structured self-monitoring of blood glucose significantly improves glycemic control in patients with type 2 diabetes mellitus after 12 weeks: a randomized, controlled pilot study. Diabetol Metab Syndr. 2015;7:2.

Williams JS, Lynch CP, Knapp RG, Egede LE. Technology-Intensified Diabetes Education Study (TIDES) in African Americans with type 2 diabetes: study protocol for a randomized controlled trial. Trials. 2014;15:460.

Willis TA, Hartley S, Glidewell L, et al. Action to Support Practices Implement Research Evidence (ASPIRE): protocol for a cluster-randomised evaluation of adaptable implementation packages targeting ‘high impact’ clinical practice recommendations in general practice. Implement Sci. 2016;11:25.

Dobson R, Whittaker R, Jiang Y, et al. Effectiveness of text message based, diabetes self management support programme (SMS4BG): two arm, parallel randomised controlled trial. BMJ. 2018;361:k1959.

Lynch CP, Williams JS, Ruggiero KJ, Knapp RG, Egede LE. Tablet-Aided BehavioraL intervention EffecT on Self-management skills (TABLETS) for diabetes. Trials. 2016;17:157.

Article   PubMed   PubMed Central   CAS   Google Scholar  

Wild SH, Hanley J, Lewis SC, et al. Supported telemonitoring and glycemic control in people with type 2 diabetes: the Telescot diabetes pragmatic multicenter randomized controlled trial. PLoS Med. 2016;13:e1002098.

Di Bartolo P, Nicolucci A, Cherubini V, Iafusco D, Scardapane M, Rossi MC. Young patients with type 1 diabetes poorly controlled and poorly compliant with self-monitoring of blood glucose: can technology help? Results of the i-NewTrend randomized clinical trial. Acta Diabetol. 2017;54:393–402.

Article   PubMed   CAS   Google Scholar  

Lalic NM, Lalic K, Jotic A, et al. The impact of structured self-monitoring of blood glucose combined with intensive education on HbA1c levels, hospitalizations, and quality-of-life parameters in insulin-treated patients with diabetes at primary care in Serbia: the multicenter SPA-EDU study. J Diabetes Sci Technol. 2017;11:746–52.

Agarwal P, Mukerji G, Desveaux L, et al. Mobile app for improved self-management of type 2 diabetes: multicenter pragmatic randomized controlled trial. JMIR Mhealth Uhealth. 2019;7:e10321.

Polonsky W, Fisher L, Schikman C, et al. The value of episodic, intensive blood glucose monitoring in non-insulin treated persons with type 2 diabetes: design of the Structured Testing Program (STeP) study, a cluster-randomised, clinical trial [NCT00674986]. BMC Fam Pract. 2010;11:37.

Kulzer B, Daenschel W, Daenschel I, et al. Integrated personalized diabetes management (PDM): design of the ProValue studies: prospective, cluster-randomized, controlled, intervention trials for evaluation of the effectiveness and benefit of PDM in patients with insulin-treated type 2 diabetes. J Diabetes Sci Technol. 2016;10:772–81.

Polonsky WH, Fisher L, Schikman CH, et al. Structured self-monitoring of blood glucose significantly reduces A1C levels in poorly controlled, noninsulin-treated type 2 diabetes: results from the Structured Testing Program study. Diabetes Care. 2011;34:262–7.

Kulzer B, Daenschel W, Daenschel I, et al. Integrated personalized diabetes management improves glycemic control in patients with insulin-treated type 2 diabetes: results of the PDM-ProValue study program. Diabetes Res Clin Pract. 2018;144:200–12.

Bosi E, Scavini M, Ceriello A, et al. Intensive structured self-monitoring of blood glucose and glycemic control in noninsulin-treated type 2 diabetes: the PRISMA randomized trial. Diabetes Care. 2013;36:2887–94.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Gale EA, Beattie SD, Hu J, Koivisto V, Tan MH. Recruitment to a clinical trial improves glycemic control in patients with diabetes. Diabetes Care. 2007;30:2989–92.

Campbell R, Peters T, Grant C, Quilty B, Dieppe P. Adapting the randomized consent (Zelen) design for trials of behavioural interventions for chronic disease: feasibility study. J Health Serv Res Policy. 2005;10:220–5.

Heinemann L. Future of diabetes technology. J Diabetes Sci Technol. 2017;11:863–9.

Klonoff DC. Behavioral theory: the missing ingredient for digital health tools to change behavior and increase adherence. J Diabetes Sci Technol. 2019;13:276–81.

Martin LR, Williams SL, Haskard KB, Dimatteo MR. The challenge of patient adherence. Ther Clin Risk Manag. 2005;1:189–99.

PubMed   PubMed Central   Google Scholar  

Reach G. Can technology improve adherence to long-term therapies? J Diabetes Sci Technol. 2009;3:492–9.

Michie S, van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci. 2011;6:42.

Mohr DC, Spring B, Freedland KE, et al. The selection and design of control conditions for randomized controlled trials of psychological interventions. Psychother Psychosom. 2009;78:275–84.

van Breukelen GJP, Candel MJJM. Efficiency loss because of varying cluster size in cluster randomized trials is smaller than literature suggests. Stat Med. 2012;31:397–400.

Abraham MB, Nicholas JA, Crone M, Ly TT, Davis EA, Jones TW. The importance of the Hawthorne effect on psychological outcomes unveiled in a randomized controlled trial of diabetes technology. J Diabetes Sci Technol. 2018;12:735–6.

Campbell BF, Sengupta S, Santos C, Lorig KR. Balanced incomplete block design: description, case study, and implications for practice. Health Educ Q. 1995;22:201–10.

Wilding S, Conner M, Sandberg T, et al. The question-behaviour effect: a theoretical and methodological review and meta-analysis. Eur Rev Soc Psychol. 2016;27:196–230.

Tai MC-T. Deception and informed consent in social, behavioral, and educational research (SBER). Tzu Chi Med J. 2012;24:218–22.

Byrd-Bredbenner C, Wu F, Spaccarotella K, Quick V, Martin-Biggers J, Zhang Y. Systematic review of control groups in nutrition education intervention research. Int J Behav Nutr Phys Act. 2017;14:91.

Ayling K, Brierley S, Johnson B, Heller S, Eiser C. How standard is standard care? Exploring control group outcomes in behaviour change interventions for young people with type 1 diabetes. Psychol Health. 2015;30:85–103.

House A, Bryant L, Russell AM, et al. Managing with learning disability and diabetes: OK-Diabetes - a case-finding study and feasibility randomised controlled trial. Health Technol Assess (Winchester, England). 2018;22:1–328.

Karlsen B, Oftedal B, Stangeland Lie S, et al. Assessment of a web-based guided self-determination intervention for adults with type 2 diabetes in general practice: a study protocol. BMJ Open. 2016;6:e013026.

McCambridge J, Witton J, Elbourne DR. Systematic review of the Hawthorne effect: new concepts are needed to study research participation effects. J Clin Epidemiol. 2014;67:267–77.

Becker H, Roberts G, Voelmeck W. Explanations for improvement in both experimental and control groups. West J Nurs Res. 2003;25:746–55.

Eccles M, Grimshaw J, Campbell M, Ramsay C. Research designs for studies evaluating the effectiveness of change and improvement strategies. Qual Saf Health Care. 2003;12:47–52.

Kim H, Park EM, Henry C, Ward SE, Song MK. Control conditions that are neither usual care nor no treatment in randomized trials of psychoeducational palliative care interventions: a systematic review. Am J Hosp Palliat Care. 2019;36:339–47.

Schwartz CE, Chesney MA, Irvine MJ, Keefe FJ. The control group dilemma in clinical research: applications for psychosocial and behavioral medicine trials. Psychosom Med. 1997;59:362–71.

Viele K, Berry S, Neuenschwander B, et al. Use of historical control data for assessing treatment effects in clinical trials. Pharm Stat. 2014;13:41–54.

Galwey NW. Supplementation of a clinical trial by historical control data: is the prospect of dynamic borrowing an illusion? Stat Med. 2017;36:899–916.

Stott DJ, Langhorne P, Rodgers H. Informed consent. Two stage randomisation and consent would overcome many problems. BMJ (Clinical research ed). 1997;315:253.

CAS   Google Scholar  

Pallmann P, Bedding AW, Choodari-Oskooei B, et al. Adaptive designs in clinical trials: why use them, and how to run and report them. BMC Med. 2018;16:29.

Chow SC. Adaptive clinical trial design. Annu Rev Med. 2014;65:405–15.

Wang SJ, Hung HM, O'Neill R. Paradigms for adaptive statistical information designs: practical experiences and strategies. Stat Med. 2012;31:3011–23.

Download references

Acknowledgements

The authors would like to thank Guido Freckmann for the support in writing the manuscript.

Scientific writing was funded by Roche Diabetes Care. The reported STeP study was funded by Roche Diabetes Care. Roche Diabetes Care was involved in the concept and design of the reported STeP and ProValue studies. Roche Diabetes Care contributed to subsequent revisions of the manuscript and approved the final version.

Author information

Authors and affiliations.

Roche Diabetes Care GmbH, Mannheim, Germany

Bettina Petersen, Iris Vesper, Bernhild Pachwald & Nicole Dagenbach

Institut für Diabetes-Technologie, Forschungs- und Entwicklungsgesellschaft mbH an der Universität Ulm, Lise-Meitner-Strasse 8/2, 89081, Ulm, Germany

Sina Buck & Delia Waldenmaier

Science Consulting in Diabetes GmbH, Neuss, Germany

Lutz Heinemann

You can also search for this author in PubMed   Google Scholar

Contributions

BP, IV, BP, and ND were involved in the concept and design of the STeP and ProValue studies. DW performed a literature search and wrote the first draft of the manuscript, and all authors contributed to subsequent revisions of the manuscript and approved the final version.

Corresponding author

Correspondence to Sina Buck .

Ethics declarations

Ethics approval and consent to participate, consent for publication, competing interests.

DW and SB are employees of the IfDT (Institut für Diabetes-Technologie, Forschungs- und Entwicklungsgesellschaft mbH an der Universität Ulm, Ulm, Germany), which carries out clinical studies on the evaluation of BG meters and medical devices for diabetes therapy on its own initiative and on behalf of various companies. IfDT has received speakers’ honoraria or consulting fees from Abbott, Ascensia, Dexcom, LifeScan, Menarini Diagnostics, Metronom Health, Novo Nordisk, Roche, Sanofi, Sensile, and Ypsomed.

BP, IV, BP, and ND are employees of Roche Diabetes Care.

LH is a consultant for a number of companies that develop novel diagnostic and therapeutic options for diabetes treatment, e.g., Roche Diabetes Care, Becton Dickinson, Lifecare, and Berlin-Chemie. LH is a shareholder of the Profil Institut für Stoffwechselforschung, Neuss, Germany, and ProSciento, San Diego, USA.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Petersen, B., Vesper, I., Pachwald, B. et al. Diabetes management intervention studies: lessons learned from two studies. Trials 22 , 61 (2021). https://doi.org/10.1186/s13063-020-05017-3

Download citation

Received : 21 January 2020

Accepted : 31 December 2020

Published : 18 January 2021

DOI : https://doi.org/10.1186/s13063-020-05017-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Study performance
• Behavior-based
• Medical device studies
• Adaptive study design
• Study effect

ISSN: 1745-6215

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

• © 2023

Advances in Diabetes Research and Management

• Rana Noor 0

Faculty of Dentistry Department of Biochemistry, Jamia Millia Islamia, New Delhi, India

You can also search for this editor in PubMed   Google Scholar

• Presents latest developments in Diabetes research and management
• Reviews role of nanotechnology for diagnosis and treatment for Diabetes
• Discusses pathophysiology of Diabetes Mellitus complications

6348 Accesses

7 Citations

2 Altmetric

• Table of contents

About this book

Editors and affiliations, about the editor, bibliographic information.

• Publish with us

Buying options

• Available as EPUB and PDF
• Read on any device
• Instant download
• Own it forever
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Other ways to access

This is a preview of subscription content, log in via an institution to check for access.

Table of contents (11 chapters)

Front matter, biochemical assay for measuring diabetes mellitus.

• Taniya Sur, Aishwariya Das, Sheerin Bashar, Sambit Tarafdar, Bhanumati Sarkar, Nithar Ranjan Madhu

Diabetes and Other Comorbidities: Microvascular and Macrovascular Diseases Diabetes and Cancer

• V. Nithya, P. Sangavi, R. Srinithi, K. T. Nachammai, S. Gowtham Kumar, D. Prabu et al.

Diabetes and Cardiovascular Disorder

• S. Santhi Priya, K. Kumar Ebenezar

Diabetes and Neurological Disorder

• Iyshwarya Bhaskar Kalarani, Ramakrishnan Veerabathiran

Diabetic and Nephropathy

• Langeswaran Kulanthaivel, Geevaprabhakaran Ganesan, Chandrashekar Kirubhanand, Gowtham Kumar Subbaraj

Technology in the Management of Type 1 and Type 2 Diabetes Mellitus: Recent Status and Future Prospects

• Titas Biswas, Biplab Kumar Behera, Nithar Ranjan Madhu

The Broader Aspects of Treating Diabetes with the Application of Nanobiotechnology

• Rupak Roy, Aditi Chakraborty, Kartik Jana, Bhanumati Sarkar, Paramita Biswas, Nithar Ranjan Madhu

A Comprehensive Pharmacological Appraisal of Indian Traditional Medicinal Plants with Anti-diabetic Potential

• Chandan Kumar Acharya, Balaram Das, Nithar Ranjan Madhu, Somnath Sau, Manna De, Bhanumati Sarkar

Diabetes Management: From “Painful” Pricks to “Pain-Free” Bliss

• Bhuvaneswari Ponnusamy, Ponnulakshmi Rajagopal, Raktim Mukherjee, Swetha Panneerselvam, Selvaraj Jayaraman

Diabetes Mellitus and iPSC-Based Therapy

• Dibyashree Chhetri, Rajesh Nanda Amarnath, Sunita Samal, Kanagaraj Palaniyandi, Dhanavathy Gnanasampanthapandian

Influence of Ketogenic Diet on Diabetes

• Natesan Sella Raja, Varsha Singh, Subhashree Sivakumar
• Diabetes Mellitus
• Nanotechnology
• Glucose sensors
• Insulin Delivery
• Microvascular and Macrovascular complications

Prof. Rana Noor is currently working at Faculty of Dentistry, Jamia Millia Islamia, New Delhi, India. She pursued her Ph.D. in Biochemistry from Aligarh Muslim University, India in the year 2003. Her research work is highly interdisciplinary, spanning a wide range in protein biochemistry, enzymology and application of biochemistry in dentistry. She is an editorial board member of various reputed journals.

Book Title : Advances in Diabetes Research and Management

Editors : Rana Noor

DOI : https://doi.org/10.1007/978-981-19-0027-3

Publisher : Springer Singapore

eBook Packages : Biomedical and Life Sciences , Biomedical and Life Sciences (R0)

Copyright Information : The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023

Hardcover ISBN : 978-981-19-0026-6 Published: 10 March 2023

Softcover ISBN : 978-981-19-0029-7 Published: 10 March 2024

eBook ISBN : 978-981-19-0027-3 Published: 09 March 2023

Edition Number : 1

Number of Pages : XII, 256

Number of Illustrations : 1 b/w illustrations

Topics : Human Physiology , Chemistry/Food Science, general , Physiology , Biological Techniques , Metabolic Diseases

Policies and ethics

• Find a journal
• Track your research

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals

Diabetes articles from across Nature Portfolio

Diabetes describes a group of metabolic diseases characterized by high blood sugar levels. Diabetes can be caused by the pancreas not producing insulin (type 1 diabetes) or by insulin resistance (cells do not respond to insulin; type 2 diabetes).

Macrophage vesicles in antidiabetic drug action

Thiazolidinediones (TZDs) are potent insulin-sensitizing drugs, but their use is accompanied by adverse side-effects. Rohm et al. now report that TZD-stimulated macrophages release miR-690-containing vesicles that improve insulin sensitization and bypass unwanted side-effects.

• Rinke Stienstra
• Eric Kalkhoven

Genetic risk variants lead to type 2 diabetes development through different pathways

The largest genome-wide association study for type 2 diabetes so far, which included several ancestry groups, led to the identification of eight clusters of genetic risk variants. The clusters capture different biological pathways that contribute to the disease, and some clusters are associated with vascular complications.

Blood glucose concentration measurement without finger pricking

A new sensor that detects optoacoustic signals generated by mid-infrared light enables measurement of glucose concentration from intracutaneous tissue rich in blood. This technology does not rely on glucose measurements in interstitial fluid or blood sampling and might yield the next generation of non-invasive glucose-sensing devices for improved diabetes management.

Related Subjects

• Diabetes complications
• Diabetes insipidus
• Gestational diabetes
• Type 1 diabetes
• Type 2 diabetes

Latest Research and Reviews

Nicotinamide Mononucleotide improves oocyte maturation of mice with type 1 diabetes

• Fucheng Guo
• Xiaoling Zhang

Folic acid supplementation on inflammation and homocysteine in type 2 diabetes mellitus: systematic review and meta-analysis of randomized controlled trials

• Kabelo Mokgalaboni
• Given. R. Mashaba
• Sogolo. L. Lebelo

The usefulness of quantitative 99m Tc-HMPAO WBC SPECT/CT for predicting lower extremity amputation in diabetic foot infection

• Soo Bin Park
• Chae Hong Lim
• Jung Mi Park

An estrogen receptor α-derived peptide improves glucose homeostasis during obesity

Type 2 diabetes mellitus is an increasing global health issue, which is caused by systemic insulin resistance. Here, the authors show a ligand-independent effect of hepatic ERα in regulating insulin sensitivity and identify an ERα-derived peptide that functions as an insulin sensitizer.

• Wanbao Yang
• Shaodong Guo

Effective interventions in preventing gestational diabetes mellitus: A systematic review and meta-analysis

Takele et al. conduct a systematic review and meta-analysis of the effects of intervention characteristics on preventing gestational diabetes. Group or healthcare facility-based physical activity interventions are found to be more effective in preventing gestational diabetes than community-based interventions.

• Wubet Worku Takele
• Kimberly K. Vesco

Illuminating the complete ß-cell mass of the human pancreas- signifying a new view on the islets of Langerhans

The pancreatic islets of Langerhans play a pivotal role in regulating blood glucose homeostasis through the regulated secretion of the hormones insulin and glucagon. Here, the authors use deep tissue 3D imaging to re-construct the entire human pancreas at microscopic resolution and display previously unrecognized heterogeneities in the islet’s cellularity with pre-clinical and clinical implications.

• Joakim Lehrstrand
• Wayne I. L. Davies
• Ulf Ahlgren

News and Comment

Repurposing a diabetes drug to treat Parkinson’s disease

In a multicenter clinical trial, patients with early-stage Parkinson’s disease treated with lixisenatide, a drug currently used for the treatment of diabetes, showed improvement in their motor scores compared with those on placebo.

• Sonia Muliyil

A novel system for non-invasive measurement of blood levels of glucose

• Olivia Tysoe

Diabetes drug slows development of Parkinson’s disease

The drug, which is in the same family as blockbuster weight-loss drugs such as Wegovy, slowed development of symptoms by a small but statistically significant amount.

Metformin acts through appetite-suppressing metabolite: Lac-Phe

• Shimona Starling

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

• Previous Article
• Next Article

Research Design and Methods

Article information, literature review of type 2 diabetes management and health literacy.

• Split-Screen
• Article contents
• Figures & tables
• Supplementary Data
• Peer Review
• Open the PDF for in another window
• Cite Icon Cite
• Get Permissions

Rulla Alsaedi , Kimberly McKeirnan; Literature Review of Type 2 Diabetes Management and Health Literacy. Diabetes Spectr 1 November 2021; 34 (4): 399–406. https://doi.org/10.2337/ds21-0014

Download citation file:

• Ris (Zotero)
• Reference Manager

The purpose of this literature review was to identify educational approaches addressing low health literacy for people with type 2 diabetes. Low health literacy can lead to poor management of diabetes, low engagement with health care providers, increased hospitalization rates, and higher health care costs. These challenges can be even more profound among minority populations and non-English speakers in the United States.

A literature search and standard data extraction were performed using PubMed, Medline, and EMBASE databases. A total of 1,914 articles were identified, of which 1,858 were excluded based on the inclusion criteria, and 46 were excluded because of a lack of relevance to both diabetes management and health literacy. The remaining 10 articles were reviewed in detail.

Patients, including ethnic minorities and non-English speakers, who are engaged in diabetes education and health literacy improvement initiatives and ongoing follow-up showed significant improvement in A1C, medication adherence, medication knowledge, and treatment satisfaction. Clinicians considering implementing new interventions to address diabetes care for patients with low health literacy can use culturally tailored approaches, consider ways to create materials for different learning styles and in different languages, engage community health workers and pharmacists to help with patient education, use patient-centered medication labels, and engage instructors who share cultural and linguistic similarities with patients to provide educational sessions.

This literature review identified a variety of interventions that had a positive impact on provider-patient communication, medication adherence, and glycemic control by promoting diabetes self-management through educational efforts to address low health literacy.

Diabetes is the seventh leading cause of death in the United States, and 30.3 million Americans, or 9.4% of the U.S. population, are living with diabetes ( 1 , 2 ). For successful management of a complicated condition such as diabetes, health literacy may play an important role. Low health literacy is a well-documented barrier to diabetes management and can lead to poor management of medical conditions, low engagement with health care providers (HCPs), increased hospitalizations, and, consequently, higher health care costs ( 3 – 5 ).

The Healthy People 2010 report ( 6 ) defined health literacy as the “degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.” Diabetes health literacy also encompasses a wide range of skills, including basic knowledge of the disease state, self-efficacy, glycemic control, and self-care behaviors, which are all important components of diabetes management ( 3 – 5 , 7 ). According to the Institute of Medicine’s Committee on Health Literacy, patients with poor health literacy are twice as likely to have poor glycemic control and were found to be twice as likely to be hospitalized as those with adequate health literacy ( 8 ). Associations between health literacy and health outcomes have been reported in many studies, the first of which was conducted in 1995 in two public hospitals and found that many patients had inadequate health literacy and could not perform the basic reading tasks necessary to understand their treatments and diagnoses ( 9 ).

Evaluation of health literacy is vital to the management and understanding of diabetes. Several tools for assessing health literacy have been evaluated, and the choice of which to use depends on the length of the patient encounter and the desired depth of the assessment. One widely used literacy assessment tool, the Test of Functional Health Literacy in Adults (TOFHLA), consists of 36 comprehension questions and four numeric calculations ( 10 ). Additional tools that assess patients’ reading ability include the Rapid Estimate of Adult Literacy in Medicine (REALM) and the Literacy Assessment for Diabetes. Tests that assess diabetes numeracy skills include the Diabetes Numeracy Test, the Newest Vital Sign (NVS), and the Single-Item Literacy Screener (SILS) ( 11 ).

Rates of both diabetes and low health literacy are higher in populations from low socioeconomic backgrounds ( 5 , 7 , 12 ). People living in disadvantaged communities face many barriers when seeking health care, including inconsistent housing, lack of transportation, financial difficulties, differing cultural beliefs about health care, and mistrust of the medical professions ( 13 , 14 ). People with high rates of medical mistrust tend to be less engaged in their care and to have poor communication with HCPs, which is another factor HCPs need to address when working with their patients with diabetes ( 15 ).

The cost of medical care for people with diabetes was $327 billion in 2017, a 26% increase since 2012 ( 1 , 16 ). Many of these medical expenditures are related to hospitalization and inpatient care, which accounts for 30% of total medical costs for people with diabetes ( 16 ).

People with diabetes also may neglect self-management tasks for various reasons, including low health literacy, lack of diabetes knowledge, and mistrust between patients and HCPs ( 7 , 15 ).

These challenges can be even more pronounced in vulnerable populations because of language barriers and patient-provider mistrust ( 17 – 19 ). Rates of diabetes are higher among racial and ethnic minority groups; 15.1% of American Indians and Alaskan Natives, 12.7% of Non-Hispanic Blacks, 12.1% of Hispanics, and 8% of Asian Americans have diagnosed diabetes, compared with 7.4% of non-Hispanic Whites ( 1 ). Additionally, patient-provider relationship deficits can be attributed to challenges with communication, including HCPs’ lack of attention to speaking slowly and clearly and checking for patients’ understanding when providing education or gathering information from people who speak English as a second language ( 15 ). White et al. ( 15 ) demonstrated that patients with higher provider mistrust felt that their provider’s communication style was less interpersonal and did not feel welcome as part of the decision-making process.

To the authors’ knowledge, there is no current literature review evaluating interventions focused on health literacy and diabetes management. There is a pressing need for such a comprehensive review to provide a framework for future intervention design. The objective of this literature review was to gather and summarize studies of health literacy–based diabetes management interventions and their effects on overall diabetes management. Medication adherence and glycemic control were considered secondary outcomes.

Search Strategy

A literature review was conducted using the PubMed, Medline, and EMBASE databases. Search criteria included articles published between 2015 and 2020 to identify the most recent studies on this topic. The search included the phrases “diabetes” and “health literacy” to specifically focus on health literacy and diabetes management interventions and was limited to original research conducted in humans and published in English within the defined 5-year period. Search results were exported to Microsoft Excel for evaluation.

Study Selection

Initial screening of the articles’ abstracts was conducted using the selection criteria to determine which articles to include or exclude ( Figure 1 ). The initial search results were reviewed for the following inclusion criteria: original research (clinical trials, cohort studies, and cross-sectional studies) conducted in human subjects with type 2 diabetes in the United States, and published in English between 2015 and 2020. Articles were considered to be relevant if diabetes was included as a medical condition in the study and an intervention was made to assess or improve health literacy. Studies involving type 1 diabetes or gestational diabetes and articles that were viewpoints, population surveys, commentaries, case reports, reviews, or reports of interventions conducted outside of the United States were excluded from further review. The criteria requiring articles to be from the past 5 years and from the United States were used because of the unique and quickly evolving nature of the U.S. health care system. Articles published more than 5 years ago or from other health care systems may have contributed information that was not applicable to or no longer relevant for HCPs in the United States. Articles were screened and reviewed independently by both authors. Disagreements were resolved through discussion to create the final list of articles for inclusion.

PRISMA diagram of the article selection process.

Data Extraction

A standard data extraction was performed for each included article to obtain information including author names, year of publication, journal, study design, type of intervention, primary outcome, tools used to assess health literacy or type 2 diabetes knowledge, and effects of intervention on overall diabetes management, glycemic control, and medication adherence.

A total of 1,914 articles were collected from a search of the PubMed, MEDLINE, and EMBASE databases, of which 1,858 were excluded based on the inclusion and exclusion criteria. Of the 56 articles that met criteria for abstract review, 46 were excluded because of a lack of relevance to both diabetes management and health literacy. The remaining 10 studies identified various diabetes management interventions, including diabetes education tools such as electronic medication instructions and text message–based interventions, technology-based education videos, enhanced prescription labels, learner-based education materials, and culturally tailored interventions ( 15 , 20 – 28 ). Figure 1 shows the PRISMA diagram of the article selection process, and Table 1 summarizes the findings of the article reviews ( 15 , 20 – 28 ).

Findings of the Article Reviews (15,20–28)

SAHLSA, Short Assessment of Health Literacy for Spanish Adults.

Medical mistrust and poor communication are challenging variables in diabetes education. White et al. ( 15 ) examined the association between communication quality and medical mistrust in patients with type 2 diabetes. HCPs at five health department clinics received training in effective health communication and use of the PRIDE (Partnership to Improve Diabetes Education) toolkit in both English and Spanish, whereas control sites were only exposed to National Diabetes Education Program materials without training in effective communication. The study evaluated participant communication using several tools, including the Communication Assessment Tool (CAT), Interpersonal Processes of Care (IPC-18), and the Short Test of Functional Health Literacy in Adults (s-TOFHLA). The authors found that higher levels of mistrust were associated with lower CAT and IPC-18 scores.

Patients with type 2 diabetes are also likely to benefit from personalized education delivery tools such as patient-centered labeling (PCL) of prescription drugs, learning style–based education materials, and tailored text messages ( 24 , 25 , 27 ). Wolf et al. ( 27 ) investigated the use of PCL in patients with type 2 diabetes and found that patients with low health literacy who take medication two or more times per day have higher rates of proper medication use when using PCL (85.9 vs. 77.4%, P = 0.03). The objective of the PCL intervention was to make medication instructions and other information on the labels easier to read to improve medication use and adherence rates. The labels incorporated best-practice strategies introduced by the Institute of Medicine for the Universal Medication Schedule. These strategies prioritize medication information, use of larger font sizes, and increased white space. Of note, the benefits of PCL were largely seen with English speakers. Spanish speakers did not have substantial improvement in medication use or adherence, which could be attributed to language barriers ( 27 ).

Nelson et al. ( 25 ) analyzed patients’ engagement with an automated text message approach to supporting diabetes self-care activities in a 12-month randomized controlled trial (RCT) called REACH (Rapid Education/Encouragement and Communications for Health) ( 25 ). Messages were tailored based on patients’ medication adherence, the Information-Motivation-Behavioral Skills model of health behavior change, and self-care behaviors such as diet, exercise, and self-monitoring of blood glucose. Patients in this trial were native English speakers, so further research to evaluate the impact of the text message intervention in patients with limited English language skills is still needed. However, participants in the intervention group reported higher engagement with the text messages over the 12-month period ( 25 ).

Patients who receive educational materials based on their learning style also show significant improvement in their diabetes knowledge and health literacy. Koonce et al. ( 24 ) developed and evaluated educational materials based on patients’ learning style to improve health literacy in both English and Spanish languages. The materials were made available in multiple formats to target four different learning styles, including materials for visual learners, read/write learners, auditory learners, and kinesthetic learners. Spanish-language versions were also available. Researchers were primarily interested in measuring patients’ health literacy and knowledge of diabetes. The intervention group received materials in their preferred learning style and language, whereas the control group received standard of care education materials. The intervention group showed significant improvement in diabetes knowledge and health literacy, as indicated by Diabetes Knowledge Test (DKT) scores. More participants in the intervention group reported looking up information about their condition during week 2 of the intervention and showed an overall improvement in understanding symptoms of nerve damage and types of food used to treat hypoglycemic events. However, the study had limited enrollment of Spanish speakers, making the applicability of the results to Spanish-speaking patients highly variable.

Additionally, findings by Hofer et al. ( 22 ) suggest that patients with high A1C levels may benefit from interventions led by community health workers (CHWs) to bridge gaps in health literacy and equip patients with the tools to make health decisions. In this study, Hispanic and African American patients with low health literacy and diabetes not controlled by oral therapy benefited from education sessions led by CHWs. The CHWs led culturally tailored support groups to compare the effects of educational materials provided in an electronic format (via iDecide) and printed format on medication adherence and self-efficacy. The study found increased adherence with both formats, and women, specifically, had a significant increase in medication adherence and self-efficacy. One of the important aspects of this study was that the CHWs shared cultural and linguistic characteristics with the patients and HCPs, leading to increased trust and satisfaction with the information presented ( 22 ).

Kim et al. ( 23 ) found that Korean-American participants benefited greatly from group education sessions that provided integrated counseling led by a team of nurses and CHW educators. The intervention also had a health literacy component that focused on enhancing skills such as reading food package labels, understanding medical terminology, and accessing health care services. This intervention led to a significant reduction of 1–1.3% in A1C levels in the intervention group. The intervention established the value of collaboration between CHW educators and nurses to improve health information delivery and disease management.

A collaboration between CHW educators and pharmacists was also shown to reinforce diabetes knowledge and improve health literacy. Sharp et al. ( 26 ) conducted a cross-over study in four primary care ambulatory clinics that provided care for low-income patients. The study found that patients with low health literacy had more visits with pharmacists and CHWs than those with high health literacy. The CHWs provided individualized support to reinforce diabetes self-management education and referrals to resources such as food, shelter, and translation services. The translation services in this study were especially important for building trust with non-English speakers and helping patients understand their therapy. Similar to other studies, the CHWs shared cultural and linguistic characteristics with their populations, which helped to overcome communication-related and cultural barriers ( 23 , 26 ).

The use of electronic tools or educational videos yielded inconclusive results with regard to medication adherence. Graumlich et al. ( 20 ) implemented a new medication planning tool called Medtable within an electronic medical record system in several outpatient clinics serving patients with type 2 diabetes. The tool was designed to organize medication review and patient education. Providers can use this tool to search for medication instructions and actionable language that are appropriate for each patient’s health literacy level. The authors found no changes in medication knowledge or adherence, but the intervention group reported higher satisfaction. On the other hand, Yeung et al. ( 28 ) showed that pharmacist-led online education videos accessed using QR codes affixed to the patients’ medication bottles and health literacy flashcards increased patients’ medication adherence in an academic medical hospital.

Goessl et al. ( 21 ) found that patients with low health literacy had significantly higher retention of information when receiving evidence-based diabetes education through a DVD recording than through an in-person group class. This 18-month RCT randomized participants to either the DVD or in-person group education and assessed their information retention through a teach-back strategy. The curriculum consisted of diabetes prevention topics such as physical exercise, food portions, and food choices. Participants in the DVD group had significantly higher retention of information than those in the control (in-person) group. The authors suggested this may have been because participants in the DVD group have multiple opportunities to review the education material.

Management of type 2 diabetes remains a challenge for HCPs and patients, in part because of the challenges discussed in this review, including communication barriers between patients and HCPs and knowledge deficits about medications and disease states ( 29 ). HCPs can have a positive impact on the health outcomes of their patients with diabetes by improving patients’ disease state and medication knowledge.

One of the common themes identified in this literature review was the prevalence of culturally tailored diabetes education interventions. This is an important strategy that could improve diabetes outcomes and provide an alternative approach to diabetes self-management education when working with patients from culturally diverse backgrounds. HCPs might benefit from using culturally tailored educational approaches to improve communication with patients and overcome the medical mistrust many patients feel. Although such mistrust was not directly correlated with diabetes management, it was noted that patients who feel mistrustful tend to have poor communication with HCPs ( 20 ). Additionally, Latino/Hispanic patients who have language barriers tend to have poor glycemic control ( 19 ). Having CHWs work with HCPs might mitigate some patient-provider communication barriers. As noted earlier, CHWs who share cultural and linguistic characteristics with their patient populations have ongoing interactions and more frequent one-on-one encounters ( 12 ).

Medication adherence and glycemic control are important components of diabetes self-management, and we noted that the integration of CHWs into the diabetes health care team and the use of simplified medication label interventions were both successful in improving medication adherence ( 23 , 24 ). The use of culturally tailored education sessions and the integration of pharmacists and CHWs into the management of diabetes appear to be successful in reducing A1C levels ( 12 , 26 ). Electronic education tools and educational videos alone did not have an impact on medication knowledge or information retention in patients with low health literacy, but a combination of education tools and individualized sessions has the potential to improve diabetes medication knowledge and overall self-management ( 20 , 22 , 30 ).

There were several limitations to our literature review. We restricted our search criteria to articles published in English and studies conducted within the United States to ensure that the results would be relevant to U.S. HCPs. However, these limitations may have excluded important work on this topic. Additional research expanding this search beyond the United States and including articles published in other languages may demonstrate different outcomes. Additionally, this literature review did not focus on A1C as the primary outcome, although A1C is an important indicator of diabetes self-management. A1C was chosen as the method of evaluating the impact of health literacy interventions in patients with diabetes, but other considerations such as medication adherence, impact on comorbid conditions, and quality of life are also important factors.

The results of this work show that implementing health literacy interventions to help patients manage type 2 diabetes can have beneficial results. However, such interventions can have significant time and monetary costs. The potential financial and time costs of diabetes education interventions were not evaluated in this review and should be taken into account when designing interventions. The American Diabetes Association estimated the cost of medical care for people with diabetes to be $327 billion in 2017, with the majority of the expenditure related to hospitalizations and nursing home facilities ( 16 ). Another substantial cost of diabetes that can be difficult to measure is treatment for comorbid conditions and complications such as cardiovascular and renal diseases.

Interventions designed to address low health literacy and provide education about type 2 diabetes could be a valuable asset in preventing complications and reducing medical expenditures. Results of this work show that clinicians who are considering implementing new interventions may benefit from the following strategies: using culturally tailored approaches, creating materials for different learning styles and in patients’ languages, engaging CHWs and pharmacists to help with patient education, using PCLs for medications, and engaging education session instructors who share patients’ cultural and linguistic characteristics.

Diabetes self-management is crucial to improving health outcomes and reducing medical costs. This literature review identified interventions that had a positive impact on provider-patient communication, medication adherence, and glycemic control by promoting diabetes self-management through educational efforts to address low health literacy. Clinicians seeking to implement diabetes care and education interventions for patients with low health literacy may want to consider drawing on the strategies described in this article. Providing culturally sensitive education that is tailored to patients’ individual learning styles, spoken language, and individual needs can improve patient outcomes and build patients’ trust.

Duality of Interest

No potential conflicts of interest relevant to this article were reported.

Author Contributions

Both authors conceptualized the literature review, developed the methodology, analyzed the data, and wrote, reviewed, and edited the manuscript. R.A. collected the data. K.M. supervised the review. K.M. is the guarantor of this work and, as such, has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation

Portions of this research were presented at the Washington State University College of Pharmacy and Pharmaceutical Sciences Honors Research Day in April 2019.

Email alerts

• Online ISSN 1944-7353
• Print ISSN 1040-9165
• Diabetes Care
• Clinical Diabetes
• Diabetes Spectrum
• Standards of Medical Care in Diabetes
• Scientific Sessions Abstracts
• BMJ Open Diabetes Research & Care
• ShopDiabetes.org
• ADA Professional Books

Clinical Compendia

• Clinical Compendia Home
• Latest News
• DiabetesPro SmartBrief
• Special Collections
• DiabetesPro®
• Diabetes Food Hub™
• Insulin Affordability
• Know Diabetes By Heart™
• About the ADA
• Journal Policies
• For Reviewers
• Advertising in ADA Journals
• Reprints and Permission for Reuse
• Copyright Notice/Public Access Policy
• ADA Professional Membership
• ADA Member Directory
• Diabetes.org
• X (Twitter)
• Cookie Policy
• Accessibility
• Terms & Conditions
• Get Adobe Acrobat Reader
• © Copyright American Diabetes Association

This Feature Is Available To Subscribers Only

Sign In or Create an Account

ORIGINAL RESEARCH article

A self-report measure of diabetes self-management for type 1 and type 2 diabetes: the diabetes self-management questionnaire-revised (dsmq-r) – clinimetric evidence from five studies.

• 1 Research Institute of the Diabetes Academy Mergentheim (FIDAM), Diabetes Center Mergentheim (DZM), Bad Mergentheim, Germany
• 2 German Center for Diabetes Research (DZD), Neuherberg, Germany
• 3 Department of Clinical Psychology and Psychotherapy, University of Bamberg, Bamberg, Germany

Aims: Measurement tools to evaluate self-management behavior are useful for diabetes research and clinical practice. The Diabetes Self-Management Questionnaire (DSMQ) was introduced in 2013 and has become a widely used tool. This article presents a revised and updated version, DSMQ-R, and evaluates its properties in assessing self-management practices in type 1 diabetes (T1D) and type 2 diabetes (T2D).

Methods: The DSMQ-R is a multidimensional questionnaire with 27 items regarding essential self-management practices for T1D and T2D (including diabetes-adjusted eating, glucose testing/monitoring, medication taking, physical activity and cooperation with the diabetes team). For the revised form, the original items were partially amended and the wording was updated; eleven items were newly added. The tool was applied as part of health-related surveys in five clinical studies (two cross-sectional, three prospective) including a total of 1,447 people with T1D and T2D. Using this data base, clinimetric properties were rigorously tested.

Results: The analyses showed high internal and retest reliability coefficients for the total scale and moderate to high coefficients for the subscales. Reliability coefficients for scales including the new items were consistently higher. Correlations with convergent criteria and related variables supported validity. Responsiveness was supported by significant short to medium term changes in prospective studies. Significant associations with glycemic outcomes were observed for DSMQ-R-assessed medication taking, glucose monitoring and eating behaviors.

Conclusions: The results support good clinimetric properties of the DSMQ-R. The tool can be useful for research and clinical practice and may facilitate the identification of improvable self-management practices in individuals.

Introduction

Diabetes mellitus is a chronic metabolic disease characterized by elevated blood glucose levels due to absolute [type 1 diabetes (T1D)] or relative [type 2 diabetes (T2D)] insulin deficiency ( 1 ). The International Diabetes Federation estimates that 537 million adult people (20–79 years) are currently living with diabetes worldwide; the number is expected to rise to 643 million by 2030 ( 2 ). Diabetes care aims to help people with diabetes achieve near-normal glycemic levels in order to reduce the risk of long-term (e.g., vascular) complications of diabetes while avoiding acute metabolic risks and preserving best possible quality of life ( 3 ).

The key factor to achieving good glycemic levels is the person with diabetes’s self-management of their condition. People with diabetes may need to control carbohydrate intake via their selection of foods, adapt eating behaviors with regard to glycemic load, fats and healthy nutrition, manage blood glucose using glucose-lowering medications, monitor glucose levels using blood tests or sensors, engage in sufficient physical exercise (to optimize glycemia, manage weight or maintain good health) and arrange their activities around current glycemic levels and treatment requirements, as recommended by current guidelines ( 4 – 6 ). Where rapid acting insulin is used (to cover glucose rises after meals), estimating carbohydrate loads of the meals, dose-adjusting insulin doses and correcting elevated glucose levels are additional required practices of daily diabetes self-management.

Persistent or recurrent hyperglycemia increases the risk for developing serious long-term complications of diabetes such as diabetic retinopathy, neuropathy, nephropathy and foot syndrome; further, suboptimal glycemic management is associated with increased risks of acute metabolic complications such as severe hypoglycemia or severe hyperglycemia with the risk of ketoacidosis or hyperosmolar coma ( 7 – 9 ). Therefore, the adoption and maintenance of functional self-management behaviors to achieve good glycemic outcome is decisive for maintaining good health and preventing complications and morbidity ( 10 ). However, evidence supports that people with diabetes’ self-management practices and overall performance are often improvable ( 11 , 12 ); this may be particularly true for people with comorbid mental conditions such as depression and diabetes-specific distress ( 13 – 15 ).

Since self-management is the decisive determinant of the course of diabetes, reflecting/monitoring relevant behaviors in individuals to identify areas of potential improvement and offer suitable education and support may be useful for routine clinical practice. The assessment and evaluation of diabetes self-management behaviors may be of particular interest in people with persistent suboptimum diabetes outcomes where possible problems and barriers are to be detected. Furthermore, measuring self-management may be required as part of research where facilitators and barriers to optimal diabetes care, including mental factors, shall be analyzed [e.g. ( 15 , 16 )] or effects of interventions (e.g., diabetes self-management education) are to be evaluated. Thus, suitable measurement tools are required.

Several systematic reviews of available measurement tools for diabetes self-management confirm that many different tools have been developed; however, most instruments have been applied in limited numbers of studies and the testing of measurement properties was often limited, with few scales meeting rigorous appraisal criteria, according to the reviewers’ conclusions ( 17 – 20 ). These problems may limit the available tools’ usability for research and practice.

In 2013, the Diabetes Self-Management Questionnaire [DSMQ ( 21 )] was introduced to provide a multidimensional measure of diabetes self-management behaviors relevant for the control of glycemia in both major types of diabetes and to overcome limitations of contemporary questionnaires [e.g. ( 22 )]. In direct comparisons, the DSMQ explained significantly more glycemic variation than an established standard self-care scale ( 21 , 23 ). Since then, it has been translated into diverse languages and used in many studies, supporting its potential value for research and practice. A recent systematic review listed the DSMQ as one of only three scales on diabetes self-management which met the COSMIN (COnsensus-based Standards for the selection of health Measurement Instruments) guidelines for measurement tools that can be recommended for use and results obtained with can be trusted ( 20 ).

However, technological innovations such as continuous glucose monitoring and automatic insulin delivery have changed terms and expressions in diabetes care. Furthermore, a shift in diabetes-related language has taken place ( 24 ). Also, some specific self-management aspects should be better covered by the tool. For these reasons, a revision of the DSMQ was needed. The present article presents a revised and updated version of the tool and rigorous testing of its clinimetric properties and functions. Experiences with the tool’s use within five clinical studies provides a broad evidence base to inform about its characteristics and potentials.

Materials and Methods

Diabetes self-management questionnaire (dsmq).

The DSMQ is a multidimensional questionnaire consisting of self-descriptive statements from the person’s point of view ( Table 1 ). Respondents are asked to reflect their self-management behaviors over the past weeks and rate to which extent each statement applies to them. An eight-week reference period was chosen to cover behaviors explaining present HbA 1c ; however, a shorter period (e.g., four weeks) might support the reflection of short-term changes, thus adaption of the instruction, where needed, might be considered. Responses are given on a four-point scale (from 0–’does not apply to me’ to 3–’applies to me very much’). Item scores are summed to scale scores reflecting the following specific activities: adjusting one’s diet towards diabetes (subscale ‘eating behavior’), taking medications consistently (subscale ‘medication taking’), testing/monitoring blood glucose or interstitial glucose (subscale ‘glucose monitoring’), being physically active to improve diabetes and health (subscale ‘physical activity’) and interacting with one’s diabetes-treating physician/healthcare professionals (subscale ‘cooperation with diabetes team’). A total score as a global measure of diabetes self-management can be calculated. Raw sum scores are transformed to a range from 0–10 for better interpretability and comparability (by dividing the raw sum score by the maximum possible sum of the scale [i.e., item number * 3] and multiplying with 10; details on scoring in Supplementary Table 1 ). The tool contains positively and negatively keyed items for greater validity and reliability (e.g., avoidance of one-sided, biased responses); negatively keyed items are reverse-scored before summing, thus higher scale scores reflect more optimal behavior. Since its introduction in 2013, the tool has been widely adopted and used for research and practice across countries and languages ( Supplementary Table 2 ).

Table 1 Items of the original and revised versions of the DSMQ compared.

Original Version

The original version of the DSMQ consists of 16 items ( Table 1 ) which were developed and selected in a systematic, iterative process: A set of newly developed and qualitatively piloted items were initially tested on a sample of 110 people and successively excluded until only those with good properties remained ( 21 ). The resulting questionnaire was then administered to 261 people with T1D or T2D to evaluate measurement properties against a convergent standard measure; results supported reliability and validity ( 21 ). A subsequent study yielded further supportive evidence ( 23 ).

Reasons for the revision were: i) wording considered as improvable in single items, ii) findings suggesting limited reliability for the ‘cooperation with diabetes team’ subscale in some studies and iii) practices of dose-adjusting insulin injections and correcting glucose levels (where intensive insulin treatment applies) being insufficiently covered. The original scale was amended accordingly, that is: i) items were updated to conform with new technologies such as continuous glucose monitoring (CGM) and data management software; the potentially misleading term ‘blood sugar levels’ was replaced with ‘glucose levels’, referring to both blood and interstitial glucose; some items were revised to avoid compliance-oriented expressions (e.g., ‘strictly follow’ or ‘as prescribed’); ii) the ‘cooperation with diabetes team’ items were harmonized and one additional item was added to improve reliability; iii) seven items covering practices of intensive insulin treatment were added as an optional extra. Item-level amendments are given in detail in Supplementary Table 3 ; old and new items are compared in Table 1 . In summary, two items remained unchanged, seven items were slightly revised and seven items were significantly altered but the essential meaning was kept ( Table 1 ). The original item order was kept, except for item 16 which was repositioned as number 20. A total of eleven items were newly added, thereof four regarding general behaviors (no. 16–19) and seven (no. 21–27) regarding intensive insulin treatment practices specifically (e.g., adjusting insulin; correcting glucose levels), the latter given in a separate section with specific instruction.

The DSMQ-R thus contains a total of 27 items, 20 on general behaviors relevant for most people with diabetes and seven on specific insulin treatment behaviors. A total score is estimated using the 20 general items; where applicable, a 27-item total score including the optional items can be calculated. The subscale ‘eating behavior’ contains now six items and the subscale ‘cooperation with diabetes team’ four; ‘medication taking’, ‘glucose monitoring’ and ‘physical activity’ remain unchanged with two, three and three items, respectively; two of the 20 general items request global statements and are included in the total scale only ( Table 1 ).

Study Design and Data Collection

This evaluation of the DSMQ-R includes T1D and T2D. The analyzed data were acquired as part of five clinical studies, three cross-sectional, two prospective, conducted between 2015 and 2021. All studies were ethically approved and carried out in accordance with the Declaration of Helsinki. All participants provided written informed consent.

● Study 1 was a multi-center, cross-sectional survey to evaluate person-reported outcome measures for diabetes, conducted in 2015–16; details of the study are reported elsewhere ( 25 ). Ethical approval was obtained from the Ethics Committee of the German Psychological Society (file no. NH 032015). N =606 participants were surveyed using questionnaires including the DSMQ-R, DAS, PAID-5, PHQ-9, DTSQ (explained below); 606 people participated; data of n =588 (56.6% T1D) could be used for this evaluation.

● Study 2 (‘Depression and Diabetes Control Trial’) was a randomized controlled trial testing a diabetes-specific treatment program for people with depressive symptoms (CES-D ≥16) and hyperglycemia (HbA 1c >7.5%/59 mmol/mol) against diabetes care as usual. It was approved by the Ethics Committee of the State Medical Chamber of Baden-Wuerttemberg (file no. F-2015-056) and is registered at ClinicalTrials.gov (ID no. NCT02675257). Participants were enrolled from 1/2016–3/2017. The first follow-up (FU) assessment after six months was used for retest analysis in this evaluation. Questionnaire assessments included the DSMQ-R, SDSCA, PAID, DDS, DAS and PHQ-9 (below). HbA 1c was assessed in a central laboratory. N =213 were enrolled and 198 (66.2% T1D) provided suitable data for this evaluation.

● Study 3: a cross-sectional FU survey of the ‘DIAMOS’ and ‘ECCE HOMO’ trial participants was conducted in 2017–18, on average five years after participating in the original trials. Participants had been enrolled using equivalent inclusion and exclusion criteria, enabling aggregation to one cohort; all had elevated depressive symptoms (CES-D ≥16) at baseline [study details accessible elsewhere ( 26 – 28 )]. The FU was approved by the Ethics Committee of the State Medical Chamber of Baden-Wuerttemberg (file no. F-2017-071). A total of 323 people (68.1% of the total cohort) could be followed up using questionnaires including the DSMQ-R, PAID, DDS, DAS and PHQ-9; HbA 1c was estimated. N =298 people (64.0% T1D) provided sufficient data for this evaluation.

● Study 4 (‘DIA-LINK1’) was a prospective observational study analyzing links between mental health and glycemic outcomes in T1D (ClinicalTrials.gov ID no. NCT03811132); participants were enrolled from 3/2019–3/2020 and followed over three months. Measurements comprised repeated surveys (including DSMQ-R, PAID, T1-DDS, CES-D), HbA 1c estimation, 17-day ecological momentary assessment (EMA) with daily diabetes-related questions and 4-week continuous glucose monitoring (CGM) ( 29 ). The study was approved by the Ethics Committee of the German Psychological Society (file no. NH 082018). N =203 participants were enrolled.

● Study 5: the ‘DIA-LINK2’ study (2020–21) is a prospective observational study regarding mental health and glycemia with the same design as DIA-LINK1 but regarding T2D (ClinicalTrials.gov ID no. NCT04438018). Ethical approval was obtained from the Ethics Committee of the German Psychological Society (file no. HermannsNorbert2020-03-05AM). A total of 190 people with T2D have been enrolled, and n =180 provided suitable data for this evaluation.

Variables and Measurements

Besides the DSMQ-R, the following variables were assessed as part of the studies:

Glycemic outcome: Glycated hemoglobin (HbA 1c ) was estimated from venous blood samples taken at the same time as the questionnaire assessments in all studies. HbA 1c was usually estimated in a central laboratory (at the Diabetes Center Mergentheim) using high performance liquid chromatography (performed with the Bio-Rad Variant II Turbo analyzer in studies 2 and 3 and the Tosoh Automated Glycohemoglobin Analyzer HLC-723G11 in studies 4 and 5), meeting IFCC standard [laboratory normal range 4.3–6.1% (24–43 mmol/mol)]; study 1 included four different laboratory cites.

Study 4 additionally assessed glycemic levels over four weeks using intermittently scanned CGM. The following CGM-derived parameters were calculated: mean sensor glucose (in mg/dl), time in range (% values between 70–180 mg/dl, 3.9–10 mmol/l), time below range (% values <70 mg/dl, <3.9 mmol/l), time above range (% values >180 mg/dl, >10 mmol/l), and glucose variability [coefficient of variation (CV)].

Diabetes self-care activities: The 10-item Summary of Diabetes Self-Care Activities Measure [SDSCA ( 22 , 30 )] was applied as a convergent measure of diabetes self-management in study 2. The tool requests on how many days of the past week the person engaged in healthy eating, exercising, blood sugar testing and foot care. Responses are averaged to scales (e.g., Diet, Exercise, Blood Sugar Testing) with scores ranging from 0–7 and higher values reflecting more frequent activity.

Diabetes distress and diabetes-specific problems: The 20-item Problem Areas in Diabetes Scale (PAID) measuring diabetes-related distress ( 31 ) was applied in all studies. The questionnaire requests ratings of diabetes-specific emotional problems on a five-point scale (0–’not a problem’ to 4–’serious problem’). The item scores are summed and transformed to a total score ranging from 0–100; higher scores reflect higher distress; scores ≥40 suggest meaningful distress ( 32 ). In study 1, the 5-item short form [PAID-5 ( 33 )] was used.

In studies 2–5, the Diabetes Distress Scale [DDS ( 34 )] or T1-Diabetes Distress Scale [T1-DDS ( 35 )] was administered in addition to the PAID. The DDS/T1-DDS items address a range of diabetes-specific problems; however, it also includes items and scales whose relations to the construct of diabetes distress have been questioned ( 14 , 32 , 36 ). Therefore, we did not estimate a total score but rather selected specific items whose contents regarding self-management-related problems could be used for the correlation analysis (i.e., DDS items 6, 8 and 12 on ‘not testing blood sugars frequently enough’, ‘often failing with diabetes routine’ and ‘not sticking closely enough to a good meal plan’, and T1-DDS items 2, 8, 12, 23 and 28 on ‘not eating as carefully as one should’, ‘not taking as much insulin as one should’, ‘not checking blood glucose as often as one should’, ‘eating being out of control’ and ‘not giving diabetes as much attention as one should’); these aspects were assessed as convergent criteria for corresponding DSMQ-R scales. Items regarding doctor-related problems (i.e., DDS item 15 on ‘not having a doctor who one can see regularly about diabetes’ and T1-DDS items 7 and 18, ‘can’t tell diabetes doctor what is really on my mind’, ‘diabetes doctor doesn’t really understand what it’s like to have diabetes’) were used for correlation with the DSMQ-R scale ‘cooperation with diabetes team’. Responses in the DDS/T1-DDS are given on a six-point scale (1–’not a problem’ to 6–’a very serious problem’), thus higher scores reflect greater problems.

Diabetes acceptance , a measure of psychological adjustment to living with diabetes, was assessed using the Diabetes Acceptance Scale (DAS); in studies 1–3, the full 20-item version was used, in studies 4–5, the 10-item short form ( 25 ). The items request aspects of acceptance and integration (e.g., ‘I accept diabetes as part of my life’) versus avoidance, neglect and demotivation (e.g., ‘I avoid dealing with topics related to diabetes’). Responses are given on a four-point scale (0–’never true for me’ to 3–’always true for me’). Item scores are summed so that higher scores reflect higher acceptance (range 0–60). Higher acceptance scores have been associated with more optimal self-management ( 25 , 37 ). Besides the total score, items specifically related to treatment motivation (e.g., ‘I have difficulties to motivate myself to perform good diabetes self-care’) and treatment neglect (e.g., ‘I neglect diabetes self-care because I want to avoid topics related to diabetes’) were aggregated to subscales (Cronbach’s α =0.71 and 0.83, respectively).

Diabetes treatment satisfaction was measured using the Diabetes Treatment Satisfaction Questionnaire (DTSQ) in study 1, including six satisfaction-related items and a 7-point scale (0–’very dissatisfied’ to 6–’very satisfied’). Items are summed to a total score from 0–64; higher scores reflect higher satisfaction ( 38 ). Higher treatment satisfaction was expected to be associated with more optimal treatment behavior (DSMQ-R).

Depressive symptoms were assessed in all studies due to their high prevalence in diabetes as well as the studies focusing on depression and mental health. Studies included either the Patient Health Questionnaire-9 (PHQ-9) or the Center for Epidemiologic Studies Depression Scale (CES-D); both have excellent properties ( 39 ). The PHQ-9 assesses the nine symptoms of major depression according to DSM-5 during the past two weeks. Responses are given on a four-point scale (0–’not at all’ to 3–’nearly every day’). Total score range is 0–27; higher scores indicate more symptoms. The CES-D assesses 20 depressive symptoms during the past week; responses are given on a four-point scale (0–’rarely or none of the time’ to 3–’most or all of the time’), resulting in a total score from 0–60 (higher scores=more symptoms). Depressive symptoms have been consistently associated with less optimal self-management across behaviors [e.g. ( 13 )].

Daily diabetes problems/burdens: The DIA-LINK studies included a smartphone-based EMA with daily diabetes-related questions over 17 days ( 29 ). Items constituting likely correlates of the DSMQ-R were used as convergent criteria (e.g., ‘How much have you felt guilty when neglecting your diabetes treatment today?’; full item details in Supplementary Table 4 ). Responses were given on a scale from 0–’not at all’ to 10–’very much’. Daily responses were averaged.

Demographic and person-related variables comprised sex, age, BMI, diabetes type, diabetes duration and treatment regimen. Long-term and acute complications of diabetes (study 1) were based on medical examinations, laboratory assessments and interviews (assessed were diabetic retinopathy, neuropathy, nephropathy, foot syndrome; treated ketoacidosis, past 12 months). Mean numbers of daily insulin injections (where applicable) and daily glucose tests or scans/readings as well as frequencies of diabetologist visits per past six months were assessed in face-to-face interviews.

Statistical Analyses

Statistical analyses were performed using SPSS 26.0.0 (IBM SPSS Statistics). P values < 0.05 (two-tailed) were considered to indicate statistical significance. For the DSMQ-R, total and subscale scores were calculated as per scoring instruction ( Supplementary Table 1 ) with scores ranging between 0 and 10. Negatively-keyed items were reverse-scored so that higher scale scores suggest more optimal behavior. Where applicable, a 27-item total score was calculated in addition to the 20-item total; yet the optional items were not included in subscale scores to warrant comparisons of scores between subgroups. Measurement functions were analyzed according to clinimetric criteria ( 40 ). Internal reliability was analyzed using Cronbach’s α ; since potential preference of McDonald’s ω over α has been discussed ( 41 ), ω was additionally estimated [using Hayes’ OMEGA macro for SPSS ( 41 )]. Reproducibility was tested using retest correlations in the prospective studies. Construct validity was evaluated via correlations with convergent measures and related variables to develop a nomological network. Since adjusting eating behaviors towards diabetes, taking medications consistently and checking glucose levels regularly can be expected to result in better glycemic levels, associations between the corresponding DSMQ-R scales and glycemic outcomes were analyzed as indicators of validity. Similarly, associations with acute and long-term complications were assessed in study 1. Further, associations between the DSMQ-R scales and convergent measures of self-care activities, treatment satisfaction, treatment motivation and neglect as well as diabetes acceptance, diabetes distress and depressive symptoms were analyzed. Structural validity was assessed using confirmatory factor analyses (AMOS 26.0.0, IBM SPSS Statistics). Model fit was evaluated according to Comparative Fit Index (CFI) ≥ 0.95, Tucker Lewis Index (TLI) ≥ 0.95, Standardized Root Mean Square Residual (SRMR) ≤ 0.08 and Root Mean Square Error of Approximation (RMSEA) ≤ 0.06. Responsiveness , the ability to detect change, was assessed via changes of the DSMQ-R scales in prospective studies, given as Cohen’s d . Where applicable, changes were compared between treatment groups (i.e., study 2, with participants randomized to either depression treatment or diabetes care as usual).

Sample Characteristics

The sample characteristics are given in Table 2 . Studies 1–3 had mixed samples including people with T1D and T2D (T1D being overrepresented in line with secondary and tertiary care enrolment), study 4 and 5 assessed only T1D or T2D, respectively. Sample sizes varied between 180 and 588. Study 1 contained a more general sample, whereas other studies overrepresented people with specific mental aspects: study 2 contained people with current depressive symptoms, study 3 contained people with a history of depressive symptoms and studies 4 and 5 included majorities with either depressive symptoms or diabetes distress. All samples had a wide age range with a mean age between 45 and 53 years, except for study 4 (T1D only) whose sample’s mean age was 39 years. The mean diabetes duration reflected relatively long-standing diabetes throughout. HbA 1c levels were generally elevated with mean values around 7.8 to 9.3% (62 to 78 mmol/mol) across the studies.

Table 2 Study sample characteristics and normative data and reliability indices for the DSMQ-R scales by study and diabetes type.

Internal Reliability

Cronbach’s α of the 20-item total scale varied from 0.88–0.92 (mean=0.90) in T1D and from 0.84–0.89 (mean=0.87) in T2D across studies. Coefficients were slightly higher for the 27-item total scale, where applicable ( Table 2 ). For the subscales, mean coefficients α for T1D (T2D) were: ‘eating behavior’=0.76 (0.78), ‘medication taking’=0.79 (0.75), ‘glucose monitoring’=0.76 (0.82), ‘physical activity’=0.87 (0.80) and ‘cooperation with diabetes team’=0.82 (0.67). McDonald’s ω yielded consistent results ( Table 2 ). Direct comparisons of scale reliabilities estimated including the newly added items versus original ones only yielded consistently higher reliability coefficients for the new scales ( Supplementary Table 5 ).

Reproducibility

Retest correlations over three to six months reflected sufficient intra-individual stability of the measurement over time. Mean correlations for 20-item total scale were 0.64 in T1D and 0.53 in T2D; mean correlations for the subscales were from 0.53–0.69 in T1D and from 0.43–0.61 in T2D ( Table 2 ).

Construct Validity

Correlations with convergent criteria were generally in line with expectations towards validity of the measurement and a meaningful nomological network.

Total scale: Higher DSMQ-R total scores (suggesting more optimal self-management) were consistently associated with better HbA 1c values across studies and diabetes types; however, the sizes of associations varied (e.g., from -0.29 to -0.57, mean = -0.41, in T1D and from -0.20 to -0.36, mean=-0.30, in T2D; 20-item total). Higher DSMQ-R total scores were also associated with lower mean sensor glucose, more time in range, less time above range and lower glucose variability in T1D (study 4). Further, higher DSMQ-R total scores were associated with lower rates of long-term complications and less events of ketoacidosis (T1D). DSMQ-R total scores were highly positively associated with convergent measures of treatment motivation, treatment satisfaction and self-management performance according to the SDSCA questionnaire and corresponding DDS/T1-DDS items ( Table 3 ); and highly negatively with items reflecting suboptimal treatment behavior. In studies 4 and 5, significant correlations with EMA items reflecting self-management were observed. Finally, higher DSMQ total scores were seen in people with better mental health, lower diabetes distress and less depressive symptoms.

Table 3 Validity and responsiveness indices for the DSMQ-R scales by study and diabetes type.

Subscales: The subscales ‘eating behavior’, ‘medication taking’ and ‘glucose monitoring’ showed significant associations with corresponding convergent criteria for diabetes-adjusted eating (e.g., SDSCA scale on healthy eating, DDS/T1-DDS items regarding sticking to a good meal plan and eating carefully), medication taking (e.g., T1-DDS item on insulin taking, mean number of daily insulin injections in T1D), glucose monitoring (e.g., SDSCA scale on blood sugar testing, DDS/T1-DDS items on glucose checking, mean number of daily glucose checks/scans). Each of the scales showed significant associations with better HbA 1c in several studies, however not all. The subscale ‘physical activity’ showed high correlations with the convergent SDSCA scale on past-week physical exercise and small-to-moderate associations with BMI. The subscale ‘cooperation with diabetes team’ showed significant correlations with self-reported frequencies of diabetologist visits as well as corresponding DDS/T1-DDS items on doctor-related problems. ‘Eating behavior’, ‘medication taking’ and ‘physical activity’ were also significantly associated with corresponding EMA ratings in studies 4 and 5.

Structural Validity

Confirmatory factor analyses supported a five-factor structure representing the five subscales with excellent fit to the data for both T1D and T2D ( Supplementary Figures 1–2 ). One-factor models representing the total scale showed good fit as well; however, with slightly lower fit indices and lower factor loadings ( Supplementary Figures 3–6 ).

Responsiveness

The ability to detect change was supported by significant changes over time in the total score and most subscale scores in the prospective studies. Greater changes were seen in the total scale and ‘eating behavior’ and ‘glucose monitoring’ subscales, while changes in ‘medication taking’ were modest and changes in ‘physical activity’ and ‘cooperation with diabetes team’ were small or lacking ( Table 3 ). Between-group comparisons for people receiving depression treatment versus diabetes care as usual in study 2 suggested similar changes in DSMQ-R scores without significant differences between the groups at six-month follow-up.

Main Findings

The evaluation of the DSMQ-R using data from diverse studies suggests very good properties in measuring diabetes self-management behavior in both T1D and T2D according to clinimetric criteria ( 40 ). Results suggest that the tool has good reliability, validity and responsiveness to change. The terms and expressions used in the questionnaire were updated to conform with modern diabetes-related language. The revised scales with newly added items showed higher internal reliability than the original version’s item sets.

The DSMQ-R total scale constitutes a reliable and valid measure of overall self-management. Yet it is a global measure; thus assessing the specific behaviors using the subscales may be preferred and even necessary for understanding individual aspects. For the subscales, however, differential properties and options should be considered: First, the numbers of items per scale differ which may affect reliability of the measurement. In this evaluation, most subscales yielded satisfactory to good reliability estimates; however, lower reliability coefficients were seen for subscales with fewer items (e.g., medication taking) in some of the studies. Furthermore, coefficients varied across studies and patient groups, suggesting that the utilization of subscales in research might benefit from affirming reliability within a given study data set. Notably, despite specific revisions and improved internal reliability, the ‘cooperation with diabetes team’ subscale still showed subthreshold reliability coefficients in two of five studies for T2D; yet not for T1D.

Reliability coefficients were mostly slightly higher in T1D subsamples compared to T2D which is in line with previous findings ( 21 ). This might be explained by more diverse treatment regimens and practices in T2D; for instance, prescribed medications may be diverse (oral drugs, insulin and/or incretin mimetics), glucose testing may or may not be required and dietary recommendations may vary in relevance and function. This might also explain higher associations between the DSMQ-R scales and HbA 1c in T1D [consistent with previous findings ( 23 )], where glycemic outcomes directly depend on the consistent coordination and adjustment of carbohydrate intake, activities and insulin doses; whereas in T2D, glycemic control may rely more on diet and activity and less on glucose checking and meal-specific decisions (depending on the treatment regimen); also, residual insulin action may stabilize glycemic levels and reduce hyperglycemia.

It should be noted that two-sided questioning (using both positively and negatively keyed items) may lower internal consistency as observed in some DSMQ-R subscales; at the same time, higher validity is achieved and response bias is prevented. From a clinimetric perspective, a varied assessment using items covering different aspects from different sides is more important than a highly homogeneous measurement ( 40 ).

Validity of the scale measurement was supported by high correlations with convergent scores and items from other questionnaires. However, as self-report is prone to bias, associations with objective measures constitute another important source of information. Thus, the widely consistent associations between DSMQ-R scales and HbA 1c (as well as CGM-derived glucose parameters) across studies may be seen as extra evidence favoring validity.

Relatively good explanation of variation in HbA 1c was already observed in our previous studies for both T1D and T2D ( 21 , 23 ). This might be explained by i) the reflection of behaviors over a broader, more representative reference period and ii) the items requesting behavioral evaluations (e.g., ‘with care and attention’) rather than behavior frequency (e.g., ‘on how many days…?’ as in the SDSCA). On the other hand, three studies using the DSMQ with non-Western samples ( 42 – 44 ) and one Hungarian study ( 45 ) have reported lower associations with HbA 1c , suggesting caution against generalization across cultures.

Validity of the measurement was also supported by the structural representation of assessed contents (i.e., items and scales) in the factor analyses with good model fit for both T1D and T2D.

Change scores reflecting improvements in DSMQ-R-assessed behaviors supported good responsiveness of the measurement. In study 4, similar changes were seen for people randomized to depression treatment versus diabetes care as usual; this could be explained by both groups receiving treatment with beneficial effects on self-management behavior. The tool’s ability to detect change is also supported by findings from international studies using the DSMQ which found significant self-management improvement over time and between-group differences in randomized trials ( 46 – 49 ); notably, observed changes in DSMQ scores by group were often accompanied by parallel changes in HbA 1c , which might be taken as evidence supporting validity of the changes ( 46 , 48 , 50 ). With regard to responsiveness and the tool’s reference period (eight weeks), a shorter period might facilitate the detection of short-term changes, thus adapting the instruction (e.g., four weeks), where needed, may be considered.

In terms of item amendments (e.g., revised wording), the DSMQ-R probably constitutes a relevant improvement. However, since most revisions were minor and item concepts were kept equivalent, the original 16-item version is basically included in the revised form. Estimation of scales as for the original version, where needed (e.g., to compare scores with former study results), would still be possible.

Limitations and Strengths

The inferences drawn from this research are qualified by the following limitations: first, four of the studies whose data were analyzed here focused on diabetes-comorbid mental conditions, thus rates of depressive symptoms and/or diabetes distress were elevated and the samples may not be representative for the general population with diabetes (i.e., risk of spectrum bias). Second, we assessed cross-sectional associations between self-reported behavior and diabetes outcomes, thus inferences towards causation are not possible; in fact, associations with glycemic outcomes might be bidirectional; for instance, knowing of glycemic levels (e.g., last HbA 1c ) might influence self-management self-appraisal in the questionnaire. Third, the study samples were recruited within secondary or tertiary care, thus samples may not represent the primary care population; based on this, people with T2D assessed here used advanced medical treatments often including insulin and even basal-bolus therapy with multiple daily injections, whereas people with diet-and-exercise regimens and/or oral antidiabetic treatment alone were less represented.

The strengths of the evaluation may be seen in the standardized assessment using validated scales and items, temporal coincidence of questionnaire self-reports, interviews and laboratory measures and the inclusion of multiple methods including CGM and EMA for the assessment of convergent criteria. Furthermore, the stratified analyses for T1D and T2D using sufficiently large samples support evaluation for both major types of diabetes. Due to potential advantages of McDonald’s ω over Cronbach’s α ( 41 ), we calculated both estimates, yielding highly consistent results. Finally, the evaluation across different study samples, both general and specific, yields a more comprehensive and representative total evidence base; the fact that indices of reliability and validity, including associations with clinical criteria, were relatively consistent across studies may favor generalizability.

Conclusions

In summary, the results support good clinimetric properties of the DSMQ-R. The tool can be used for research and clinical practice. It may help understand barriers and facilitators of functional self-management in T1D and T2D, facilitate the identification improvable practices in individuals and monitor behavior change following treatment in practical care or research trials.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics Statement

The studies involving human participants were reviewed and approved by Ethics Committee of the German Psychological Society or the Ethics Committee of the State Medical Chamber of Baden-Wuerttemberg. The patients/participants provided their written informed consent to participate in this study.

Author Contributions

AS: developed and revised the DSMQ; planned and designed study 1; co-planned and designed studies 2–5; collected the data; performed the evaluation, analyzed and interpreted the data; wrote the manuscript. BK and NH: planned and designed studies 2–5; discussed the findings and revised the manuscript. DE: planned and designed studies 4–5; co-planned and designed studies 2–3; discussed the findings and revised the manuscript. TH: discussed the findings and revised the manuscript. All authors contributed to the article and approved the submitted version.

Study 1 was supported by the German Diabetes Foundation (DDS) (grant number 375.10.15). Studies 2–3 were supported by the German Center for Diabetes Research (DZD) (grant number 82DZD01102). Studies 4–5 were supported by the German Center for Diabetes Research (DZD) [grant number 82DZD11A02]. The funders were not involved in decisions regarding study design; collection, analysis and interpretation of data; writing of the report; and submission of the article for publication.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Acknowledgments

We thank the persons who kindly participated in the studies enabling this research. We acknowledge the valuable contributions to participant enrolment and data collection of André Reimer (studies 1–3), Paula Rubertus (study 4) and Fabienne Schmid (study 5).

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcdhc.2021.823046/full#supplementary-material

Abbreviations

CES-D, Center for Epidemiologic Studies Depression Scale; CG, control group; CV, coefficient of variation; DAS, Diabetes Acceptance Scale; DDS, Diabetes Distress Scale; DM, diabetes mellitus; DSMQ, Diabetes Self-Management Questionnaire; DTSQ, Diabetes Treatment Satisfaction Questionnaire; EG, experimental group; EMA, ecological momentary assessment; HbA 1c , glycated hemoglobin; iscCGM, intermittently scanned continuous glucose monitoring; MDI, multiple daily (insulin) injections; PAID, Problem Areas in Diabetes Scale; PHQ-9, Patient Health Questionnaire-9; PWD, people with diabetes; rtCGM, real-time continuous glucose monitoring; SDSCA, Summary of Diabetes Self-Care Activities; SMBG, self-monitoring of blood glucose; T1-DDS, Type 1 Diabetes Distress Scale; T1D, type 1 diabetes; T2D, type 2 diabetes.

1. World Health Organization (WHO). Diabetes . Available at: https://www.who.int/health-topics/diabetes#tab=tab_1 (Accessed on 10.11.2021).

Google Scholar

2. International Diabetes Federation. IDF Diabetes Atlas . 10th edn. Brussels, Belgium (2021). Available at: https://www.diabetesatlas.org .

3. American Diabetes Association. Facilitating Behavior Change and Well-Being to Improve Health Outcomes: Standards of Medical Care in Diabetes-2021. Diabetes Care (2021) 44:S53–72. doi: 10.2337/dc21-S005

PubMed Abstract | CrossRef Full Text | Google Scholar

4. American Association of Diabetes Educators (AADE). AADE7 Self-Care Behaviors. Diabetes Educ (2008) 34:445–9. doi: 10.1177/0145721708316625

5. American Diabetes Association. Standards of Medical Care in Diabetes—2020 Abridged for Primary Care Providers. Clin. Diabetes (2020) 38:10–38. doi: 10.2337/cd20-as01

6. Davies MJ, D’Alessio DA, Fradkin J, Kernan WN, Mathieu C, Mingrone G, et al. Management of Hyperglycaemia in Type 2 Diabetes, 2018. A Consensus Report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetologia (2018) 61:2461–98. doi: 10.1007/s00125-018-4729-5

7. Akalin S, Berntorp K, Ceriello A, Das AK, Kilpatrick ES, Koblik T, et al. Intensive Glucose Therapy and Clinical Implications of Recent Data: A Consensus Statement From the Global Task Force on Glycaemic Control. Int J Clin Pract (2009) 63:1421–25. doi: 10.1111/j.1742-1241.2009.02165.x

8. Yu PC, Bosnyak Z, Ceriello A. The Importance of Glycated Haemoglobin (HbA(1c)) and Postprandial Glucose (PPG) Control on Cardiovascular Outcomes in Patients With Type 2 Diabetes. Diabetes Res Clin Pract (2010) 89:1–9. doi: 10.1016/j.diabres.2009.12.009

9. Sina M, Graffy J, Simmons D. Associations Between Barriers to Self-Care and Diabetes Complications Among Patients With Type 2 Diabetes. Diabetes Res Clin Pract (2018) 141:126–31. doi: 10.1016/j.diabres.2018.04.031

10. Houle J, Beaulieu M-D, Chiasson J-L, Lespérance F, Côté J, Strychar I, et al. Glycaemic Control and Self-Management Behaviours in Type 2 Diabetes: Results From a 1-Year Longitudinal Cohort Study. Diabetes Med (2015) 32:1247–54. doi: 10.1111/dme.12686

CrossRef Full Text | Google Scholar

11. Coyle ME, Francis K, Chapman Y. Self-Management Activities in Diabetes Care: A Systematic Review. Aust Health Rev (2013) 37:513–22. doi: 10.1071/AH13060

12. Alexandre K, Vallet F, Peytremann-Bridevaux I, Desrichard O. Identification of Diabetes Self-Management Profiles in Adults: A Cluster Analysis Using Selected Self-Reported Outcomes. PloS One (2021) 16:e0245721. doi: 10.1371/journal.pone.0245721

13. Gonzalez JS, Peyrot M, McCarl LA, Collins EM, Serpa L, Mimiaga MJ, et al. Depression and Diabetes Treatment Nonadherence: A Meta-Analysis. Diabetes Care (2008) 31:2398–403. doi: 10.2337/dc08-1341

14. Gonzalez JS, Shreck E, Psaros C, Safren SA. Distress and Type 2 Diabetes-Treatment Adherence: A Mediating Role for Perceived Control. Health Psychol (2015) 34:505–13. doi: 10.1037/hea0000131

15. Schmitt A, Bendig E, Baumeister H, Hermanns H, Kulzer B. Associations of Depression and Diabetes Distress With Self-Management Behavior and Glycemic Control. Health Psychol (2021) 40:113–24. doi: 10.1037/hea0001037

16. Osborn CY, Egede LE. Validation of an Information-Motivation-Behavioral Skills Model of Diabetes Self-Care (IMB-DSC). Patient Educ Couns (2010) 79:49–54. doi: 10.1016/j.pec.2009.07.016

17. Eigenmann CA, Colagiuri R, Skinner TC, Trevena L. Are Current Psychometric Tools Suitable for Measuring Outcomes of Diabetes Education? Diabetes Med (2009) 26:425–36. doi: 10.1111/j.1464-54912009.02697.x

18. Caro-Bautista J, Martín-Santos FJ, Morales-Asencio JM. Systematic Review of the Psychometric Properties and Theoretical Grounding of Instruments Evaluating Self-Care in People With Type 2 Diabetes Mellitus. J Adv Nurs (2014) 70:1209–27. doi: 10.1111/jan.12298

19. Lu Y, Xu J, Zhao W, Han H-R. Measuring Self-Care in Persons With Type 2 Diabetes: A Systematic Review. Eval Health Prof (2016) 39:131–84. doi: 10.1177/0163278715588927

20. Wee PJL, Kwan YH, Loh DHF, Phang JK, Puar TH, Østbye T, et al. Measurement Properties of Patient-Reported Outcome Measures for Diabetes: Systematic Review. J Med Internet Res (2021) 23:e25002. doi: 10.2196/25002

21. Schmitt A, Gahr A, Hermanns N, Kulzer B, Huber J, Haak T. The Diabetes Self-Management Questionnaire (DSMQ): Development and Evaluation of an Instrument to Assess Diabetes Self-Care Activities Associated With Glycaemic Control. Health Qual Life Outcomes (2013) 11:138. doi: 10.1186/1477-7525-11-138

22. Kamradt M, Bozorgmehr K, Krisam J, Freund T, Kiel M, Qreini M, et al. Assessing Self-Management in Patients With Diabetes Mellitus Type 2 in Germany: Validation of a German Version of the Summary of Diabetes Self-Care Activities Measure (SDSCA-G). Health Qual Life Outcomes (2014) 12:185. doi: 10.1186/s12955-014-0185-1

23. Schmitt A, Reimer A, Hermanns N, Huber J, Ehrmann D, Schall S, et al. Assessing Diabetes Self-Management With the Diabetes Self-Management Questionnaire (DSMQ) Can Help Analyse Behavioural Problems Related to Reduced Glycaemic Control. PloS One (2016) 11:e0150774. doi: 10.1371/journal.pone.0150774

24. Speight J, Conn J, Dunning T, Skinner TC, Diabetes Australia. Diabetes Australia Position Statement. A New Language for Diabetes: Improving Communications With and About People With Diabetes. Diabetes Res Clin Pract (2012) 97:425–31. doi: 10.1016/j.diabres.2012.03.015

25. Schmitt A, Reimer A, Kulzer B, Icks A, Paust R, Roelver KM, et al. Measurement of Psychological Adjustment to Diabetes With the Diabetes Acceptance Scale. J. Diabetes Complications (2018) 32:384–92. doi: 10.1016/j.jdiacomp.2018.01.005

26. Hermanns N, Schmitt A, Gahr A, Herder C, Nowotny B, Roden M, et al. The Effect of a Diabetes-Specific Cognitive Behavioral Treatment Program (DIAMOS) for Patients With Diabetes and Subclinical Depression: Results of a Randomized Controlled Trial. Diabetes Care (2015) 38:551–60. doi: 10.2337/dc14-1416

27. Schmitt A, Kulzer B, Reimer A, Herder C, Roden M, Haak T, et al. Evaluation of a Stepped Care Approach to Manage Depression and Diabetes Distress in Patients With Type 1 Diabetes and Type 2 Diabetes: Results of a Randomized Controlled Trial (ECCE HOMO Study). Psychother Psychosom (2021). doi: 10.1159/000520319

28. Herder C, Schmitt A, Budden F, Reimer A, Kulzer B, Roden M, et al. Association Between Pro- and Anti-Inflammatory Cytokines and Depressive Symptoms in Patients With Diabetes-Potential Differences by Diabetes Type and Depression Scores. Transl Psychiatry (2018) 7:1. doi: 10.1038/s41398-017-0009-2

29. Ehrmann D, Schmitt A, Priesterroth L, Kulzer B, Hermanns N. Time in Diabetes Distress and Glycaemia-Specific Distress: New Patient-Reported Outcome Measures for Psychosocial Burden of Diabetes Using Ecological Momentary Assessment in an Observational Study. Diabetes Care (2022).

30. Toobert DJ, Hampson SE, Glasgow RE. The Summary of Diabetes Self-Care Activities Measure: Results From 7 Studies and a Revised Scale. Diabetes Care (2000) 23:943–50. doi: 10.2337/diacare.23.7.943

31. Polonsky WH, Anderson BJ, Lohrer PA, Welch G, Jacobson AM, Aponte JE, et al. Assessment of Diabetes-Related Distress. Diabetes Care (1995) 18:754–60. doi: 10.2337/diacare.18.6.754

32. Schmitt A, Reimer A, Kulzer B, Haak T, Ehrmann D, Hermanns N. How to Assess Diabetes Distress: Comparison of the Problem Areas in Diabetes Scale (PAID) and the Diabetes Distress Scale (DDS). Diabetes Med (2016) 33:835–43. doi: 10.1111/dme.12887

33. McGuire BE, Morrison TG, Hermanns N, Skovlund S, Eldrup E, Gagliardino J, et al. Short-Form Measures of Diabetes-Related Emotional Distress: The Problem Areas In Diabetes Scale (PAID)-5 and PAID-1. Diabetologia (2010) 53:66–9. doi: 10.1007/s00125-009-1559-5

34. Polonsky WH, Fisher L, Earles J, Dudl RJ, Lees J, Mullan J, et al. Assessing Psychosocial Distress in Diabetes: Development of the Diabetes Distress Scale. Diabetes Care (2005) 28:626–31. doi: 10.2337/diacare.28.3.626

35. Fisher L, Polonsky WH, Hessler DM, Masharani U, Blumer I, Peters AL, et al. Understanding the Sources of Diabetes Distress in Adults With Type 1 Diabetes. J Diabetes Complications (2015) 29:572–7. doi: 10.1016/j.jdiacomp.2015.01

36. Fenwick EK, Rees G, Holmes-Truscott E, Browne JL, Pouwer F, Speight J. What is the Best Measure for Assessing Diabetes Distress? A Comparison of the Problem Areas in Diabetes and Diabetes Distress Scale: Results From Diabetes MILES–Australia. J Health Psychol (2018) 23:667–80. doi: 10.1177/1359105316642006

37. Schmitt A, Reimer A, Kulzer B, Haak T, Gahr A, Hermanns N. Assessment of Diabetes Acceptance can Help Identify Patients With Ineffective Diabetes Self-Care and Poor Diabetes Control. Diabetes Med (2014) 31:1446–51. doi: 10.1111/dme.12553

38. Bradley C. Handbook of Psychology and Diabetes: A Guide to Psychological Measurement in Diabetes Research and Practice. In: Bradley C, editor. Diabetes Treatment Satifaction Questionnaire (DTSQ) . London: Overseas Publishers Association (1994). p. 111–32.

39. van Dijk SEM, Adriaanse MC, van der Zwaan L, Bosmans JE, van Marwijk HWJ, van Tulder MW, et al. Measurement Properties of Depression Questionnaires in Patients With Diabetes: A Systematic Review. Qual Life Res (2018) 27:1415–30. doi: 10.1007/s11136-018-1782-y

40. Carrozzino D, Patierno C, Guidi J, Berrocal Montiel C, Cao J, Charlson ME, et al. Clinimetric Criteria for Patient-Reported Outcome Measures. Psychother Psychosom (2021) 90:222–32. doi: 10.1159/000516599

41. Hayes AF, Coutts JJ. Use Omega Rather Than Cronbach’s Alpha for Estimating Reliability. But… Commun Methods Meas (2020) 14:1–24. doi: 10.1080/19312458.2020.1718629

42. Hammad S, Darawad M, Hourani E, Demeh W. Predictors of Glycated Hemoglobin Among Jordanian Diabetic Patients. Iran J Public Health (2015) 44:1482–91.

PubMed Abstract | Google Scholar

43. Bukhsh A, Khan TM, Nawaz MS, Ahmed HS, Chan KG, Lee L-H, et al. Association of Diabetes-Related Self-Care Activities With Glycemic Control of Patients With Type 2 Diabetes in Pakistan. Patient Prefer Adherence (2018) 12:2377–85. doi: 10.2147/PPA.S177314

44. Totesora D, Ramos-Rivera MI, Villegas-Florencio MQ, Reyes-Sia PN. Association of Diabetes-Related Emotional Distress With Diabetes Self-Care and Glycemic Control Among Adult Filipinos With Type 2 Diabetes Mellitus at a Tertiary Hospital in Manila, Philippines. J ASEAN Fed Endocr Soc (2019) 34:189–96. doi: 10.15605/jafes.034.02.10

45. Vincze A, Losonczi A, Stauder A. The Validity of the Diabetes Self-Management Questionnaire (DSMQ) in Hungarian Patients With Type 2 Diabetes. Health Qual Life Outcomes (2020) 18:344. doi: 10.1186/s12955-020-01595-7

46. Eroglu N, Sabuncu N. The Effect of Education Given to Type 2 Diabetic Individuals on Diabetes Self-Management and Self-Efficacy: Randomized Controlled Trial. Prim Care Diabetes (2021) 15:451–58. doi: 10.1016/j.pcd.2021.02.011

47. Fearon-Lynch JA, Sethares KA, Asselin ME, Batty K, Stover CM. Effects of Guided Reflection on Diabetes Self-Care: A Randomized Controlled Trial. Diabetes Educ (2019) 45:66–79. doi: 10.1177/0145721718816632

48. Sayin Kasar K, Duru Asiret G, Kutmec Yilmaz C, Canlar Ş. The Effect of Model-Based Telephone Counseling on HbA1c and Self-Management for Individuals With Type 2 Diabetes: A Randomized Controlled Trial. Prim Care Diabetes (2021) 10. doi: 10.1016/j.pcd.2021.09.005

49. Schnell O, Klausmann G, Gutschek B, Garcia-Verdugo RM, Hummel M. Impact on Diabetes Self-Management and Glycemic Control of a New Color-Based SMBG Meter. J Diabetes Sci Technol (2017) 11:1218–25. doi: 10.1177/1932296817706376

50. Ebert DD, Nobis S, Lehr D, Baumeister H, Riper H, Auerbach RP, et al. The 6-Month Effectiveness of Internet-Based Guided Self-Help for Depression in Adults With Type 1 and 2 Diabetes Mellitus. Diabetes Med (2017) 34:99–107. doi: 10.1111/dme.13173

Keywords: diabetes, treatment behavior, self-managament, health behavior, clinimetric, measurement instrument, questionnaire, evaluation

Citation: Schmitt A, Kulzer B, Ehrmann D, Haak T and Hermanns N (2022) A Self-Report Measure of Diabetes Self-Management for Type 1 and Type 2 Diabetes: The Diabetes Self-Management Questionnaire-Revised (DSMQ-R) – Clinimetric Evidence From Five Studies. Front. Clin. Diabetes Healthc. 2:823046. doi: 10.3389/fcdhc.2021.823046

Received: 26 November 2021; Accepted: 17 December 2021; Published: 13 January 2022.

Reviewed by:

Copyright © 2022 Schmitt, Kulzer, Ehrmann, Haak and Hermanns. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Andreas Schmitt, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

• - Google Chrome

Intended for healthcare professionals

• Access provided by Google Indexer
• My email alerts
• BMA member login
• Username * Password * Forgot your log in details? Need to activate BMA Member Log In Log in via OpenAthens Log in via your institution

Search form

• Advanced search
• Search responses
• Search blogs
• News & Views
• Dietary and...

Dietary and nutritional approaches for prevention and management of type 2 diabetes

Food for thought, click here to read other articles in this collection.

• Related content
• Peer review
• Nita G Forouhi , professor 1 ,
• Anoop Misra , professor 2 ,
• Viswanathan Mohan , professor 3 ,
• Roy Taylor , professor 4 ,
• William Yancy , director 5 6 7
• 1 MRC Epidemiology Unit, University of Cambridge School of Clinical Medicine, Cambridge, UK
• 2 Fortis-C-DOC Centre of Excellence for Diabetes, Metabolic Diseases and Endocrinology, and National Diabetes, Obesity and Cholesterol Foundation, New Delhi, India
• 3 Dr Mohan’s Diabetes Specialities Centre and Madras Diabetes Research Foundation, Chennai, India
• 4 Magnetic Resonance Centre, Institute of Cellular Medicine, Newcastle University, Newcastle, UK
• 5 Duke University Diet and Fitness Center, Durham, North Carolina, USA
• 6 Department of Medicine, Duke University School of Medicine, Durham, North Carolina, USA
• 7 Center for Health Services Research in Primary Care, Department of Veterans Affairs, Durham, North Carolina, USA
• Correspondence to: N G Forouhi nita.forouhi{at}mrc-epid.cam.ac.uk

Common ground on dietary approaches for the prevention, management, and potential remission of type 2 diabetes can be found, argue Nita G Forouhi and colleagues

Dietary factors are of paramount importance in the management and prevention of type 2 diabetes. Despite progress in formulating evidence based dietary guidance, controversy and confusion remain. In this article, we examine the evidence for areas of consensus as well as ongoing uncertainty or controversy about dietary guidelines for type 2 diabetes. What is the best dietary approach? Is it possible to achieve remission of type 2 diabetes with lifestyle behaviour changes or is it inevitably a condition causing progressive health decline? We also examine the influence of nutrition transition and population specific factors in the global context and discuss future directions for effective dietary and nutritional approaches to manage type 2 diabetes and their implementation.

Why dietary management matters but is difficult to implement

Diabetes is one of the biggest global public health problems: the prevalence is estimated to increase from 425 million people in 2017 to 629 million by 2045, with linked health, social, and economic costs. 1 Urgent solutions for slowing, or even reversing, this trend are needed, especially from investment in modifiable factors including diet, physical activity, and weight. Diet is a leading contributor to morbidity and mortality worldwide according to the Global Burden of Disease Study carried out in 188 countries. 2 The importance of nutrition in the management and prevention of type 2 diabetes through its effect on weight and metabolic control is clear. However, nutrition is also one of the most controversial and difficult aspects of the management of type 2 diabetes.

The idea of being on a “diet” for a chronic lifelong condition like diabetes is enough to put many people off as knowing what to eat and maintaining an optimal eating pattern are challenging. Medical nutrition therapy was introduced to guide a systematic and evidence based approach to the management of diabetes through diet, and its effectiveness has been demonstrated, 3 but difficulties remain. Although most diabetes guidelines recommend starting pharmacotherapy only after first making nutritional and physical activity lifestyle changes, this is not always followed in practice globally. Most physicians are not trained in nutrition interventions and this is a barrier to counselling patients. 4 5 Moreover, talking to patients about nutrition is time consuming. In many settings, outside of specialised diabetes centres where trained nutritionists/educators are available, advice on nutrition for diabetes is, at best, a printed menu given to the patient. In resource poor settings, when type 2 diabetes is diagnosed, often the patient leaves the clinic with a list of new medications and little else. There is wide variation in the use of dietary modification alone to manage type 2 diabetes: for instance, estimates of fewer than 5-10% of patients with type 2 diabetes in India 6 and 31% in the UK are reported, although patients treated by lifestyle measures may be less closely managed than patients on medication for type 2 diabetes. 7 Although systems are usually in place to record and monitor process measures for diabetes care in medical records, dietary information is often neglected, even though at least modest attention to diet is needed to achieve adequate glycaemic control. Family doctors and hospital clinics should collect this information routinely but how to do this is a challenge. 5 8

Progress has been made in understanding the best dietary advice for diabetes but broader problems exist. For instance, increasing vegetable and fruit intake is recommended by most dietary guidelines but their cost is prohibitively high in many settings: the cost of two servings of fruits and three servings of vegetables a day per individual (to fulfil the “5-a-day” guidance) accounted for 52%, 18%, 16%, and 2% of household income in low, low to middle, upper to middle, and high income countries, respectively. 9 An expensive market of foods labelled for use by people with diabetes also exists, with products often being no healthier, and sometimes less healthy, than regular foods. After new European Union legislation, food regulations in some countries, including the UK, were updated as recently as July 2016 to ban such misleading labels. This is not the case elsewhere, however, and what will happen to such regulation after the UK leaves the European Union is unclear, which highlights the importance of the political environment.

Evidence for current dietary guidelines

In some, mostly developed, countries, dietary guidelines for the management of diabetes have evolved from a focus on a low fat diet to the recognition that more important considerations are macronutrient quality (that is, the type versus the quantity of macronutrient), avoidance of processed foods (particularly processed starches and sugars), and overall dietary patterns. Many systematic reviews and national dietary guidelines have evaluated the evidence for optimal dietary advice, and we will not repeat the evidence review. 10 11 12 13 14 15 16 17 18 We focus instead in the following sections on some important principles where broad consensus exists in the scientific and clinical community and highlight areas of uncertainty, but we begin by outlining three underpinning features.

Firstly, an understanding of healthy eating for the prevention and management of type 2 diabetes has largely been derived from long term prospective studies and limited evidence from randomised controlled trials in general populations, supplemented by evidence from people with type 2 diabetes. Many published guidelines and reviews have applied grading criteria and this evidence is often of moderate quality in the hierarchy of evidence that places randomised controlled trials at the top. Elsewhere, it is argued that different forms of evidence evaluating consistency across multiple study designs including large population based prospective studies of clinical endpoints, controlled trials of intermediate pathways, and where feasible randomised trials of clinical endpoints should be used collectively for evidence based nutritional guidance. 19

Secondly, it is now recognised that dietary advice for both the prevention and management of type 2 diabetes should converge, and they should not be treated as different entities ( fig 1 ). However, in those with type 2 diabetes, the degree of glycaemic control and type and dose of diabetes medication should be coordinated with dietary intake. 12 With some dietary interventions, such as very low calorie or low carbohydrate diets, people with diabetes would usually stop or reduce their diabetes medication and be monitored closely, as reviewed in a later section.

Dietary advice for different populations for the prevention and management of type 2 diabetes

• Download figure
• Open in new tab
• Download powerpoint

Thirdly, while recognising the importance of diet for weight management, there is now greater understanding 10 of the multiple pathways through which dietary factors exert health effects through both obesity dependent and obesity independent mechanisms. The influence of diet on weight, glycaemia, and glucose-insulin homeostasis is directly relevant to glycaemic control in diabetes, while other outcomes such as cardiovascular complications are further influenced by the effect of diet on blood lipids, apolipoproteins, blood pressure, endothelial function, thrombosis, coagulation, systemic inflammation, and vascular adhesion. The effect of food and nutrients on the gut microbiome may also be relevant to the pathogenesis of diabetes but further research is needed. Therefore, diet quality and quantity over the longer term are relevant to the prevention and management of diabetes and its complications through a wide range of metabolic and physiological processes.

Areas of consensus in guidelines

Weight management.

Type 2 diabetes is most commonly associated with overweight or obesity and insulin resistance. Therefore, reducing weight and maintaining a healthy weight is a core part of clinical management. Weight loss is also linked to improvements in glycaemia, blood pressure, and lipids and hence can delay or prevent complications, particularly cardiovascular events.

Energy balance

Most guidelines recommend promoting weight loss among overweight or obese individuals by reducing energy intake. Portion control is one strategy to limit energy intake together with a healthy eating pattern that focuses on a diet composed of whole or unprocessed foods combined with physical activity and ongoing support.

Dietary patterns

The evidence points to promoting patterns of food intake that are high in vegetables, fruit, whole grains, legumes, nuts, and dairy products such as yoghurt but with some cautions. Firstly, some dietary approaches (eg, low carbohydrate diets) recommend restricting the intake of fruits, whole grains, and legumes because of their sugar or starch content. For fruit intake, particularly among those with diabetes, opinion is divided among scientists and clinicians (see appendix on bmj.com). Many guidelines continue to recommend fruit, however, on the basis that fructose intake from fruits is preferable to isocaloric intake of sucrose or starch because of the additional micronutrient, phytochemical, and fibre content of fruit. Secondly, despite evidence from randomised controlled trials and prospective studies 10 that nuts may help prevent type 2 diabetes, some (potentially misplaced) concern exists about their high energy content. Further research in people with type 2 diabetes should help to clarify this.

There is also consensus on the benefits of certain named dietary patterns such as the Mediterranean diet for prevention and management of type 2 diabetes. Expert guidelines also support other healthy eating patterns that take account of local sociocultural factors and personal preferences.

Foods to avoid

Consensus exists on reducing or avoiding the intake of processed red meats, refined grains and sugars (especially sugar sweetened drinks) both for prevention and management of type 2 diabetes, again with some cautions. Firstly, for unprocessed red meat, the evidence of possible harm because of the development of type 2 diabetes is less consistent and of a smaller magnitude. More research is needed on specific benefits or harms in people with type 2 diabetes. Secondly, evidence is increasing on the relevance of carbohydrate quality: that is that whole grains and fibre are better choices than refined grains and that fibre intake should be at least as high in people with type 2 diabetes as recommended for the general population, that diets that have a higher glycaemic index and load are associated with an increased risk of type 2 diabetes, and that there is a modest glycaemic benefit in replacing foods with higher glycaemic load with foods with low glycaemic load. However, debate continues about the independence of these effects from the intake of dietary fibre. Some evidence exists that consumption of potato and white rice may increase the risk of type 2 diabetes but this is limited and further research is needed.

Moreover, many guidelines also highlight the importance of reducing the intake of in foods high in sodium and trans fat because of the relevance of these specifically for cardiovascular health.

Areas of uncertainty in guidelines

Optimal macronutrient composition.

One of the most contentious issues about the management of type 2 diabetes has been on the best macronutrient composition of the diet. Some guidelines continue to advise macronutrient quantity goals, such as the European or Canadian recommendation of 45–60% of total energy as carbohydrate, 10–20% as protein, and less than 35% as fat, 13 20 or the Indian guidelines that recommend 50-60% energy from carbohydrates, 10-15% from protein, and less than 30% from fat. 21 In contrast, the most recent nutritional guideline from the American Diabetes Association concluded that there is no ideal mix of macronutrients for all people with diabetes and recommended individually tailored goals. 12 Alternatively, a low carbohydrate diet for weight and glycaemic control has gained popularity among some experts, clinicians, and the public (reviewed in a later section). Others conclude that a low carbohydrate diet combined with low saturated fat intake is best. 22

For weight loss, three points are noteworthy when comparing dietary macronutrient composition. Firstly, evidence from trials points to potentially greater benefits from a low carbohydrate than a low fat diet but the difference in weight loss between diets is modest. 23 Secondly, a comparison of named diet programmes with different macronutrient composition highlighted that the critical factor in effectiveness for weight loss was the level of adherence to the diet over time. 24 Thirdly, the quality of the diet in low carbohydrate or low fat diets is important. 25 26

Research to date on weight or metabolic outcomes in diabetes is complicated by the use of different definitions for the different macronutrient approaches. For instance, the definition of a low carbohydrate diet has ranged from 4% of daily energy intake from carbohydrates (promoting nutritional ketosis) to 40%. 15 Similarly, low fat diets have been defined as fat intake less than 30% of daily energy intake or substantially lower. Given these limitations, the best current approach may be an emphasis on the use of individual assessment for dietary advice and a focus on the pattern of eating that most readily allows the individual to limit calorie intake and improve macronutrient quality (such as avoiding refined carbohydrates).

Regular fish intake of at least two servings a week, including one serving of oily fish (eg, salmon, mackerel, and trout) is recommended for cardiovascular risk prevention but fish intake has different associations with the risk of developing type 2 diabetes across the world—an inverse association, no association, and a positive association. 27 It is thought that the type of fish consumed, preparation or cooking practices, and possible contaminants (eg, methyl mercury and polychlorinated biphenyls) vary by geographical location and contributed to this heterogeneity. More research is needed to resolve whether fish intake should be recommended for the prevention of diabetes. However, the current evidence supports an increase in consumption of oily fish for individuals with diabetes because of its beneficial effects on lipoproteins and prevention of coronary heart disease. Most guidelines agree that omega 3 polyunsaturated fatty acid (fish oil) supplementation for cardiovascular prevention in people with diabetes should not be recommended but more research is needed and the results of the ASCEND (A Study of Cardiovascular Events in Diabetes) trial should help to clarify this. 28

Dairy foods are encouraged for the prevention of type 2 diabetes, with more consistent evidence of the benefits of fermented dairy products, such as yoghurt. Similar to population level recommendations about limiting the intake of foods high in saturated fats and replacing them with foods rich in polyunsaturated fat, the current advice for diabetes also favours low fat dairy products but this is debated. More research is needed to resolve this question.

Uncertainty continues about certain plant oils and tropical oils such as coconut or palm oil as evidence from prospective studies or randomised controlled trials on clinical events is sparse or non-existent. However, olive oil, particularly extra virgin olive oil, has been studied in greater detail with evidence of potential benefits for the prevention and management of type 2 diabetes 29 and the prevention of cardiovascular disease within the context of a Mediterranean diet 30 (see article in this series on dietary fats). 31

Difficulties in setting guidelines

Where dietary guidelines exist (in many settings there are none, or they are adapted from those in developed countries and therefore may not be applicable to the local situation), they vary substantially in whether they are evidence based or opinion pieces, and updated in line with scientific progress or outdated. Their accessibility—both physical availability (eg, through a website or clinic) and comprehensibility— for patients and healthcare professionals varies. They vary also in scope, content, detail, and emphasis on the importance of individualised dietary advice, areas of controversy, and further research needs. The quality of research that informs dietary guidelines also needs greater investment from the scientific community and funders. Moreover, lack of transparency in the development of guidelines and bias in the primary nutritional studies can undermine the development of reliable dietary guidelines; recommendations for their improvement must be heeded. 32

Reversing type 2 diabetes through diet

Type 2 diabetes was once thought to be irreversible and progressive after diagnosis, but much interest has arisen about the potential for remission. Consensus on the definition of remission is a sign of progress: glucose levels lower than the diagnostic level for diabetes in the absence of medications for hyperglycaemia for a period of time (often proposed to be at least one year). 33 34 However, the predominant role of energy deficit versus macronutrient composition of the diet in achieving remission is still controversial.

Remission through a low calorie energy deficit diet

Although the clinical observation of the lifelong, steadily progressive nature of type 2 diabetes was confirmed by the UK Prospective Diabetes Study, 35 rapid normalisation of fasting plasma glucose after bariatric surgery suggested that deterioration was not inevitable. 36 As the main change was one of sudden calorie restriction, a low calorie diet was used as a tool to study the mechanisms involved. In one study of patients with type 2 diabetes, fasting plasma glucose normalised within seven days of following a low calorie diet. 37 This normalisation through diet occurred despite simultaneous withdrawal of metformin therapy. Gradually over eight weeks, glucose stimulated insulin secretion returned to normal. 37 Was this a consequence of calorie restriction or composition of the diet? To achieve the degree of weight loss obtained (15 kg), about 610 kcal a day was provided—510 kcal as a liquid formula diet and about 100 kcal as non-starchy vegetables. The formula diet consisted of 59 g of carbohydrate (30 g as sugars), 11.4 g of fat, and 41 g of protein, including required vitamins and minerals. This high “sugar” approach to controlling blood glucose may be surprising but the critical aspect is not what is eaten but the gap between energy required and taken in. Because of this deficit, the body must use previously stored energy. Intrahepatic fat is used first, and the 30% decrease in hepatic fat in the first seven days appears sufficient to normalise the insulin sensitivity of the liver. 37 In addition, pancreatic fat content fell over eight weeks and beta cell function improved. This is because insulin secretory function was regained by re-differentiation after fat removal. 38

The permanence of these changes was tested by a nutritional and behavioural approach to achieve long term isocaloric eating after the acute weight loss phase. 39 It was successful in keeping weight steady over the next six months of the study. Calorie restriction was associated with both hepatic and pancreatic fat content remaining at the low levels achieved. The initial remission of type 2 diabetes was closely associated with duration of diabetes, and the individuals with type 2 diabetes of shorter duration who achieved normal levels of blood glucose maintained normal physiology during the six month follow-up period. Recently, 46% of a UK primary care cohort remained free of diabetes at one year during a structured low calorie weight loss programme (the DiRECT trial). 40 These results are convincing, and four years of follow-up are planned.

A common criticism of the energy deficit research has been that very low calorie diets may not be achievable or sustainable. Indeed, adherence to most diets in the longer term is an important challenge. 24 However, Look-AHEAD, the largest randomised study of lifestyle interventions in type 2 diabetes (n=5145), randomised individuals to intensive lifestyle management, including the goal to reduce total calorie intake to 1200-1800 kcal/d through a low fat diet assisted by liquid meal replacements, and this approach achieved greater weight loss and non-diabetic blood glucose levels at year 1 and year 4 in the intervention than the control group. 41

Considerable interest has arisen about whether low calorie diets associated with diabetes remission can also help to prevent diabetic complications. Evidence is sparse because of the lack of long term follow-up studies but the existing research is promising. A return to the non-diabetic state brings an improvement in cardiovascular risk (Q risk decreasing from 19.8% to 5.4%) 39 ; case reports of individuals facing foot amputation record a return to a low risk state over 2-4 years with resolution of painful neuropathy 42 43 ; and retinal complications are unlikely to occur or progress. 44 However, other evidence highlights that worsening of treatable maculopathy or proliferative retinopathy may occur following a sudden fall in plasma glucose levels, 45 46 so retinal imaging in 4-6 months is recommended for individuals with more than minimal retinopathy if following a low calorie remission diet. Annual review is recommended for all those in the post-diabetic state, and a “diabetes in remission” code (C10P) is now available in the UK. 34

Management or remission through a low carbohydrate diet

Before insulin was developed as a therapy, reducing carbohydrate intake was the main treatment for diabetes. 47 48 Carbohydrate restriction for the treatment of type 2 diabetes has been an area of intense interest because, of all the macronutrients, carbohydrates have the greatest effect on blood glucose and insulin levels. 49

In a review by the American Diabetes Association, interventions of low carbohydrate (less than 40% of calories) diets published from 2001 to 2010 were identified. 15 Of 11 trials, eight were randomised and about half reported greater improvement in HbA1c on the low carbohydrate diet than the comparison diet (usually a low fat diet), and a greater reduction in the use of medicines to lower glucose. Notably, calorie reduction coincided with carbohydrate restriction in many of the studies, even though it was not often specified in the dietary counselling. One of the more highly controlled studies was an inpatient feeding study, 50 which reported a decline in mean HbA1c from 7.3% to 6.8% (P=0.006) over just 14 days on a low carbohydrate diet.

For glycaemia, other reviews of evidence from randomised trials on people with type 2 diabetes have varying conclusions. 51 52 53 54 55 56 Some concluded that low carbohydrate diets were superior to other diets for glycaemic control, or that a dose response relationship existed, with stricter low carbohydrate restriction resulting in greater reductions in glycaemia. Others cautioned about short term beneficial effects not being sustained in the longer term, or found no overall advantage over the comparison diet. Narrative reviews have generally been more emphatic on the benefits of low carbohydrate diets, including increased satiety, and highlight the advantages for weight loss and metabolic parameters. 57 58 More recently, a one year clinic based study of the low carbohydrate diet designed to induce nutritional ketosis (usually with carbohydrate intake less than 30 g/d) was effective for weight loss, and for glycaemic control and medication reduction. 59 However, the study was not randomised, treatment intensity differed substantially in the intervention versus usual care groups, and participants were able to select their group.

Concerns about potential detrimental effects on cardiovascular health have been raised as low carbohydrate diets are usually high in dietary fat, including saturated fat. For lipid markers as predictors of future cardiovascular events, several studies found greater improvements in high density lipoprotein cholesterol and triglycerides with no relative worsening of low density lipoprotein cholesterol in patients with type 2 diabetes following carbohydrate restriction, 15 with similar conclusions in non-diabetic populations. 57 60 61 62 Low density lipoprotein cholesterol tends to decline more, however, in a low fat comparison diet 61 63 and although low density lipoprotein cholesterol may not worsen with a low carbohydrate diet 63 in the short term, the longer term effects are unclear. Evidence shows that low carbohydrate intake can lower the more atherogenic small, dense low density lipoprotein particles. 57 64 Because some individuals may experience an increase in serum low density lipoprotein cholesterol when following a low carbohydrate diet high in saturated fat, monitoring is important.

Another concern is the effect of the potentially higher protein content of low carbohydrate diets on renal function. Evidence from patients with type 2 diabetes with normal baseline renal function and from individuals without diabetes and with normal or mildly impaired renal function has not shown worsening renal function at one or up to two years of follow-up, respectively. 22 65 66 67 Research in patients with more severely impaired renal function, with or without diabetes, has not been reported to our knowledge. Other potential side effects of a very low carbohydrate diet include headache, fatigue, and muscle cramping but these side effects can be avoided by adequate fluid and sodium intake, particularly in the first week or two after starting the diet when diuresis is greatest. Concern about urinary calcium loss and a possible contribution to increased future risk of kidney stones or osteoporosis 68 have not been verified 69 but evidence is sparse and warrants further investigation. The long term effects on cardiovascular disease and chronic kidney disease in patients with diabetes need further evaluation.

Given the hypoglycaemic effect of carbohydrate restriction, patients with diabetes who adopt low carbohydrate diets and their clinicians must understand how to avoid hypoglycaemia by appropriately reducing glucose lowering medications. Finally, low carbohydrate diets can restrict whole grain intake and although some low carbohydrate foods can provide the fibre and micronutrients contained in grains, it may require greater effort to incorporate such foods. This has led some experts to emphasise restricting refined starches and sugars but retaining whole grains.

Nutrition transition and population specific factors

Several countries in sub-Saharan Africa, South America, and Asia (eg, India and China) have undergone rapid nutrition transition in the past two decades. These changes have paralleled economic growth, foreign investment in the fast food industry, urbanisation, direct-to-consumer marketing of foods high in calories, sale of ultraprocessed foods, and as a result, lower consumption of traditional diets. The effect of these factors on nutrition have led to obesity and type 2 diabetes on the one hand, and co-existing undernutrition and micronutrient deficiencies on the other.

Dietary shifts in low and middle income countries have been stark: in India, these include a substantial increase in fat intake in the setting of an already high carbohydrate intake, with a slight increase in total energy and protein, 70 and a decreasing intake of coarse cereals, pulses, fruits, and vegetables 71 ; in China, animal protein and fat as a percentage of energy has also increased, while cereal intake has decreased. 72 An almost universal increase in the intake of caloric beverages has also occurred, with sugar sweetened soda drinks being the main beverage contributing to energy intake, for example among adults and children in Mexico, 73 or the substantial rise in China in sales of sugar sweetened drinks from 10.2 L per capita in 1998 to 55.0 L per capita in 2012. 74 The movement of populations from rural to urban areas within a country may also be linked with shifts in diets to more unhealthy patterns, 75 while acculturation of immigrant populations into their host countries also results in dietary shifts. 76

In some populations, such as South Asians, rice and wheat flour bread are staple foods, with a related high carbohydrate intake (60-70% of calories). 77 Although time trends show that intake of carbohydrate has decreased among South Asian Indians, the quality of carbohydrates has shifted towards use of refined carbohydrates. 71 The use of oils and traditional cooking practices also have specific patterns in different populations. For instance, in India, the import and consumption of palm oil, often incorporated in the popular oil vanaspati (partially hydrogenated vegetable oil, high in trans fats), is high. 78 Moreover, the traditional Indian cooking practice of frying at high temperatures and re-heating increases trans fatty acids in oils. 79 Such oils are low cost, readily available, and have a long shelf life, and thus are more attractive to people from the middle and low socioeconomic strata but their long term effects on type 2 diabetes are unknown.

Despite the nutrition transition being linked to an increasing prevalence of type 2 diabetes, obesity and other non-communicable diseases, strong measures to limit harmful foods are not in place in many countries. Regulatory frameworks including fiscal policies such as taxation for sugar sweetened beverages need to be strengthened to be effective and other preventive interventions need to be properly implemented. Efforts to control trans fatty acids in foods have gained momentum but are largely confined to developed countries. To reduce consumption in low and middle income countries will require both stringent regulations and the availability and development of alternative choices of healthy and low cost oils, ready made food products, and consumer education. 80 The need for nutritional labelling is important but understanding nutrition labels is a problem in populations with low literacy or nutrition awareness, which highlights the need for educational activities and simpler forms of labelling. The role of dietary/nutritional factors in the predisposition of some ethnic groups to developing type 2 diabetes at substantially lower levels of obesity than European populations 81 is poorly researched and needs investigation.

Despite the challenges of nutritional research, considerable progress has been made in formulating evidence based dietary guidance and some common principles can be agreed that should be helpful to clinicians, patients, and the public. Several areas of uncertainty and controversy remain and further research is needed to resolve these. While adherence to dietary advice is an important challenge, weight management is still a cornerstone in diabetes management, supplemented with new developments, including the potential for the remission of type 2 diabetes through diet.

Future directions

Nutritional research is difficult. Although much progress has been made to improve evidence based dietary guidelines, more investment is needed in good quality research with a greater focus on overcoming the limitations of existing research. Experts should also strive to build consensus using research evidence based on a combination of different study designs, including randomised experiments and prospective observational studies

High quality research is needed that compares calorie restriction and carbohydrate restriction to assess effectiveness and feasibility in the long term. Consensus is needed on definitions of low carbohydrate nutrition. Use of the findings must take account of individual preferences, whole diets, and eating patterns

Further research is needed to resolve areas of uncertainty about dietary advice in diabetes, including the role of nuts, fruits, legumes, fish, plant oils, low fat versus high fat dairy, and diet quantity and quality

Given recent widespread recommendations (such as from the World Health Organization 82 and the UK Scientific Advisory Committee on Nutrition 83 ) to reduce free sugars to under 10% or even 5% of total energy intake in the general population and to avoid sugar sweetened drinks, we need targeted research on the effect of non-nutritive sweeteners on health outcomes in people with diabetes and in the whole population

Most dietary guidelines are derived from evidence from Western countries. Research is needed to better understand the specific aetiological factors that link diet/nutrition and diabetes and its complications in different regions and different ethnic groups. This requires investment in developing prospective cohorts and building capacity to undertake research in low and middle income settings and in immigrant ethnic groups. Up-to-date, evidence based dietary guidelines are needed that are locally relevant and readily accessible to healthcare professionals, patients, and the public in different regions of the world. Greater understanding is also needed about the dietary determinants of type 2 diabetes and its complications at younger ages and in those with lower body mass index in some ethnic groups

We need investment in medical education to train medical students and physicians in lifestyle interventions, including incorporating nutrition education in medical curricula

Individual, collective, and upstream factors are important. Issuing dietary guidance does not ensure its adoption or implementation. Research is needed to understand the individual and societal drivers of and barriers to healthy eating. Educating and empowering individuals to make better dietary choices is an important strategy; in particular, the social aspects of eating need attention as most people eat in family or social groups and counselling needs to take this into account. Equally important is tackling the wider determinants of individual behaviour—the “foodscape”, sociocultural and political factors, globalisation, and nutrition transition

Key messages

Considerable evidence supports a common set of dietary approaches for the prevention and management of type 2 diabetes, but uncertainties remain

Weight management is a cornerstone of metabolic health but diet quality is also important

Low carbohydrate diets as the preferred choice in type 2 diabetes is controversial. Some guidelines maintain that no single ideal percentage distribution of calories from different macronutrients (carbohydrates, fat, or protein) exists, but there are calls to review this in light of emerging evidence on the potential benefits of low carbohydrate diets for weight management and glycaemic control

The quality of carbohydrates such as refined versus whole grain sources is important and should not get lost in the debate on quantity

Recognition is increasing that the focus of dietary advice should be on foods and healthy eating patterns rather than on nutrients. Evidence supports avoiding processed foods, refined grains, processed red meats, and sugar sweetened drinks and promoting the intake of fibre, vegetables, and yoghurt. Dietary advice should be individually tailored and take into account personal, cultural, and social factors

An exciting recent development is the understanding that type 2 diabetes does not have to be a progressive condition but instead there is potential for remission with dietary intervention

Acknowledgments

We thank Sue Brown as a patient representative of Diabetes UK for her helpful comments and insight into this article.

Contributors and sources: The authors have experience and research interests in the prevention and management of type 2 diabetes (NGF, AM, VM, RT, WY), in guideline development (NGF, AM, VM, WY), and in nutritional epidemiology (NGF, VM). Sources of information for this article included published dietary guidelines or medical nutrition therapy guidelines for diabetes, and systematic reviews and primary research articles based on randomised clinical trials or prospective observational studies. All authors contributed to drafting this manuscript, with NGF taking a lead role and she is also the guarantor of the manuscript. All authors gave intellectual input to improve the manuscript and have read and approved the final version.

Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following: NGF receives funding from the Medical Research Council Epidemiology Unit (MC_UU_12015/5). NGF is a member (unpaid) of the Joint SACN/NHS-England/Diabetes-UK Working Group to review the evidence on lower carbohydrate diets compared with current government advice for adults with type 2 diabetes and is a member (unpaid) of ILSI-Europe Qualitative Fat Intake Task Force Expert Group on update on health effects of different saturated fats. AM received honorarium and research funding from Herbalife and Almond Board of California. VM has received funding from Abbott Health Care for meal replacement studies, the Cashew Export Promotion Council of India, and the Almond Board of California for studies on nuts. RT has received funding from Diabetes UK for the Diabetes Remission Clinical Trial and he is a member (unpaid) of the Joint SACN/NHS-England/Diabetes-UK Working Group to review the evidence on lower carbohydrate diets compared to current government advice for adults with type 2 diabetes. WY has received funding from the Veterans Affairs for research projects examining a low carbohydrate diet in patients with diabetes.

Provenance and peer review: Commissioned, externally peer reviewed

This article is one of a series commissioned by The BMJ . Open access fees for the series were funded by Swiss Re, which had no input in to the commissioning or peer review of the articles. The BMJ thanks the series advisers, Nita Forouhi and Dariush Mozaffarian, for valuable advice and guiding selection of topics in the series.

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ .

• International Diabetes Federation
• Forouzanfar MH ,
• Alexander L ,
• Anderson HR ,
• GBD 2013 Risk Factors Collaborators
• Pastors JG ,
• Warshaw H ,
• DiabCare India 2011 Study Group
• Hippisley-Cox J ,
• England CY ,
• Andrews RC ,
• Thompson JL
• Mozaffarian D
• Boucher JL ,
• Cypress M ,
• American Diabetes Association
• Dworatzek PD ,
• Gougeon R ,
• Sievenpiper JL ,
• Williams SL ,
• Canadian Diabetes Association Clinical Practice Guidelines Expert Committee
• English P ,
• Wheeler ML ,
• Dunbar SA ,
• Jaacks LM ,
• Diabetes UK Nutrition Working Group
• MacLeod J ,
• Schwingshackl L ,
• Hoffmann G ,
• Lampousi AM ,
• Mozaffarian D ,
• De Leeuw I ,
• Hermansen K ,
• Diabetes and Nutrition Study Group (DNSG) of the European Association
• National Dietary Guidelines Consensus Group
• Luscombe-Marsh ND ,
• Thompson CH ,
• Tobias DK ,
• Manson JE ,
• Ludwig DS ,
• Willett W ,
• Johnston BC ,
• Kanters S ,
• Bandayrel K ,
• Gardner CD ,
• Bersamin A ,
• Trepanowski JF ,
• Del Gobbo LC ,
• Di Giuseppe D ,
• Forouhi NG ,
• ASCEND Study Collaborative Group
• Portillo MP ,
• Romaguera D ,
• Estruch R ,
• Salas-Salvadó J ,
• PREDIMED Study Investigators
• Krauss RM ,
• Cefalu WT ,
• McCombie L ,
• Turner RC ,
• Holman RR ,
• UK Prospective Diabetes Study (UKPDS) Group
• Guidone C ,
• Valera-Mora E ,
• Hollingsworth KG ,
• Aribisala BS ,
• Mathers JC ,
• Al-Mrabeh A ,
• Leslie WS ,
• Barnes AC ,
• Wagenknecht LE ,
• Look AHEAD Research Group
• Whittington J
• Pearce IA ,
• The Kroc Collaborative Study Group
• Westman EC ,
• Yancy WS Jr . ,
• Humphreys M
• Bisschop PH ,
• De Sain-Van Der Velden MG ,
• Stellaard F ,
• Sargrad K ,
• Mozzoli M ,
• van Wyk HJ ,
• Snorgaard O ,
• Poulsen GM ,
• Andersen HK ,
• Graves DE ,
• Craven TE ,
• Lipkin EW ,
• Margolis KL
• Chaimani A ,
• Schwedhelm C ,
• Noakes TD ,
• Feinman RD ,
• Pogozelski WK ,
• Hallberg SJ ,
• McKenzie AL ,
• Williams PT ,
• de Melo IS ,
• de Oliveira SL ,
• da Rocha Ataide T
• Mansoor N ,
• Vinknes KJ ,
• Veierød MB ,
• Retterstøl K
• Santos FL ,
• Esteves SS ,
• da Costa Pereira A ,
• Sharman MJ ,
• Forsythe CE
• Friedman AN ,
• Foster GD ,
• Jesudason DR ,
• Pedersen E ,
• Brinkworth GD ,
• Buckley JD ,
• Sakhaee K ,
• Brinkley L ,
• Wycherley TP ,
• Singhal N ,
• Sivakumar B ,
• Jaiswal A ,
• Piernas C ,
• Barquera S ,
• Rivera JA ,
• Ebrahim S ,
• De Stavola B ,
• Holmboe-Ottesen G ,
• Dehghan M ,
• Rangarajan S ,
• Prospective Urban Rural Epidemiology (PURE) study investigators
• Bhardwaj S ,
• Ghosh-Jerath S
• Godsland IF ,
• Hughes AD ,
• Chaturvedi N ,
• ↵ World Health Organization. Sugars intake for adults and children. Guideline. WHO, 2015. http://www.who.int/nutrition/publications/guidelines/sugars_intake/en/
• ↵ Scientific Advisory Committee on Nutrition. SACN Carbohydrates and Health Report. Public Health England. London, 2015. https://www.gov.uk/government/publications/sacn-carbohydrates-and-health-report

• Research article
• Open access
• Published: 16 August 2014

A qualitative synthesis of diabetes self-management strategies for long term medical outcomes and quality of life in the UK

• Julia Frost 1 ,
• Ruth Garside 2 ,
• Chris Cooper 3 &
• Nicky Britten 1  

BMC Health Services Research volume  14 , Article number:  348 ( 2014 ) Cite this article

12k Accesses

43 Citations

8 Altmetric

Metrics details

Qualitative research on self-management for people with Type 2 Diabetes Mellitus (T2DM) has typically reported one-off retrospective accounts of individuals’ strategies. The aim of this research was to identify the ways in which self-management strategies are perceived by people with T2DM as being either supportive or unsupportive over time, by using qualitative findings from both longitudinal intervention studies and usual care.

A systematic review of qualitative literature, published between 2000 and 2013, was conducted using a range of searching techniques. 1374 prospective qualitative papers describing patients’ experiences of self-management strategies for T2DM were identified and screened. Of the 98 papers describing qualitative research conducted in the UK, we identified 4 longitudinal studies (3 intervention studies, 1 study of usual care). Key concepts and themes were extracted, reviewed and synthesised using meta-ethnography techniques.

Aspects of self-management strategies in clinical trials (e.g. supported exercise regimens) can be perceived as enabling the control of biomarkers and facilitative of quality of life. In contrast, aspects of self-management strategies outwith trial conditions (e.g. self-monitoring) can be perceived of as negative influences on quality of life. For self-management strategies to be sustainable in the long term, patients require a sense of having a stake in their management that is appropriate for their beliefs and perceptions, timely information and support, and an overall sense of empowerment in managing their diabetes in relation to other aspects of their life. This enables participants to develop flexible diabetes management strategies that facilitate quality of life and long term medical outcomes.

Conclusions

This synthesis has explored how patients give meaning to the experiences of interventions for T2DM and subsequent attempts to balance biomarkers with quality of life in the long term. People with T2DM both construct and draw upon causal accounts as a resource, and a means to counter their inability to balance medical outcomes and quality of life. These accounts can be mediated by the provision of timely and tailored information and support over time, which can allow people to develop a flexible regimen that can facilitate both quality of life and medical outcomes.

Peer Review reports

By 2030, 500 million adults worldwide will have diabetes, with 2.5 million predicted in the UK [ 1 ]. UK policymakers have described the burden associated with the progressive nature of diabetes in terms of direct costs to the NHS and associated healthcare support services; indirect costs to the economy due to loss of productivity; and the personal impact of diabetes, and complications for patients and their families [ 2 ]. The National Service Framework for Diabetes called for a ‘skills-based approach [to] support self-care by improving knowledge, blood glucose control, weight and dietary management, physical activity and wellbeing’ ([ 3 ]:16); and the associated NICE guideline for Type 2 Diabetes Mellitus (T2DM) recommends that people with diabetes should be offered structured education [ 4 ]. However, a review of service organisation and delivery in the UK identified a lack of health services research in diabetes; variability in the quality and range of support that is provided to people with diabetes; and a recognition that supporting self-care behaviours is challenging for many groups of patients with diabetes [ 5 ].

Randomised controlled trials of educational interventions, specifically for T2DM in the UK, have proved inconclusive. A trial of an expert patient education intervention found that it improved glycaemic control, reduced total cholesterol level, body weight, BMI and waist circumference, reduced requirement for diabetes medication, increased consumption of fruit and vegetables, enjoyment of food, knowledge of diabetes, self-empowerment, self-management skills and treatment satisfaction; but there was no overall improvement in quality of life at fourteen months [ 6 ]. A trial of diabetes education and self-management for ongoing and newly diagnosed (DESMOND) programme found that while those in the intervention group reported positive improvements in beliefs about their illness, there was no difference in HbA1c or dimensions of quality of life at twelve months [ 7 ]. Furthermore, a recent systematic review established that tailored interventions for T2DM, hypertension and heart disease had no impact on medication adherence, self-monitoring, exercise, smoking or diet control, while having a modest impact on screening, dietary fat intake and levels of physical exercise [ 8 ].

An observational study found that people with T2DM rationalise their understanding and response to diabetes by externalising control of their condition to health care professionals who are responsible for their care [ 9 ]; while a qualitative study with participants from the DESMOND trial identified that individual orientations (degrees of acceptance or resistance to either their new ‘diabetic’ identity or any perceived consequences of that identity, and perceptions of the required degree of personal responsibility) may mediate both education preference (e.g. group education that is peer or professional led) and ultimately self-management [ 10 ]. While both provide useful insights, they only collected retrospective accounts via one-off interviews [ 11 ]. Given the long term, chronic nature of T2DM this may not capture aspects of change or development in attitudes and behaviours over time.

Existing syntheses of qualitative research (including only non-intervention studies) indicate that patients with T2DM often prioritise the maintenance of their current quality of life over future improvements to their biomarkers. Paterson et al. demonstrated that people learn to balance their diabetes by combining experience with experimenting with strategies for managing their illness [ 12 ]; while Campbell et al. identified that balance required a complex process of understanding and ‘an ability to manipulate dietary and medication regimens in order to live life as fully as possible, rather than limiting social and work activities in order to adhere rigidly to medical advice’ ([ 13 ]:681). More recently, Gommersall et al. have established the salience of culture and gender roles, as well as perceptions of threats to selfhood [ 14 ]. However, we are not aware of any syntheses that have explicitly compared observational and intervention-linked longitudinal qualitative studies of diabetes self-management. Such a synthesis could illuminate the long term sequelae of diabetes self-management strategies, and prove crucial for understanding the impact of the progressive and degenerative nature of diabetes [ 15 , 16 ].

The aim of this qualitative synthesis was to identify the ways in which self-management strategies are perceived by people with T2DM as being either supportive or unsupportive, from prospective qualitative research using data collected on two or more occasions over a twelve month period, and to compare experiences of those taking part in intervention studies and those receiving usual care.

Systematic reviews use a structured approach to identify, appraise and synthesise research [ 17 ]. Here the meta-ethnographic method described by Noblit and Hare was used to conduct this qualitative synthesis [ 18 ]. Meta-ethnography is an interpretive rather than an aggregative approach, which involves the translation of individual qualitative studies into each other, through the re-interpretation and transformation of their theoretical and substantive concepts [ 13 , 16 , 19 ]. We adhered to the enhancing transparency in reporting the synthesis of qualitative research (ENTREQ) statement [ 20 ].

A database search (MEDLINE, EMBASE, Social Policy and Practice, HMIC and PsycINFO, all via OVID) was conducted in February 2013 using terms for diabetes/diabetics (diabet*) combined with a qualitative research methods literature search filter written by the Health Information Research Unit, McMaster University strategy [ 21 ], In addition:

Meta-syntheses of diabetes qualitative research were identified and forwards and backwards citations found;

Key authors were identified and their papers were located and screened;

Key trials were citation chased and searches were made using the trial acronyms as search terms;

Experts in the field were contacted for published and unpublished papers.

References, including titles and abstracts, were then loaded into Endnote X5 [ 22 ], and the search was updated in May 2013.

Papers were included at the title and abstract screening stage [by JF] if they:

Were among adults with T2DM,

Conducted in the UK (reflecting current practice in the National Health Service),

Published since 2000.

Full text papers were obtained and were further screened [by RG, NB] to establish if they:

Used recognised qualitative methods of data collection and analyses,

Used longitudinal qualitative data collection – with two or more periods of data collection

Used follow–up qualitative data collection of at least 12 months.

We appraised the quality of the included papers using a validated appraisal tool [ 23 ] as a means to assess the rigour and validity of the data collection and analysis techniques employed by the research authors (Table  1 ), which informed our analysis of the data [ 24 ].

The research team repeatedly read the studies to identify key themes and concepts, and to identify areas of consonance and dissonance between the included studies. JF produced a structured summary for each paper, and with input from RG and NB, tabulated the key results, concepts and themes for each study, so that they could be compared and contrasted. Quotations from patients and carers (first order data), as well as author interpretations of the data (second order interpretations) were interpreted and integrated by the research team (into third order concepts) to produce a ‘line of argument’ [ 13 , 18 , 19 ].

The purpose of meta-ethnography is to identify where similar concepts and themes from different studies or papers refer to the same entity (congruent synthesis) or to opposing findings (refutational synthesis); this is referred to as the papers’ findings being ‘translated’ into each other [ 18 , 46 , 47 ]. Thus, as interpretations emerged, they were subject to systematic testing within and between the studies, and in consultation with a wider stakeholder group. This group included People with Type 1 or Type 2 Diabetes, members of the South West Peninsula Diabetes Research Network, Diabetologists, General Practitioners (with and without a special interest in diabetes), the Director of Public Health, and community or hospital based Diabetes Specialist Nurses.

The searches resulted in 1374 abstracts (Figure  1 ). Preliminary screening identified that 98 of these papers described qualitative research conducted in the UK. A full text screen identified only four studies (reported in 15 papers) that used longitudinal data methods over 12 months or more. An updated search in May 2013 identified 280 further abstracts of which seven conference abstracts about the four studies were identified. Similar to a meta-ethnography of patients’ experience of managing anti-depressants we identified two groups of papers early in the synthesis process and grouped them along a timeline [ 48 ]. Reflecting the growing use of longitudinal qualitative methods in medical research more generally, the first group concerned serial qualitative studies that were nested within randomized controlled trials of interventions in order to elucidate causal pathways, while the second were stand-alone studies exploring the influence of health services on the conceptualization of illness over time [ 49 ].

Search flowchart. A database search was conducted using terms for diabetes and qualitative research methods, Additional methods included citation chasing and key author/paper identification. Papers were included at the screening stage if they were among adults with T2DM; conducted in the UK; and published since 2000. Full text papers were obtained and screened to establish if recognised longitudinal qualitative methods of data collection and analyses were used.

Six papers concerned three studies where qualitative methods were nested in trials of self-management interventions. The DALY study employed focus groups to explore the impact of an eight week educational programme administered to all trial participants (n = 89) via five trial courses running over a period of one year [ 25 , 26 ]. The Early ACTID study purposively sampled participants from both trial arms and various recruitment sites to elicit patient perspectives of a trial of diet versus diet and exercise versus usual care via face to face and telephone interviews [ 28 ]. The PACCTS study purposively sampled participants from a deprived urban area, and allocated them to four study groups (where HbA1c control was deemed to be ‘good’, ‘bad’, ‘improving’ or ‘deteriorating’), in order to determine the utility of a tele-care support intervention that was titrated to HbA1c results [ 31 – 33 ]. Although the qualitative data collection in the paper by Malpass et al. was only over a period of nine months [ 28 ], its findings were substantiated by data in an associated quantitative paper that extended over twelve months [ 29 ].

Nine papers were about a single sample of patients, recruited from both primary and secondary care within six months of diagnosis with Type 2 Diabetes, and concerning their perceptions of self-management in the context of changes to diabetes service delivery in Scotland. Four papers utilised interviews at baseline six months and 12 months [ 38 – 41 ]; and five papers providing additional data collected at 48 months [ 11 , 42 – 45 ] (Table  2 ). This set of papers provided a rich data set, which variously employed grounded theory, thematic analysis and longitudinal data analysis techniques to interrogate the sixty interviews, in terms of the patients’ perspectives of service delivery [ 39 ] as well as discrete concepts such as control and causality [ 42 , 43 ].

The inclusion of these papers allowed the reviewers to explore any potential influence of either trial design (e.g. a supportive trial environment may enhance the positive impact of the interventions) or study duration (e.g. ongoing information and support, which are central to the underpinning philosophies of the intervention studies, and foster understanding and confidence, and ultimately an enduring self-efficacy). Combining studies with diversity of purpose provides opportunities for comparisons and potentially more fruitful and meaningful insights, than that which may be obtained by synthesising papers that are more obviously methodologically or substantively ‘similar’ [ 50 ].

Using the established synthesis techniques outlined above, the reviewers identified and translated the key concepts in each of the papers (e.g. the individual study authors’ second order interpretations of the ‘raw’ or ‘first order’ patients’ words) into three interlocking third order constructs (Table  3 , first column): Patient as stakeholder (a patient actively engaged with their service provision), Timeliness of support (the appropriate provision of provision and support, often repeatedly, tailored to the current needs and preferences of the patient, and their subjective understanding of their condition), and Empowerment (the willingness and ability of people with diabetes to self-manage in order to achieve purposive and meaningful behaviour change strategies); and from these produced a line of argument about the support needs of people with T2DM (Table  3 ). Subsequent columns in Table  3 summarise the second order concepts, and the narrative ‘translation’ of those concepts, and details of the papers from which they were drawn, which were listed and tabulated so that they could be explored, compared and juxtaposed.We conceive of these three aspects as interlinked and supporting each other, and there is some overlap, but they represent the cornerstones of good support that allow people with T2DM to successfully manage their condition. These aspects can be mutually supporting, leading to a virtuous circle and improved self-management, while their absence may lead to a vicious circle for patients as they become increasingly disempowered, unable to participate in their own self- care and unable to access the support they need to change (Figure  2 ).

Three interlocking concepts. This qualitative synthesis identified three interlocking third order constructs (Patient as stakeholder, Timeliness of support, and Empowerment), and a line of argument that stated that for self-management strategies to be effective, people with diabetes require a sense of ownership of the management of their disease. This can be fostered through the timely provision of information and advice that acknowledges and accounts for their individual circumstances (e.g. disease duration, and prior experience of diabetes management).

Patient as stakeholder

By comparing and synthesising the key concepts in the intervention and usual care studies, a third order concept was developed about the notion of patient as a stakeholder: a patient actively engaged with their service provision. Participants in the intervention studies were strategically encouraged to ‘own’ and actively manage their diabetes, while those in receipt of usual care did not have the advantage of benefiting from the intervention, nor of staff with additional expertise. In its absence they provided their own interpretation of both their condition and their role in its management.

Within the trial of an educational support intervention (PACCTS) self-management was fostered by tele-carers tailoring information to individuals’ needs [ 28 ]. An example is provided of a woman who had had a stroke, where information was provided to her husband to optimise ‘portion control’ [Participant quote, [ 31 ]: 183] at mealtimes, as exercise was not a viable weight-reduction strategy. This advice was subsequently reinforced by the tele-carer acknowledging that weight loss is difficult (using ‘empathy’) and encouragement to the couple to maintain their efforts (‘positive reinforcement’) [Author quotes, [ 31 ]: 183].

Participants receiving an eight week diabetes educational intervention (DALY) similarly valued receiving information that was tailored to their circumstances, in contrast to their previous experience of care:

One participant described the course as an ‘eye-opener’, whilst another said, ‘I’ve learnt more in the first hour here than I’ve learnt in nearly 5 years’. The course provided participants with the details of managing their disease within the context of their everyday lives, with frequent references to learning about ‘individual’ and ‘small things’ [Author quotes, [ 25 ]: 199]

Participants suggested that having nurse tutors who showed ‘integrity, respect and compassion toward them, as well as demonstrating their nursing expertise in diabetes’ fostered a ‘process of moral interaction’ [Author quotes, [ 25 ]: 200]. This facilitated patient participation in both their education, and development of the skills for effective self-management, while nurturing reciprocal health care relationships which are fundamental to ‘quality of life for people with chronic illness’ [Author quote, [ 25 ]: 200].

Respect and reciprocity were also viewed as fundamental precursors to enabling patients to be honest about their ability and desire to self-manage, deemed essential by the PACCTS study authors, for planning sustainable behaviour change beyond the duration of a trial. The tele-carers in the PACCTS study acknowledged that if patients were too regimented in their routines there was the potential to ‘fall off the bandwagon’ [Tele-carer quote, [ 32 ]: 223]; while the participants’ narratives demonstrated how some felt able to try out new management strategies and modify them to fit into their daily routines, and report this to a non-judgmental professional:

‘Other things we discussed were regular food intake. I felt that she really wanted me to have my breakfast at 8, lunch at 12, tea at 4. But I explained that I can not do that. Even when you are getting older, you still have your own way of doing things. We had a discussion about the gap. I would have high readings at lunchtime and she wanted to know the cause. It was probably because I was trying to comply with her regulations but I was not getting out of bed until late, finishing my breakfast 9:30 or 9:45 and then having dinner at 1. So, we decided not to do it like that.’ [Participant quote, [ 32 ]: 224]

In contrast, reflections on ‘usual care’ from participants in the health education study (DALY) reported that health professionals did not always embrace participants’ self-perceived needs, and talked about the difficulties they had in getting professionals to respond in what they considered to be an appropriate fashion (e.g. when patients transferred from oral therapies to insulin). The study authors conclude that ‘the lack of an appropriate response by health professionals highlighted the interaction of factors that can affect the clinical outcomes associated with educational trials’ [Author quote, [ 26 ]:49]

However, while facilitating patient participation in the management of their diabetes, these interventions and trial environments are resource intensive. All of the participants in the EarlyACTID trial had a consultation with a doctor at baseline, six and twelve months, while those in the intervention arms also received:

‘fifteen nurse or dietician visits of twenty minutes each over a twelve month period …seeing the same nurse or dietician throughout the trial.’ [Author quote, [ 28 ]: 259].

A review of the baseline medical histories taken in EarlyACTID identified that, in contrast to NICE guidelines, many patients were not given the opportunity to make any lifestyle changes before commencing on oral hypoglycaemic agents within one month of diagnosis [ 51 ]. As such, those in the intervention arms of the trial (receiving intensive diet, or diet and exercise support), were afforded the first opportunity to set dietary goals, such as moderating fat intake or increasing the amount of fruit and vegetables that they ate [ 52 ]. This led Malpass et al. to conclude that the first year post-diagnosis is a ‘crucial period of time’ where patients can be supported to modify their lifestyles in ways that are necessary to develop a sense of control ‘over time’ [ 28 ] (Italics added).

This is reflected in the levels of support described by some of the participants in the usual care study who, without adequate support and information, developed their own interpretation of the relationship between their experience of diabetes symptoms and service provision over time.

When T2DM is first diagnosed, people often make intuitive inferences about the severity of their condition, based upon common sense notions of both symptoms and locations of service delivery:

‘Mary, for example, gave the very strong impression in all of her interviews that there was little, if anything, about the health services with which she had had contact that had indicated to her that she had a potentially serious disease…Particularly striking in her interviews, however, is the assumptions she had made about why all of her care had remained in general practice. Mary, like most other patients who took part in the study, perceived hospitals as places where “you really get looked after” (Ellen) because they are frequented by diabetes consultants (i.e. specialists) who provide “the ultimate knowledge” (Andy). Accordingly, not receiving a hospital referral and/or having to wait for what was perceived as a long time for an appointment to come through were commonly interpreted by patients as indicating that they could not have a potentially serious disease.’ [Author quote, [ 39 ]: 1428–9]

Similar interpretations were made when participants were prescribed equipment (e.g. blood glucose meters or urine testing sticks) and, in addition, some of those conducting urine testing perceived a negative test result as indicative of an absence of diabetes, which could impede subsequent self- management [ 38 ].

The intervention studies provided a supportive environment in which participants could understand and develop techniques for monitoring and managing their condition which fitted in with their lives, and which they could subsequently try out in the real world setting and discuss with trial staff. In contrast, participants receiving usual care often lacked insight into their condition, which ameliorated both the understanding and confidence required to effectively self-manage, and led to many disengaging with care providers on all but a superficial level.

Timeliness of support

We developed a third order concept around the notion of timeliness of support: the provision of appropriate support and information, often repeatedly, tailored to the current needs and preferences of the patient, and their subjective understanding of the progression of their condition. Without such timely support, patients are unable neither to be stakeholders in decision making nor to be empowered to make changes. While those in the intervention studies received frequent opportunities for re-engagement with professionals and reinforcement of diabetes knowledge, those in the usual care study were left largely to fend for themselves.

Participants receiving the educational intervention (DALY) identified the course as providing optimal conditions for ‘pressurising’ them to take more notice of their health adding that ‘the protected time for learning is very important’ [Participant quotes, [ 25 ]]. They emphasised the significance of ‘real life’ and the need for practitioners to acknowledge that compromise was required for them to conform to treatment regimens [Participant quote, [ 26 ]]. The study authors conclude that timely education can lead people to re-evaluate their perception of diabetes as a ‘mild’ disease, and that this change can allow them to view diabetes as ‘integral’ to themselves, and therefore less of a ‘threat’ to manage [Author quote, [ 26 ]].

Quantitative findings suggest that the DALY intervention produced sustained improvement in both illness attitudes and self-monitoring at 12 months, with the researchers concluding that continued reinforcement may be required to sustain behaviour change and notions of self-involvement [ 27 ].

In the PACCTS study, one of the participants was explicit about the impact of having a supportive tele-carer work with them to identify problems as they arise and negotiate strategies at their own pace:

‘Because I am conscious of the fact that I have to give those figures to somebody and it has been explained to me although I do not dwell on it, the implications if I do not control my levels the fact is I am susceptible to strokes, etc. The underlying factors of diabetes, I do not like to think about it but I have been made aware through calls and general conversation… I really, really love chocolate. I could eat four bars in the morning, and I am not saying I do not touch it but, I am more conscious of the damage.’ [Respondent quote, [ 32 ]: 224].

The authors suggest that adopting a ‘diabetic identity’ is fundamental to effective self-management; although they emphasise that this emerging and evolving identity is not one based on ‘adherence’, but rather an ‘enhanced self-agency…albeit potentially constrained by [one’s] own socio-economic circumstances, demographic profile and other ill-health or mobility restrictions’ [Author quote, [ 26 ]: 280] - illustrated by this account:

‘I don’t think I would be here if I had carried on the way I was…. Within 12 months, I was down to low numbers and now, I am in the 7 s. I have cut the drinking down by 70%… The information, changing over time, has improved me and I think it is invaluable. The call centre has met all my expectations even gone above them…’ [Participant quote, [ 33 ]: 277, Reviewer emphasis]

Both of these quotes emphasise that agency is facilitated by regular, repeated and timely contact [ 33 ].

EarlyACTID participants also identified that confronting the reality of their diagnosis was a motivator for change: ‘I’m in control because fear made me control my diabetes’, and ‘I don’t want to end up on insulin . . . if I can maintain this level of health I will be happy . . . I want to avoid even going on tablets’ [Participant quotes, [ 28 ]: 260]. However, a small number countered that making multiple lifestyle changes could be difficult, with two men concluding that increasing the amount of exercise, while reducing portion sizes, was counterproductive; while a female participant struggled to make changes without the support of staff or family [ 28 ]. This emphasises that change needs to be at a pace that is suitable for an individual, and taking into social roles, in order to achieve sustained behaviour change.

In the absence of on-going support, providing one’s own interpretation can become an enduring aspect of living with diabetes, with implications for both self-monitoring and effective management. Those in the usual care study articulated a tension between wanting to receive knowledge in the early days after diagnosis while, at the same time, being unable or unwilling to articulate their concerns. After diagnosis, some respondents were mindful of taking up the time of health professionals:

‘It takes that long to get an appointment with the GP you feel silly going in and saying er ‘Should I reduce my Metformin?’ and he’ll say ‘Nope’.’ ‘You go to your GP and you’re aware all the time that you’ve got five minutes to get this over and get out, and that’s at the back of your mind. You know, and I’m sitting there thinking ‘I’ve no got to bother her today with all these questions’…I’m thinking ‘oh well some poor soul behind me could have cancer or whatever’, you know.’ [Respondent quotes, [ 40 ]: 1248–9]

Others identified that appointments did not always coincide with their information needs:

‘As I say as time goes on you get more and more used to it and you get more and more able to deal with it yourself. But initially it really is, erm certainly was for me, a real – I was in shock. And inevitably that asks…begs many questions that you want to ask and you’ve got to kind of put in a request to see y’know, well how, wait a minute, a request to see somebody, no, hang on, why can’t I just- why can’t I just have an answer to my question, simple little thing, that’ll put my mind at rest. [Respondent quote, [ 40 ]: 1249]

Over time, some participants were more open about the fear associated with ‘knowing too much’ or adopting what they perceive as a ‘diabetic identity’:

Eric: Erm I think I know enough erm but erm I don’t feel that y’know at the moment I-I don’t need, er,don’t want others to talk to me about diabetes. I think that might suggest that I’m becoming obsessional about the damned thing and I-I don’t know if I want … if you’re sort of searching out people or organisations that are talking about diabetes all the time, you sort of become a diabetic person and erm well you’re somebody else then. Jennifer: No, I read quite a bit about it y’know on the leaflets and that. And sometimes I often think there’s a book that they advertise in all the newspapers and I think “I’m going to send away for that” but sometimes I think you can know too much. So I’ve never done it. (I: What do you mean like in terms of knowing too much? Like because it might worry you more?) Jennifer: Yeah, yes. That’s exactly what I mean.[Respondent quotes. [ 40 ]: 1431]

Furthermore, most patients considered that their need for prompt information and reassurance would attenuate over time [ 39 ], which suggests that, in the absence of appropriate information, some people are unable to appreciate the chronic nature of their diabetes and anticipate their future needs.

The benefit of exploring longitudinal studies is apparent in subsequent interviews, where participants describe the cessation of self-monitoring, although ‘none of the participants reported having been explicitly told by health professionals to stop self-monitoring, nor had they received additional education about self-monitoring after the first year following diagnosis’ [Author quote, [ 42 ]: 494]. A particular note is made of ‘older and less well educated participants’, who the authors believe are particularly vulnerable to negative attitudes of health professionals, and who may continue the process of monitoring without fully engaging in it, for the benefit of the health professional, rather than the patient:

‘Four checks a week, I do. But I write it down, and that’s as far as it goes’ [Participant quote, [ 42 ]: 495]

With time, this lack of engagement may extend from self-monitoring to self-management. With regard to the role of exercise, ‘few participants acknowledged that physical exercise overall was fundamental to their diabetes self-care’ [Author quote, 11: 572], and several emphasised the perceived lack of interest expressed by their health professionals:

‘Well they’ll ask, y’know, what exercise you get … but they haven’t said “Oh I think you should be walking twice as far” no, nothing like that.’ [Participant quote, 11:572]

Likewise, in relation to medicine taking, ‘few respondents claimed to be fully adherent, highlighting forgetfulness as the central reason for this’ [Author quote, [ 44 ]: 493]. Forgetfulness was common for asymptomatic patients where diabetes was not ‘at the forefront of your mind’, for those with busy lives who forgot to take their medication with them, and those with multiple co-morbidities who stressed that, ‘if you take a lot of tablets, you’ve no idea when you’ve taken them, and what you’ve taken’ [Participant quotes, [ 44 ]: 493].

The intervention studies provided structured opportunities for participants to ask questions and to be provided with timely and appropriate information. In contrast, participants in the usual care study describe a cascade of missed opportunities.

Empowerment

The final third order concept was developed around empowerment: the willingness and ability of people with diabetes to self-manage in order to achieve purposive and meaningful behavior change strategies. Participants in the intervention studies were often able to claim empowerment as a consequence of the support and information that they received, although not all patients were willing or able to self-manage. The study of usual care illustrated that when patients are in receipt of standard healthcare, diabetes related ‘quality of life’ may also be mediated by one’s orientation or perspective.

Participants in the DALY study valued collaborative learning:

‘I’ve learnt about other people’s ideas, other people’s problems and you find that you are not on your own. You can learn how they are overcoming the problems.’ [Participant quotes, [ 26 ]: 202].

Accepting a ‘diabetic identity’ and sharing experiences ‘between equals’ facilitated a ‘group empathy’ which enabled participants to ‘analyse motives for their current behaviour’ and provide ‘opportunities for them to learn new skills in relation to self-managing their diabetes.’ [Author quotes, [ 26 ]: 202]:

‘I am able to bend more now. I no longer find it [diabetes] a nuisance.’ [Participant quotes, [ 26 ]: 202].

A similar confidence developed among participants in the EarlyACTID trial, who were able to titrate diet and exercise in such a way that that self-management enabled both blood glucose control and quality of life:

‘For example, Wayne (DPAI) tried hard to follow dietary recommendations but enjoyed drinking alcohol and eating out. Both were key to his friendship and relationship building. To counterbalance the effects of these two behaviours, Wayne would “work a bit harder in the gym the next day”.’ [Author quote, [ 26 ]: 260]

While participants in both of the above educational intervention studies acknowledged that motivation was needed to supplement their learning, those who were supported to improve their level of physical activity were able to identify that a cycle of behaviour change had been set in place, such that improvements to their physical and mental health encouraged them to eat healthy food and persist in their self-management:

‘I always feel better when I come back (from the gym), I always feel I’ve got more energy . . . when you’re exercising you’re saying ‘I’m doing all this, I ought to cut back a bit’ (laughs).’ ‘Having gone, exercised and come back, you feel really rejuvenated, and I think it spurs you on to keep motivated.’ [Participant quotes, [ 28 ]: 260]

Thus physical activity could be viewed as both a motivator for, and integral to, diabetes self-management.

At twelve months, PACCT participants with a baseline HbA1c greater than 7% achieved significant improvement in glycaemic control [ 34 ] and high levels of satisfaction [ 35 ]; while follow-up at three years identified a continuing significant reduction of HbA1c attributed to the intervention, without additional pharmacological means in a sample drawn from a socio-economically deprived urban community [ 36 ]. Despite the PACCTS intervention being well received by both participants and health care providers, a subsequent economic evaluation of PACCTs found it to be borderline cost effective [ 37 ].

However, even within a supportive trial environment, some participants remained unempowered when it came to self-management. In the tele-support study (PACCTS), the authors describe one man who, despite having his knowledge enhanced, was unable to translate that awareness into effective control after two years of study participation:

‘He seems to have found the calls somewhat irritating: they always ask me the same question, ‘are you eating say this, this? ‘ It’s always the same. But, he remarked, if I’d had diabetes for a year I could have understood it but this is fifteen years, well, three years of this now and I know what they are going to say. He feels that he has his diabetes under control… He has not really changed the way he eats, except in relation to the amount of sugar…’ [Author quotes, [ 33 ]: 257]

Furthermore, Cooper et al. conclude that the ‘low uptake of patient education may not just reflect a cultural climate that promotes dependency’, which they attribute to a bureaucratic health care system in the UK and health professionals who lack specialist knowledge, ‘it may also reflect patients’ desires to continue with their passive role’ [[ 25 ]: 204–5].

In contrast to notions of a ‘passive’ patient, the study of usual care unpacked the dynamic relationships between patient perspectives and behaviour, in the absence behaviour change interventions. By exploring changes in causal accounts over four years, Lawton et al. contend that treatment experiences mediate respondents’ disease perceptions [ 43 ]. Whereas ‘Ellen’ maintained her causal account of her diabetes being due to her poor diet over time, ‘Fiona’ amended her account from having a dietary cause to being hereditary, on the assumption that ‘even my son’s got it now… it must be hereditary’ [Participant quote, [ 43 ]: 51]. In contrast ‘Graham’ begins with a hereditary account of his diabetes, and later acknowledges that he ‘made a pig of myself and put on a lot of weight’ [Participant quote, [ 43 ]: 51]. Thus, Lawton et al. suggest that ‘causation accounts may be informed by, and revised in light of, the perceived efficacy of treatments’ [Author quote, [ 43 ]: 51].

Importantly, these ‘causal accounts’ do not ‘simply convey respondents’ “beliefs” about disease causation… they also serve a communicative or interactional role… [and can] be used as vehicles to rationalise, legitimate and/or enable particular approaches to T2DM self-management.’ [Author quote, [ 43 ]: 52]. While ‘Mary’ attributed her diabetes to ‘bad living’, by also emphasising that her mother had lived a healthy life but died of a stroke, she provides legitimacy for not engaging in self-management activities [[ 43 ]: 53]. A legitimising account is also present in this respondent’s account, where medicine taking is viewed as controlling blood glucose to the detriment of quality of life:

‘By his third interview, Callum had “compared notes” with work colleague with T2DM who had recently moved on to insulin, and “seemed to control things a lot better”. By virtue of being able to titrate her insulin doses, this colleague appeared to have the freedom to eat and drink what she wanted, a freedom which Callum professed to desire. At this point, Callum stopped talking about being able to control his own diabetes with tablets and diet, suggesting that “sooner or later, it’s going to become an insulin issue”. He also ceased to blame any “spikes” recorded through SMBG on his continued snacking. Instead, he attributed them to “the tablets no longer working”, and used this to justify bringing forward his appointment with his consultant and negotiating a move to insulin: “I eventually convinced them I was ready for it”’ [Author quote, [ 43 ]: 53].

By re-framing his orientation, this participant is able to legitimise his transition to commencing insulin as a strategic form of self-management that will improve his quality of life.

Participants in the usual care study were able to develop causal accounts that could inhibit their sense of agency and legitimise their inability to self-manage. Furthermore, the development of lay causal narratives and the negation of a ‘diabetic identity’ (resulting from lack of timely information and support) enabled some respondents to absolve themselves of any responsibility for their diabetes and inhibit subsequent behaviour change. In contrast, with the support and information integral to intervention studies, trial participants were better able to achieve empowerment. This enabled participants to develop flexible diabetes management strategies that facilitate (rather than inhibit) quality of life and long term medical outcomes (including blood glucose and weight control.

Having identified three interlocking third order constructs (Patient as stakeholder, Timeliness of support, and Empowerment, that can be positively or negatively reinforcing), the synthesis substantiated a line of argument that stated that for self-management strategies to be effective, people with diabetes require a sense of ownership of the management of their disease. This can be fostered through the timely provision of information and advice that acknowledges and accounts for their individual circumstances (e.g. disease duration, and prior experience of diabetes management).

‘I feel better certainly. I am not getting infections […] Before, I used to get thrush and infection after infection because my blood sugars were out of control. I take more care in things like having my feet done. I do not know, I just feel healthy […] I am sure that I will continue to follow the advice.’ [Participant quote at 3 years; 32: 224]

In contrast, strategies can be undermined when health professionals do not take account of patients’ beliefs and values, or when self-management is limited to monitoring rather than the means to moderate ‘treatments’, (e.g. when advice is generic rather than tailored to an individual’s support needs). While it is acknowledged that not all people with diabetes are willing or able to self-manage their condition, a flexible regimen is associated with a balanced approach to self-management that facilitates quality of life; while a self-management strategy that is perceived as encroaching upon quality of life (e.g. by inhibiting participation in social activity, such as family meals) has little or no positive impact:

‘I walk out and into the pigeon loft at the back door, over to the shop for my cigarette papers’ [Participant quote at 4 years; 11: 573]

Noblit and Hare contend that the objective of an interpretive synthesis is either to make the obvious obvious, make the obvious dubious, or make the hidden obvious [ 18 ]. While it has long been recognised that patients with chronic illness require enduring support to effectively self-manage their condition [ 12 ], this longitudinal qualitative synthesis has demonstrated that this is still not routine practice, and that this omission may have a cumulative deficit for people with T2DM. These findings suggest that, in the absence of timely support and advice, the construction of elaborate lay models over time may have a self-protective effect, which can mitigate a sense of failure and liability. It is incredibly difficult for someone with an enduring and unchallenged hidden causal account that minimises behavioral causes of T2DM and/or the validity of self-management, to become a confident and flexible self-manager, as this requires an acknowledgement of one’s accountability.

Building upon existing retrospective cross-sectional accounts, the usual care study claims that if people do not acquire a stake in their diabetes management shortly after diagnosis, or they lose that stake due to lack of support, this can have an immediate impact upon how they frame ‘their diabetes’ which may subsequently negate their ability to make informed decisions and choices [ 9 , 10 ]. While the on-going information and support, which are central to the underpinning philosophies of the intervention studies, foster understanding and confidence and ultimately self-efficacy, the findings of the study of usual care identify that unsupported self-management can lead to people with diabetes not fully engaging with resources and behaviours, the result of which may be detrimental to their quality of life. Additionally, while the usual care study identified that the empowerment required for the effective management of quality of life and biomarkers can be mediated by a person’s orientation or perspective, even resource intensive interventions cannot guarantee that participants will willingly embrace self-management. By extending the time-frame of papers included in this synthesis (e.g. those with repeat data collection over twelve months, rather than one off interview studies), we are able to build upon the findings of existing syntheses. The ability to achieve balance [ 12 , 13 ] may be mediated by cycles of behavior change, such that with adequate support a positive feedback mechanism can develop in relation to diet and exercise which may facilitate both quality of life and control of biomarkers, while unsupported self-management can inhibit the effective management of Hba1c, as the perceived burden on one’s quality of life may be too great.

These findings suggest that people with diabetes both value and profit from ongoing support and information, from both health professionals and peers [ 53 , 54 ], when it is reciprocal and tailored to their own needs. The challenge is to deliver continuity of individualised care in the context of current changes to the healthcare system in the UK and elsewhere. The emphasis should be on small patient-centered goals, such as weight-loss or portion control, that are achievable, rather than the prioritisation of biomarkers, which some people may perceive as unachievable and burdensome. Not all people with diabetes have the propensity or ability to self-manage, and following Noblit and Hare’s assertion that the purpose of comparative translation is ‘not to achieve closure, but to enable discourse’ [ 18 ], greater understanding is now required of how health professionals may both inhibit and facilitate self-management in specific populations (such as the elderly or those with multiple co-morbidities) who may have additional needs that interact with any diabetes specific information and support. There is now a need to facilitate on-going open dialogue in usual practice, in order to achieve sustainable change in diabetes self-management.

This paper has demonstrated that it is possible to synthesise longitudinal qualitative papers in order to identify strategies for the long term management of T2DM. Although data from only four patient samples were synthesised, this rich a dataset represents over one hundred interviews and ten focus groups, collected over a four year period. This is in keeping with the amount of data synthesised in previous meta-ethnographies [ 13 ].

Combining qualitative papers associated with different objectives and methodological approaches is not without challenge [ 48 , 50 ]. The impact of the ideological and theoretical perspectives inherent in the study of usual care, and their impact on data collection and analysis were not always clear, and we were unable to assess the sustainability of change that could be attributed to educational interventions beyond three years. The authors of the intervention studies identified that the supportive trial environment may have enhanced the positive impact of the interventions [ 53 ], We contend that the comparison between the two types of studies illuminated the information and support that were missing from usual care, and which are necessary for people with diabetes to successfully self-manage. The supportive trial environments allowed patients to test out and discuss new monitoring and management strategies. Both sets of papers suggest that this does not routinely happen in ‘usual care’. Patients describe an inability to articulate or willingness to discuss their concerns early on in the disease trajectory, with few, if any, subsequent opportunities to subsequently ask key questions. This negation may lead to some people never fully grasping the nature, severity, or progressive nature of their condition, or which monitoring and management practices are most appropriate.

All four studies acknowledge that the qualitative study samples may have been self-selected [ 54 ]. Furthermore, most of the participants in the studies were white British; although we anticipate that research currently being conducted by Greenhalgh et al. [ 55 ] could further illuminate the mediating roles of migration and culture [ 56 ]. However, we were able to develop a line of argument that for self-management strategies to be sustainable beyond the duration of clinical trials, patients require timely information and support, a sense of having a role in their management that is appropriate for their beliefs and perceptions, and an overall sense of empowerment in managing their diabetes in relation to other aspects of their life.

Abbreviations

Type 2 diabetes mellitus.

Shaw J, Sicree R, Simmet P: Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Res Clin Pract. 2010, 87: 4-14. 10.1016/j.diabres.2009.10.007.

Article   CAS   PubMed   Google Scholar  

National Collaborating Centre for Chronic Conditions Type 2 DM: National clinical guideline for management in primary and secondary care (update). 2008, London: Royal College of Physicians

Google Scholar  

Department of Health: National Service Framework for Diabetes: Delivery Strategy. 2002, London: DoH

National Institute for Health and Clinical Excellence: Type 2 DM: newer agents for blood glucose control in Type 2 DM (CG87). 2009, http://guidance.nice.org.uk/CG87/Guidance .

Forbes A, Hughes J, Ismail K, While A: The organisation and delivery of diabetes service in the UK: a scoping exercise. 2010, London: HMSO

Deakin T, Cade J, Williams R, Greenwood D: Structured patient education: the diabetes X-pert programme makes a difference. Diabet Med. 2006, 23: 944-954. 10.1111/j.1464-5491.2006.01906.x.

Davis M, Skinner T, Campbell M, Carey M, Cradock S, Dallosso H, Daly H, Doherty Y, Eaton S, Fox C, Oliver L, Rantell K, Rayman G, Khunti K: Effectiveness of the diabetes education and self management for people with newly diagnosed (DESMOND) programme for people with newly diagnosed type 2 DM: cluster randomised controlled trial. Br Med J. 2008, 336: 491-495. 10.1136/bmj.39474.922025.BE.

Article   Google Scholar  

Radhkrishnan K: The efficacy of tailored interventions for self-management outcomes of type 2 DM, hypertension or heart disease: a systematic review. J Adv Nurs. 2012, 68 (3): 496-510. 10.1111/j.1365-2648.2011.05860.x.

Gillibrand W, Flynn M: Forced externalisation of control in people with diabetes: a qualitative exploratory study. J Adv Nurs. 2001, 34 (4): 501-510. 10.1046/j.1365-2648.2001.01779.x.

Ockleford E, Shaw R, Willars J, Dixon-Woods M: Education and self-management for people newly diagnosed with Type 2 DM: a qualitative study of patients’ views. Chron Illness. 2008, 4: 28-37. 10.1177/1742395307086673.

Peel E, Douglas M, Parry O, Lawton J: Type 2 DM and dog walking: patients’ longitudinal perspectives about implementing and sustaining physical activity. Br J Gen Pract. 2010, 60: 570-577. 10.3399/bjgp10X515061.

Article   PubMed   PubMed Central   Google Scholar  

Paterson B, Thorne S, Dewis M: Adapting to and managing diabetes. Image. 1998, 30 (1): 57-62.

CAS   Google Scholar  

Campbell R, Pound P, Pope C, Britten N, Pill R, Morgan M, Donovan J: Evaluating meta-ethnography: a synthesis of qualitative research on lay experiences of diabetes and diabetes care. Soc Sci Med. 2003, 56 (4): 671-684. 10.1016/S0277-9536(02)00064-3.

Article   PubMed   Google Scholar  

Gomersall T, Madill A, Summers L: A metasynthesis of the self-management of type 2 DM. Qual Health Res. 2011, 21 (6): 853-871. 10.1177/1049732311402096.

Saldana J: Longitudinal qualitative research: Analysing change through time. 2003, Walnut Creek: Rowman and Littlefield Publishers

Holland J, Thomson R, Henderson S: Qualitative longitudinal research: A discussion paper. Families and Social Capital ESRC Research Group. 2006, London: ESRC

Egger M, Smith GD, Altman D: Systematic reviews in health care: meta-analysis in context. 2001, London: BMJ, 2

Book   Google Scholar  

Noblit G, Hare R: MetaEthnography: synthesising qualitative studies. 1988, Newbury Park: Sage

Britten N, Campbell R, Pope C, Donovan J, Morgan M, Pill R: Using meta-ethnography to synthesise qualitative research: a worked example. J Health Serv Res Policy. 2002, 7 (4): 209-215. 10.1258/135581902320432732.

Tong A, Flemming K, McInnes E, Oliver S, Craig J: Enhancing transparency in reporting the synthesis of qualitative research: ENTREQ. BMC Med Res Methodol. 2012, 12: 181-10.1186/1471-2288-12-181.

McMaster University Search Filters for MEDLINE in Ovid Syntax and the PubMed translation. http://hiru.mcmaster.ca/hiru/HIRU_Hedges_MEDLINE_Strategies.aspx#Qualitative .

Thomson Reuters (2013) Endnote 5. http://thomsonreuters.com/ .

Wallace A, Croucher K, Quilgars D, Baldwin S: Meeting the challenge: developing systematic reviewing in social policy. Policy Polit. 2004, 32 (4): 455-470. 10.1332/0305573042009444.

Atkins S, Lewin S, Smith H, Engle M, Fretheim A, Volmink J: Conducting a meta-ethnography of qualitative literature: lessons learnt. BMC Med Res Methodol. 2008, 8: 21-10.1186/1471-2288-8-21.

Cooper H, Booth K, Gill G: Patients’ perspectives on diabetes health care education. Health Educ Res. 2003, 18 (2): 191-206. 10.1093/her/18.2.191.

Cooper H, Booth K, Gill G: Using combined research methods for exploring diabetes patient education. Patient Educ Couns. 2003, 51: 45-52. 10.1016/S0738-3991(02)00265-3.

Cooper H, Booth K, Gill G: A trial of empowerment-based education in type 2 DM – global rather than glycaemic benefits. Diabetes Res Clin Pract. 2008, 82: 165-171. 10.1016/j.diabres.2008.07.013.

Malpass A, Andrews R, Turner K: Patients with type 2 DM experiences of making multiple lifestyle changes: a qualitative study. Patient Educ Couns. 2009, 24: 258-263.

Andrews R, Cooper A, Montgomery A, Norcross A, Peters T, Sharp D, Jackson N, Fitssimons K, Bright J, Coulman K, England C, Gorton J, McLenaghan A, Paxton E, Polet A, Thompson C, Dayan C: Diet or diet plus physical activity versus usual care in patients with newly diagnosed type 2 DM: the early ACTID randomised controlled trial. Lancet. 2001, 378: 129-139.

Bennett-Britton I, Kingsley A, White P, Jackson N, Chen M, Andrews R: Effects of diet plus physical activity on psychological outcomes in newly diagnosed Type 2 DM mellitus: results of a randomised controlled trial. 2011, Lisbon, Portuga: European Association for the Study of Diabetes

Gambling T, Long A: Tailoring advice and optimising response: a case study of a telephone-based support for patients with type 2 DM. Fam Pract. 2010, 27: 179-185. 10.1093/fampra/cmp097.

Gambling T, Long A: The realisation of patient-centred care during a 3 year proactive telephone counselling self-care intervention for diabetes. Patient Educ Couns. 2010, 80: 219-226. 10.1016/j.pec.2009.11.007.

Long A, Gambling T: Enhancing health literacy and behavioural change within a tele-care education and support intervention for people with type 2 DM. Health Expect. 2011, 15: 267-282.

Young R, Taylor J, Friede T, Hollis S, Mason J, Lee P, Burns E, Long A, Gambling T, New J, Gibson J: Po-active call center treatment support (PACCTS) to improve glucose control in type 2 DM. Diabetes Care. 2005, 28 (2): 278-282. 10.2337/diacare.28.2.278.

Long A, Gambling T, Young R, Taylor J: PACCTS team: Acceptability and satisfaction with a telecarer approach to the management of type 2 DM. Diabetes Care. 2005, 28 (2): 283-289. 10.2337/diacare.28.2.283.

Narayanan R, Mason J, Taylor J, Long A, Gambling T, New J, Gibson J, Young R: Telemedicine to improve glycaemic control: 3 year results from the Pro-Active Call centre treatment Support (PACCTS) Trial. Diabet Med. 2012, 29: 284-285. 10.1111/j.1464-5491.2011.03352.x.

Mason J, Young R, New J, Gibson M, Long A, Gambling T, Friede T: Economic analysis of a telemedicine intervention to improve glycaemic control in patients with diabetes mellitus. Dis Manag Health Outcome. 2006, 14 (6): 377-385. 10.2165/00115677-200614060-00007.

Lawton J, Peel E, Douglas M, Parry O: ‘Urine testing is a waste of time’: newly diagnosed type 2 DM patients’ perceptions of self-monitoring. Diabet Med. 2004, 21: 1045-1048. 10.1111/j.1464-5491.2004.01286.x.

Lawton J, Peel E, Parry O, Araos G, Douglas M: Lay perceptions of type 2 DM in Scotland: bringing health services back in. Soc Sci Med. 2005, 60: 1423-1435. 10.1016/j.socscimed.2004.08.013.

Lawton J, Parry O, Peel E, Douglas M: Diabetes service provision: a qualitative study of newly diagnosed type 2 DM patients’ experiences and views. Diabet Med. 2005, 22: 1246-1251. 10.1111/j.1464-5491.2005.01619.x.

Parry O, Peel E, Douglas M, Lawton J: Issues of cause and control in patients accounts of type 2 DM. Health Educ Res. 2006, 21 (1): 97-107.

Peel E, Douglas M, Lawton J: Self-monitoring of blood glucose in type 2 DM: longitudinal qualitative study of patients’ perspectives. Br Med J. 2007, 335: 493-10.1136/bmj.39302.444572.DE.

Lawton J, Peel E, Parry O, Douglas M: Shifting accountability: a longitudinal qualitative study of diabetes causation accounts. Soc Sci Med. 2008, 67: 47-56. 10.1016/j.socscimed.2008.03.028.

Lawton J, Peel E, Parry O, Douglas M: Patients’ perceptions and experiences of taking oral glucose-lowering agents: a longitudinal qualitative study. Diabet Med. 2008, 25: 491-495. 10.1111/j.1464-5491.2008.02400.x.

Lawton J, Rankin D, Peel E, Douglas M: Patient’s perceptions and experiences of transitions in diabetes care: a longitudinal qualitative study. Health Expect. 2009, 12: 138-148.

PubMed   Google Scholar  

Garside R, Britten N, Stein K: The experience of heavy menstrual bleeding: a systematic review and meta-ethnography of qualitative studies. J Adv Nurs. 2008, 63 (6): 550-562. 10.1111/j.1365-2648.2008.04750.x.

Grose J, Frost J, Richardson J, Skirton H: Using meta-ethnography to understand the emotional impact of caring for people with increasing cognitive impairment. Nurs Health Sci. 2013, 15. 1: 113-123.

Malpass A, Shaw A, Sharp D, Walter F, Feder G, Ridd M, Jessler D: ‘Medication career’ or ‘Moral career’? The two sides of managing antidepressants: a meta-ethnography of patients’ experience of antidepressants. Soc Sci Med. 2009, 68: 154-168. 10.1016/j.socscimed.2008.09.068.

Murray S, Kendall M, Carduff E, Worth A, Harris F, Lloyd A, Cavers D, Grant L: Use of serial qualitative interviews to understand patients evolving experiences and needs. Br Med J. 2009, 339: b3702-10.1136/bmj.b3702.

Reid B, Sinclair M, Barr O: Reflections on the methodological challenges of undertaking a meta-synthesis: a response to Heyman. Soc Sci Med. 2009, 69: 1577-1579. 10.1016/j.socscimed.2009.09.005.

Thompson C, Andrews R: Conference: Diabetes UK’s Annual Professional Conference. Are we giving lifestyle management a chance? Diabetic Medicine. 2009, Glasgow: SECC

Gorton J, Paxton L, England C, Coulman K, Andrews R: Conference: Diabetes UK’s Annual Professional Conference. Dietary goal setting in the Early Activity in Diabetes (EarlyACTID study), Diabetic Medicine. 2009, Glasgow: SECC

Rogers A, Lee V, Kennedy A: Continuity and change? Exploring reactions to guided self-management in a randomized controlled trial for IBS with reference to prior experience of managing a long term condition. Trials. 2007, 8: 6-10.1186/1745-6215-8-6. doi: 10.1186/1745-6215-8-6

Clark T: On ‘being researched’: why do people engage with qualitative research?. Qual Res. 2010, 10 (4): 399-419. 10.1177/1468794110366796.

Greenhalgh T, Collard A, Campbell-Richards D, Vijayaraghavan S, Malik F, Morris J, Claydon A: Storylines of self-management: narratives of people with diabetes from a multiethnic inner city population. J Health Serv Res Pol. 2010, 16 (1): 37-43.

Lawton J, Ahmad N, Hanna L, Douglas M, Hallowell N: Diabetes service provision: a qualitative study of the experiences and views of Pakistani and Indian patients with type 2 diabetes. Diabet Med. 2005, 23: 1003-1007.

Pre-publication history

The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1472-6963/14/348/prepub

Download references

Acknowledgements

We thank all of the patients, clinicians and commissioners who participated in the advisory group and stakeholder workshops, as part of this research.

This is a summary of independent research funded by the National Institute for Health Research (NIHR)’s Research for Patient Benefit programme (Grant Reference Number PB-PG-0909-19257). NB and RG are partially supported by the NIHR Collaboration for Leadership in Applied Health Research and Care (CLAHRC) for the South West Peninsula. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health in England. RG is partially funded by The European Centre for Environment and Human Health, which is supported by investment from the European Regional Development Fund and the European Social Fund Convergence Programme for Cornwall and the Isles of Scilly.

Author information

Authors and affiliations.

Institute of Health Research, University of Exeter Medical School, Veysey Building, Salmon Pool Lane, Exeter, EX2 4SG, UK

Julia Frost & Nicky Britten

European Centre for Environment and Human Health, University of Exeter Medical School, Knowledge Spa, Royal Cornwall Hospital, Truro, TR1 3HD, UK

Ruth Garside

Peninsula Technology Assessment Group (PenTAG), University of Exeter Medical School, Veysey Building, Salmon Pool Lane, Exeter, EX2 4SG, UK

Chris Cooper

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Julia Frost .

Additional information

Competing interests.

The authors declare that they have no competing interests.

Authors’ contributions

JF carried out preliminary screening and data extraction and drafted the manuscript. CC conceived the search strategy and undertook the searches. RG participated in the design of the study and undertook second screening and data extraction coordination and helped to draft the manuscript. NB conceived the study, and participated in its design and coordination and helped to draft the manuscript. JF, RG and NB conducted the data analysis and synthesis. All authors read and approved the final manuscript.

Authors’ original submitted files for images

Below are the links to the authors’ original submitted files for images.

Authors’ original file for figure 1

Authors’ original file for figure 2, rights and permissions.

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( http://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Cite this article.

Frost, J., Garside, R., Cooper, C. et al. A qualitative synthesis of diabetes self-management strategies for long term medical outcomes and quality of life in the UK. BMC Health Serv Res 14 , 348 (2014). https://doi.org/10.1186/1472-6963-14-348

Download citation

Received : 13 December 2013

Accepted : 04 August 2014

Published : 16 August 2014

DOI : https://doi.org/10.1186/1472-6963-14-348

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Meta-ethnography
• Self-management
• Longitudinal
• Qualitative
• Quality of life

BMC Health Services Research

ISSN: 1472-6963

• General enquiries: [email protected]

• Program Finder
• Admissions Services
• Course Directory
• Academic Calendar
• Hybrid Campus
• Lecture Series
• Convocation
• Strategy and Development
• Implementation and Impact
• Integrity and Oversight
• In the School
• In the Field
• In Baltimore
• Resources for Practitioners
• Articles & News Releases
• In The News
• Statements & Announcements
• At a Glance
• Student Life
• Strategic Priorities
• Inclusion, Diversity, Anti-Racism, and Equity (IDARE)
• What is Public Health?

Research Gaps Around Type 1 Diabetes

A large body of research on Type 2 diabetes has helped to develop guidance, informing how patients are diagnosed, treated, and manage their lifestyle. In contrast, Type 1 diabetes, often mistakenly associated only with childhood, has received less attention.

In this Q&A, adapted from the  April 17 episode of Public Health On Call , Stephanie Desmon speaks to Johns Hopkins epidemiologists  Elizabeth Selvin , PhD '04, MPH, and  Michael Fang , PhD, professor and assistant professor, respectively, in the Department of Epidemiology, about recent findings that challenge common beliefs about type 1 diabetes. Their conversation touches on the misconception that it’s solely a childhood condition, the rise of adult-onset cases linked to obesity, and the necessity for tailored approaches to diagnosis and care. They also discuss insulin prices and why further research is needed on medications like Ozempic in treating Type 1 diabetes.

I want to hear about some of your research that challenges what we have long understood about Type 1 diabetes, which is no longer called childhood diabetes. 

MF: Type 1 diabetes was called juvenile diabetes for the longest time, and it was thought to be a disease that had a childhood onset. When diabetes occurred in adulthood it would be type 2 diabetes. But it turns out that approximately half of the cases of Type 1 diabetes may occur during adulthood right past the age of 20 or past the age of 30.

The limitations of these initial studies are that they've been in small clinics or one health system. So, it's unclear whether it's just that particular clinic or whether it applies to the general population more broadly. 

We were fortunate because the CDC has collected new data that explores Type 1 diabetes in the U.S. Some of the questions they included in their national data were, “Do you have diabetes? If you do, do you have Type 1 or Type 2? And, at what age were you diagnosed?”

With these pieces of information, we were able to characterize how the age of diagnosis of Type 1 diabetes differs in the entire U.S. population.

Are Type 1 and Type 2 diabetes different diseases?

ES:  They are very different diseases and have a very different burden. My whole career I have been a Type 2 diabetes epidemiologist, and I’ve been very excited to expand work with Type 1 diabetes.

There are about 1.5 million adults with Type 1 diabetes in the U.S., compared to 21 million adults with Type 2 diabetes. In terms of the total cases of diabetes, only 5 to 10 percent have Type 1 diabetes. Even in our largest epidemiologic cohorts, only a small percentage of people have Type 1 diabetes. So, we just don't have the same national data, the same epidemiologic evidence for Type 1 diabetes that we have for Type 2. The focus of our research has been trying to understand and characterize the general epidemiology and the population burden of Type 1 diabetes.

What is it about Type 1 that makes it so hard to diagnose?

MF: The presentation of symptoms varies by age of diagnosis. When it occurs in children, it tends to have a very acute presentation and the diagnosis is easier to make. When it happens in adulthood, the symptoms are often milder and it’s often misconstrued as Type 2 diabetes. 

Some studies have suggested that when Type 1 diabetes occurs in adulthood, about 40% of those cases are misdiagnosed initially as Type 2 cases. Understanding how often people get diagnosed later in life is important to correctly diagnose and treat patients. 

Can you talk about the different treatments?

MF:  Patients with Type 1 diabetes are going to require insulin. Type 2 diabetes patients can require insulin, but that often occurs later in the disease, as oral medications become less and less effective.

ES: Because of the epidemic of overweight and obese in the general population, we’re seeing a lot of people with Type 1 diabetes who are overweight and have obesity. This can contribute to issues around misdiagnosis because people with Type 1 diabetes will have signs and will present similarly to Type 2 diabetes. They'll have insulin resistance potentially as a result of weight gain metabolic syndrome. Some people call it double diabetes—I don't like that term—but it’s this idea that if you have Type 1 diabetes, you can also have characteristics of Type 2 diabetes as well.

I understand that Type 1 used to be considered a thin person's disease, but that’s not the case anymore.  MF:  In a separate paper, we also explored the issue of overweight and obesity in persons with Type 1 diabetes. We found that approximately 62% of adults with Type 1 diabetes were either overweight or obese, which is comparable to the general U.S. population.

But an important disclaimer is that weight management in this population [with Type 1 diabetes] is very different. They can't just decide to go on a diet, start jogging, or engage in rigorous exercise. It can be a very, very dangerous thing to do.

Everybody's talking about Ozempic and Mounjaro—the GLP-1 drugs—for diabetes or people who are overweight to lose weight and to solve their diabetes. Where does that fit in with this population?

ES: These medications are used to treat Type 2 diabetes in the setting of obesity. Ozempic and Mounjaro are incretin hormones. They mediate satiation, reduce appetite, slow gastric emptying, and lower energy intake. They're really powerful drugs that may be helpful in Type 1 diabetes, but they're  not approved for the management of obesity and Type 1 diabetes. At the moment, there aren't data to help guide their use in people with Type 1 diabetes, but I suspect they're going to be increasingly used in people with Type 1 diabetes.

MF:   The other piece of managing weight—and it's thought to be foundational for Type 1 or Type 2—is dieting and exercising. However, there isn’t good guidance on how to do this in persons with Type 1 diabetes, whereas there are large and rigorous trials in Type 2 patients. We’re really just starting to figure out how to safely and effectively manage weight with lifestyle changes for Type 1 diabetics, and I think that's an important area of research that should continue moving forward.

ES: Weight management in Type 1 diabetes is complicated by insulin use and the risk of hypoglycemia, or your glucose going too low, which can be an acute complication of exercise. In people with Type 2 diabetes, we have a strong evidence base for what works. We know modest weight loss can help prevent the progression and development of Type 2 diabetes, as well as weight gain. In Type 1, we just don't have that evidence base.

Is there a concern about misdiagnosis and mistreatment? Is it possible to think a patient has Type 2 but they actually have Type 1? 

MF: I think so. Insulin is the overriding concern. In the obesity paper, we looked at the percentage of people who said their doctors recommended engaging in more exercise and dieting. We found that people with Type 1 diabetes were less likely to receive the same guidance from their doctor. I think providers may be hesitant to say, “Look, just go engage in an active lifestyle.”

This is why it's important to have those studies and have that guidance so that patients and providers can be comfortable in improving lifestyle management.

Where is this research going next?

ES:  What's clear from these studies is that the burden of overweight and obesity is substantial in people with Type 1 diabetes and it's not adequately managed. Going forward, I think we're going to need clinical trials, clear clinical guidelines, and patient education that addresses how best to tackle obesity in the setting of Type 1 diabetes.

It must be confusing for people with Type 1 diabetes who are   hearing about people losing all this weight on these drugs, but they go to their doctor who says, “Yeah, but that's not for you.”

ES: I hope it's being handled more sensitively. These drugs are being used by all sorts of people for whom they are not indicated, and I'm sure that people with Type 1 diabetes are accessing these drugs. I think the question is, are there real safety issues? We need thoughtful discussion about this and some real evidence to make sure that we're doing more good than harm.

MF:  Dr. Selvin’s group has published a paper, estimating that about 15% of people with Type 1 diabetes are on a GLP-1. But we don't have great data on what potentially can happen to individuals.

The other big part of diabetes that we hear a lot about is insulin and its price. Can you talk about your research on this topic?

MF:  There was a survey that asked, “Has there been a point during the year when you were not using insulin because you couldn’t afford it?” About 20% of adults under the age of 65 said that at some point during the year, they couldn't afford their insulin and that they did engage in what sometimes is called “cost-saving rationing” [of insulin].

Medicare is now covering cheaper insulin for those over 65, but there are a lot of people for whom affordability is an issue. Can you talk more about that? 

MF:  The fight is not over. Just because there are national and state policies, and now manufacturers have been implementing price caps, doesn't necessarily mean that the people who need insulin the most are now able to afford it. 

A recent study in the  Annals of Internal Medicine looked at states that adopted or implemented out-of-pocket cost caps for insulin versus those that didn't and how that affected insulin use over time. They found that people were paying less for insulin, but the use of insulin didn't change over time. The $35 cap is an improvement, but we need to do more.

ES: There are still a lot of formulations of insulin that are very expensive. $35 a month is not cheap for someone who is on insulin for the rest of their lives.

• Overweight and Obesity in People With Type 1 Diabetes Nearly Same as General Population
• The Impacts of COVID-19 on Diabetes and Insulin
• Why Eli Lilly’s Insulin Price Cap Announcement Matters

Related Content

To Protect Human Health, We Must Protect the Earth’s Health

New Grant Enables Johns Hopkins Researchers to Implement Community Health Worker-Led Health Interventions for Noncommunicable Diseases in Nepal

The Omnipresence of PFAS—and What We Can Do About Them

Summer Institute in Data to Policy

Filters and Digital Health Program Reduced Participants' Arsenic Levels by Nearly Half in American Indian Households Relying on Well Water