systematic literature review management

Advertisement

How to conduct systematic literature reviews in management research: a guide in 6 steps and 14 decisions

• Review Paper
• Open access
• Published: 12 May 2023
• Volume 17 , pages 1899–1933, ( 2023 )

Cite this article

You have full access to this open access article

• Philipp C. Sauer   ORCID: orcid.org/0000-0002-1823-0723 1 &
• Stefan Seuring   ORCID: orcid.org/0000-0003-4204-9948 2  

23k Accesses

42 Citations

6 Altmetric

Explore all metrics

Systematic literature reviews (SLRs) have become a standard tool in many fields of management research but are often considerably less stringently presented than other pieces of research. The resulting lack of replicability of the research and conclusions has spurred a vital debate on the SLR process, but related guidance is scattered across a number of core references and is overly centered on the design and conduct of the SLR, while failing to guide researchers in crafting and presenting their findings in an impactful way. This paper offers an integrative review of the widely applied and most recent SLR guidelines in the management domain. The paper adopts a well-established six-step SLR process and refines it by sub-dividing the steps into 14 distinct decisions: (1) from the research question, via (2) characteristics of the primary studies, (3) to retrieving a sample of relevant literature, which is then (4) selected and (5) synthesized so that, finally (6), the results can be reported. Guided by these steps and decisions, prior SLR guidelines are critically reviewed, gaps are identified, and a synthesis is offered. This synthesis elaborates mainly on the gaps while pointing the reader toward the available guidelines. The paper thereby avoids reproducing existing guidance but critically enriches it. The 6 steps and 14 decisions provide methodological, theoretical, and practical guidelines along the SLR process, exemplifying them via best-practice examples and revealing their temporal sequence and main interrelations. The paper guides researchers in the process of designing, executing, and publishing a theory-based and impact-oriented SLR.

Similar content being viewed by others

Reporting reliability, convergent and discriminant validity with structural equation modeling: A review and best-practice recommendations

Research Methodology: An Introduction

RMSEA, CFI, and TLI in structural equation modeling with ordered categorical data: The story they tell depends on the estimation methods

Avoid common mistakes on your manuscript.

1 Introduction

The application of systematic or structured literature reviews (SLRs) has developed into an established approach in the management domain (Kraus et al. 2020 ), with 90% of management-related SLRs published within the last 10 years (Clark et al. 2021 ). Such reviews help to condense knowledge in the field and point to future research directions, thereby enabling theory development (Fink 2010 ; Koufteros et al. 2018 ). SLRs have become an established method by now (e.g., Durach et al. 2017 ; Koufteros et al. 2018 ). However, many SLR authors struggle to efficiently synthesize and apply review protocols and justify their decisions throughout the review process (Paul et al. 2021 ) since only a few studies address and explain the respective research process and the decisions to be taken in this process. Moreover, the available guidelines do not form a coherent body of literature but focus on the different details of an SLR, while a comprehensive and detailed SLR process model is lacking. For example, Seuring and Gold ( 2012 ) provide some insights into the overall process, focusing on content analysis for data analysis without covering the practicalities of the research process in detail. Similarly, Durach et al. ( 2017 ) address SLRs from a paradigmatic perspective, offering a more foundational view covering ontological and epistemological positions. Durach et al. ( 2017 ) emphasize the philosophy of science foundations of an SLR. Although somewhat similar guidelines for SLRs might be found in the wider body of literature (Denyer and Tranfield 2009 ; Fink 2010 ; Snyder 2019 ), they often take a particular focus and are less geared toward explaining and reflecting on the single choices being made during the research process. The current body of SLR guidelines leaves it to the reader to find the right links among the guidelines and to justify their inconsistencies. This is critical since a vast number of SLRs are conducted by early-stage researchers who likely struggle to synthesize the existing guidance and best practices (Fisch and Block 2018 ; Kraus et al. 2020 ), leading to the frustration of authors, reviewers, editors, and readers alike.

Filling these gaps is critical in our eyes since researchers conducting literature reviews form the foundation of any kind of further analysis to position their research into the respective field (Fink 2010 ). So-called “systematic literature reviews” (e.g., Davis and Crombie 2001 ; Denyer and Tranfield 2009 ; Durach et al. 2017 ) or “structured literature reviews” (e.g., Koufteros et al. 2018 ; Miemczyk et al. 2012 ) differ from nonsystematic literature reviews in that the analysis of a certain body of literature becomes a means in itself (Kraus et al. 2020 ; Seuring et al. 2021 ). Although two different terms are used for this approach, the related studies refer to the same core methodological references that are also cited in this paper. Therefore, we see them as identical and abbreviate them as SLR.

There are several guidelines on such reviews already, which have been developed outside the management area (e.g. Fink 2010 ) or with a particular focus on one management domain (e.g., Kraus et al. 2020 ). SLRs aim at capturing the content of the field at a point in time but should also aim at informing future research (Denyer and Tranfield 2009 ), making follow-up research more efficient and productive (Kraus et al. 2021 ). Such standalone literature reviews would and should also prepare subsequent empirical or modeling research, but usually, they require far more effort and time (Fisch and Block 2018 ; Lim et al. 2022 ). To achieve this preparation, SLRs can essentially a) describe the state of the literature, b) test a hypothesis based on the available literature, c) extend the literature, and d) critique the literature (Xiao and Watson 2019 ). Beyond guiding the next incremental step in research, SLRs “may challenge established assumptions and norms of a given field or topic, recognize critical problems and factual errors, and stimulate future scientific conversations around that topic” (Kraus et al. 2022 , p. 2578). Moreover, they have the power to answer research questions that are beyond the scope of individual empirical or modeling studies (Snyder 2019 ) and to build, elaborate, and test theories beyond this single study scope (Seuring et al. 2021 ). These contributions of an SLR may be highly influential and therefore underline the need for high-quality planning, execution, and reporting of their process and details.

Regardless of the individual aims of standalone SLRs, their numbers have exponentially risen in the last two decades (Kraus et al. 2022 ) and almost all PhD or large research project proposals in the management domain include such a standalone SLR to build a solid foundation for their subsequent work packages. Standalone SLRs have thus become a key part of management research (Kraus et al. 2021 ; Seuring et al. 2021 ), which is also underlined by the fact that there are journals and special issues exclusively accepting standalone SLRs (Kraus et al. 2022 ; Lim et al. 2022 ).

However, SLRs require a commitment that is often comparable to an additional research process or project. Hence, SLRs should not be taken as a quick solution, as a simplistic, descriptive approach would usually not yield a publishable paper (see also Denyer and Tranfield 2009 ; Kraus et al. 2020 ).

Furthermore, as with other research techniques, SLRs are based on the rigorous application of rules and procedures, as well as on ensuring the validity and reliability of the method (Fisch and Block 2018 ; Seuring et al. 2021 ). In effect, there is a need to ensure “the same level of rigour to reviewing research evidence as should be used in producing that research evidence in the first place” (Davis and Crombie 2001 , p.1). This rigor holds for all steps of the research process, such as establishing the research question, collecting data, analyzing it, and making sense of the findings (Durach et al. 2017 ; Fink 2010 ; Seuring and Gold 2012 ). However, there is a high degree of diversity where some would be justified, while some papers do not report the full details of the research process. This lack of detail contrasts with an SLR’s aim of creating a valid map of the currently available research in the reviewed field, as critical information on the review’s completeness and potential reviewer biases cannot be judged by the reader or reviewer. This further impedes later replications or extensions of such reviews, which could provide longitudinal evidence of the development of a field (Denyer and Tranfield 2009 ; Durach et al. 2017 ). Against this observation, this paper addresses the following question:

Which decisions need to be made in an SLR process, and what practical guidelines can be put forward for making these decisions?

Answering this question, the key contributions of this paper are fourfold: (1) identifying the gaps in existing SLR guidelines, (2) refining the SLR process model by Durach et al. ( 2017 ) through 14 decisions, (3) synthesizing and enriching guidelines for these decisions, exemplifying the key decisions by means of best practice SLRs, and (4) presenting and discussing a refined SLR process model.

In some cases, we point to examples from operations and supply chain management. However, they illustrate the purposes discussed in the respective sections. We carefully checked that the arguments held for all fields of management-related research, and multiple examples from other fields of management were also included.

2 Identification of the need for an enriched process model, including a set of sequential decisions and their interrelations

In line with the exponential increase in SLR papers (Kraus et al. 2022 ), multiple SLR guidelines have recently been published. Since 2020, we have found a total of 10 papers offering guidelines on SLRs and other reviews for the field of management in general or some of its sub-fields. These guidelines are of double interest to this paper since we aim to complement them to fill the gap identified in the introduction while minimizing the doubling of efforts. Table 1 lists the 10 most recent guidelines and highlights their characteristics, research objectives, contributions, and how our paper aims to complement these previous contributions.

The sheer number and diversity of guideline papers, as well as the relevance expressed in them, underline the need for a comprehensive and exhaustive process model. At the same time, the guidelines take specific foci on, for example, updating earlier guidelines to new technological potentials (Kraus et al. 2020 ), clarifying the foundational elements of SLRs (Kraus et al. 2022 ) and proposing a review protocol (Paul et al. 2021 ) or the application and development of theory in SLRs (Seuring et al. 2021 ). Each of these foci fills an entire paper, while the authors acknowledge that much more needs to be considered in an SLR. Working through these most recent guidelines, it becomes obvious that the common paper formats in the management domain create a tension for guideline papers between elaborating on a) the SLR process and b) the details, options, and potentials of individual process steps.

Our analysis in Table 1 evidences that there are a number of rich contributions on aspect b), while the aspect a) of SLR process models has not received the same attention despite the substantial confusion of authors toward them (Paul et al. 2021 ). In fact, only two of the most recent guidelines approach SLR process models. First, Kraus et al. ( 2020 ) incrementally extended the 20-year-old Tranfield et al. ( 2003 ) three-stage model into four stages. A little later, Paul et al. ( 2021 ) proposed a three-stage (including six sub-stages) SPAR-4-SLR review protocol. It integrates the PRISMA reporting items (Moher et al. 2009 ; Page et al. 2021 ) that originate from clinical research to define 14 actions stating what items an SLR in management needs to report for reasons of validity, reliability, and replicability. Almost naturally, these 14 reporting-oriented actions mainly relate to the first SLR stage of “assembling the literature,” which accounts for nine of the 14 actions. Since this protocol is published in a special issue editorial, its presentation and elaboration are somewhat limited by the already mentioned word count limit. Nevertheless, the SPAR-4-SLR protocol provides a very useful checklist for researchers that enables them to include all data required to document the SLR and to avoid confusion from editors, reviewers, and readers regarding SLR characteristics.

Beyond Table 1 , Durach et al. ( 2017 ) synthesized six common SLR “steps” that differ only marginally in the delimitation of one step to another from the sub-stages of the previously mentioned SLR processes. In addition, Snyder ( 2019 ) proposed a process comprising four “phases” that take more of a bird’s perspective in addressing (1) design, (2) conduct, (3) analysis, and (4) structuring and writing the review. Moreover, Xiao and Watson ( 2019 ) proposed only three “stages” of (1) planning, (2) conducting, and (3) reporting the review that combines the previously mentioned conduct and the analysis and defines eight steps within them. Much in line with the other process models, the final reporting stage contains only one of the eight steps, leaving the reader somewhat alone in how to effectively craft a manuscript that contributes to the further development of the field.

In effect, the mentioned SLR processes differ only marginally, while the systematic nature of actions in the SPAR-4-SLR protocol (Paul et al. 2021 ) can be seen as a reporting must-have within any of the mentioned SLR processes. The similarity of the SLR processes is, however, also evident in the fact that they leave open how the SLR analysis can be executed, enriched, and reflected to make a contribution to the reviewed field. In contrast, this aspect is richly described in the other guidelines that do not offer an SLR process, leading us again toward the tension for guideline papers between elaborating on a) the SLR process and b) the details, options, and potentials of each process step.

To help (prospective) SLR authors successfully navigate this tension of existing guidelines, it is thus the ambition of this paper to adopt a comprehensive SLR process model along which an SLR project can be planned, executed, and written up in a coherent way. To enable this coherence, 14 distinct decisions are defined, reflected, and interlinked, which have to be taken across the different steps of the SLR process. At the same time, our process model aims to actively direct researchers to the best practices, tips, and guidance that previous guidelines have provided for individual decisions. We aim to achieve this by means of an integrative review of the relevant SLR guidelines, as outlined in the following section.

3 Methodology: an integrative literature review of guidelines for systematic literature reviews in management

It might seem intuitive to contribute to the debate on the “gold standard” of systematic literature reviews (Davis et al. 2014 ) by conducting a systematic review ourselves. However, there are different types of reviews aiming for distinctive contributions. Snyder ( 2019 ) distinguished between a) systematic, b) semi-systematic, and c) integrative (or critical) reviews, which aim for i) (mostly quantitative) synthesis and comparison of prior (primary) evidence, ii) an overview of the development of a field over time, and iii) a critique and synthesis of prior perspectives to reconceptualize or advance them. Each review team needs to position itself in such a typology of reviews to define the aims and scope of the review. To do so and structure the related research process, we adopted the four generic steps for an (integrative) literature review by Snyder ( 2019 )—(1) design, (2) conduct, (3) analysis, and (4) structuring and writing the review—on which we report in the remainder of this section. Since the last step is a very practical one that, for example, asks, “Is the contribution of the review clearly communicated?” (Snyder 2019 ), we will focus on the presentation of the method applied to the initial three steps:

(1) Regarding the design, we see the need for this study emerging from our experience in reviewing SLR manuscripts, supervising PhD students who, almost by default, need to prepare an SLR, and recurring discussions on certain decisions in the process of both. These discussions regularly left some blank or blurry spaces (see Table 1 ) that induced substantial uncertainty regarding critical decisions in the SLR process (Paul et al. 2021 ). To address this gap, we aim to synthesize prior guidance and critically enrich it, thus adopting an integrative approach for reviewing existing SLR guidance in the management domain (Snyder 2019 ).

(2) To conduct the review, we started collecting the literature that provided guidance on the individual SLR parts. We built on a sample of 13 regularly cited or very recent papers in the management domain. We started with core articles that we successfully used to publish SLRs in top-tier OSCM journals, such as Tranfield et al. ( 2003 ) and Durach et al. ( 2017 ), and we checked their references and papers that cited these publications. The search focus was defined by the following criteria: the articles needed to a) provide original methodological guidance for SLRs by providing new aspects of the guideline or synthesizing existing ones into more valid guidelines and b) focus on the management domain. Building on the nature of a critical or integrative review that does not require a full or representative sample (Snyder 2019 ), we limited the sample to the papers displayed in Table 2 that built the core of the currently applied SLR guidelines. In effect, we found 11 technical papers and two SLRs of SLRs (Carter and Washispack 2018 ; Seuring and Gold 2012 ). From the latter, we mainly analyzed the discussion and conclusion parts that explicitly developed guidance on conducting SLRs.

(3) For analyzing these papers, we first adopted the six-step SLR process proposed by Durach et al. ( 2017 , p.70), which they define as applicable to any “field, discipline or philosophical perspective”. The contrast between the six-step SLR process used for the analysis and the four-step process applied by ourselves may seem surprising but is justified by the use of an integrative approach. This approach differs mainly in retrieving and selecting pertinent literature that is key to SLRs and thus needs to be part of the analysis framework.

While deductively coding the sample papers against Durach et al.’s ( 2017 ), guidance in the six steps, we inductively built a set of 14 decisions presented in the right columns of Table 2 that are required to be made in any SLR. These decisions built a second and more detailed level of analysis, for which the single guidelines were coded as giving low, medium, or high levels of detail (see Table 3 ), which helped us identify the gaps in the current guidance papers and led our way in presenting, critically discussing, and enriching the literature. In effect, we see that almost all guidelines touch on the same issues and try to give a comprehensive overview. However, this results in multiple guidelines that all lack the space to go into detail, while only a few guidelines focus on filling a gap in the process. It is our ambition with this analysis to identify the gaps in the guidelines, thereby identifying a precise need for refinement, and to offer a first step into this refinement. Adopting advice from the literature sample, the coding was conducted by the entire author team (Snyder 2019 ; Tranfield et al. 2003 ) including discursive alignments of interpretation (Seuring and Gold 2012 ). This enabled a certain reliability and validity of the analysis by reducing the within-study and expectancy bias (Durach et al. 2017 ), while the replicability was supported by reporting the review sample and the coding results in Table 3 (Carter and Washispack 2018 ).

(4) For the writing of the review, we only pointed to the unusual structure of presenting the method without a theory section and then the findings in the following section. However, this was motivated by the nature of the integrative review so that the review findings at the same time represent the “state of the art,” “literature review,” or “conceptualization” sections of a paper.

4 Findings of the integrative review: presentation, critical discussion, and enrichment of prior guidance

4.1 the overall research process for a systematic literature review.

Even within our sample of only 13 guidelines, there are four distinct suggestions for structuring the SLR process. One of the earliest SLR process models was proposed by Tranfield et al. ( 2003 ) encompassing the three stages of (1) planning the review, (2) conducting a review, and (3) reporting and dissemination. Snyder ( 2019 ) proposed four steps employed in this study: (1) design, (2) conduct, (3) analysis, and (4) structuring and writing the review. Borrowing from content analysis guidelines, Seuring and Gold ( 2012 ) defined four steps: (1) material collection, (2) descriptive analysis, (3) category selection, and (4) material evaluation. Most recently Kraus et al. ( 2020 ) proposed four steps: (1) planning the review, (2) identifying and evaluating studies, (3) extracting and synthesizing data, and (4) disseminating the review findings. Most comprehensively, Durach et al. ( 2017 ) condensed prior process models into their generic six steps for an SLR. Adding the review of the process models by Snyder ( 2019 ) and Seuring and Gold ( 2012 ) to Durach et al.’s ( 2017 ) SLR process review of four papers, we support their conclusion of the general applicability of the six steps defined. Consequently, these six steps form the backbone of our coding scheme, as shown in the left column of Table 2 and described in the middle column.

As stated in Sect.  3 , we synthesized the review papers against these six steps but experienced that the papers were taking substantially different foci by providing rich details for some steps while largely bypassing others. To capture this heterogeneity and better operationalize the SLR process, we inductively introduced the right column, identifying 14 decisions to be made. These decisions are all elaborated in the reviewed papers but to substantially different extents, as the detailed coding results in Table 3 underline.

Mapping Table 3 for potential gaps in the existing guidelines, we found six decisions on which we found only low- to medium-level details, while high-detail elaboration was missing. These six decisions, which are illustrated in Fig.  1 , belong to three steps: 1: defining the research question, 5: synthesizing the literature, and 6: reporting the results. This result underscores our critique of currently unbalanced guidance that is, on the one hand, detailed on determining the required characteristics of primary studies (step 2), retrieving a sample of potentially relevant literature (step 3), and selecting the pertinent literature (step 4). On the other hand, authors, especially PhD students, are left without substantial guidance on the steps critical to publication. Instead, they are called “to go one step further … and derive meaningful conclusions” (Fisch and Block 2018 , p. 105) without further operationalizations on how this can be achieved; for example, how “meet the editor” conference sessions regularly cause frustration among PhDs when editors call for “new,” “bold,” and “relevant” research. Filling the gaps in the six decisions with best practice examples and practical experience is the main focus of this study’s contribution. The other eight decisions are synthesized with references to the guidelines that are most helpful and relevant for the respective step in our eyes.

The 6 steps and 14 decisions of the SLR process

4.2 Step 1: defining the research question

When initiating a research project, researchers make three key decisions.

Decision 1 considers the essential tasks of establishing a relevant and timely research question, but despite the importance of the decision, which determines large parts of further decisions (Snyder 2019 ; Tranfield et al. 2003 ), we only find scattered guidance in the literature. Hence, how can a research topic be specified to allow a strong literature review that is neither too narrow nor too broad? The latter is the danger in meta-reviews (i.e., reviews of reviews) (Aguinis et al. 2020 ; Carter and Washispack 2018 ; Kache and Seuring 2014 ). In this respect, even though the method would be robust, the findings would not be novel. In line with Carter and Washispack ( 2018 ), there should always be room for new reviews, yet over time, they must move from a descriptive overview of a field further into depth and provide detailed analyses of constructs. Clark et al. ( 2021 ) provided a detailed but very specific reflection on how they crafted a research question for an SLR and that revisiting the research question multiple times throughout the SLR process helps to coherently and efficiently move forward with the research. More generically, Kraus et al. ( 2020 ) listed six key contributions of an SLR that can guide the definition of the research question. Finally, Snyder ( 2019 ) suggested moving into more detail from existing SLRs and specified two main avenues for crafting an SLR research question that are either investigating the relationship among multiple effects, the effect of (a) specific variable(s), or mapping the evidence regarding a certain research area. For the latter, we see three possible alternative approaches, starting with a focus on certain industries. Examples are analyses of the food industry (Beske et al. 2014 ), retailing (Wiese et al. 2012 ), mining and minerals (Sauer and Seuring 2017 ), or perishable product supply chains (Lusiantoro et al. 2018 ) and traceability at the example of the apparel industry (Garcia-Torres et al. 2019 ). A second opportunity would be to assess the status of research in a geographical area that composes an interesting context from a research perspective, such as sustainable supply chain management (SSCM) in Latin America (Fritz and Silva 2018 ), yet this has to be justified explicitly, avoiding the fact that geographical focus is taken as the reason per se (e.g., Crane et al. 2016 ). A third variant addresses emerging issues, such as SCM, in a base-of-the-pyramid setting (Khalid and Seuring 2019 ) and the use of blockchain technology (Wang et al. 2019 ) or digital transformation (Hanelt et al. 2021 ). These approaches limit the reviewed field to enable a more contextualized analysis in which the novelty, continued relevance, or unjustified underrepresentation of the context can be used to specify a research gap and related research question(s). This also impacts the following decisions, as shown below.

Decision 2 concerns the option for a theoretical approach (i.e., the adoption of an inductive, abductive, or deductive approach) to theory building through the literature review. The review of previous guidance on this delivers an interesting observation. On the one hand, there are early elaborations on systematic reviews, realist synthesis, meta-synthesis, and meta-analysis by Tranfield et al. ( 2003 ) that are borrowing from the origins of systematic reviews in medical research. On the other hand, recent management-related guidelines largely neglect details of related decisions, but point out that SLRs are a suitable tool for theory building (Kraus et al. 2020 ). Seuring et al. ( 2021 ) set out to fill this gap and provided substantial guidance on how to use theory in SLRs to advance the field. To date, the option for a theoretical approach is only rarely made explicit, leaving the reader often puzzled about how advancement in theory has been crafted and impeding a review’s replicability (Seuring et al. 2021 ). Many papers still leave related choices in the dark (e.g., Rhaiem and Amara 2021 ; Rojas-Córdova et al. 2022 ) and move directly from the introduction to the method section.

In Decision 3, researchers need to adopt a theoretical framework (Durach et al. 2017 ) or at least a theoretical starting point, depending on the most appropriate theoretical approach (Seuring et al. 2021 ). Here, we find substantial guidance by Durach et al. ( 2017 ) that underlines the value of adopting a theoretical lens to investigate SCM phenomena and the literature. Moreover, the choice of a theoretical anchor enables a consistent definition and operationalization of constructs that are used to analyze the reviewed literature (Durach et al. 2017 ; Seuring et al. 2021 ). Hence, providing some upfront definitions is beneficial, clarifying what key terminology would be used in the subsequent paper, such as Devece et al. ( 2019 ) introduce their terminology on coopetition. Adding a practical hint beyond the elaborations of prior guidance papers for taking up established constructs in a deductive analysis (decision 2), there would be the question of whether these can yield interesting findings.

Here, it would be relevant to specify what kind of analysis is aimed for the SLR, where three approaches might be distinguished (i.e., bibliometric analysis, meta-analysis, and content analysis–based studies). Briefly distinguishing them, the core difference would be how many papers can be analyzed employing the respective method. Bibliometric analysis (Donthu et al. 2021 ) usually relies on the use of software, such as Biblioshiny, allowing the creation of figures on citations and co-citations. These figures enable the interpretation of large datasets in which several hundred papers can be analyzed in an automated manner. This allows for distinguishing among different research clusters, thereby following a more inductive approach. This would be contrasted by meta-analysis (e.g., Leuschner et al. 2013 ), where often a comparatively smaller number of papers is analyzed (86 in the respective case) but with a high number of observations (more than 17,000). The aim is to test for statistically significant correlations among single constructs, which requires that the related constructs and items be precisely defined (i.e., a clearly deductive approach to the analysis).

Content analysis is the third instrument frequently applied to data analysis, where an inductive or deductive approach might be taken (Seuring et al. 2021 ). Content-based analysis (see decision 9 in Sect.  4.6 ; Seuring and Gold 2012 ) is a labor-intensive step and can hardly be changed ex post. This also implies that only a certain number of papers might be analyzed (see Decision 6 in Sect.  4.5 ). It is advisable to adopt a wider set of constructs for the analysis stemming even from multiple established frameworks since it is difficult to predict which constructs and items might yield interesting insights. Hence, coding a more comprehensive set of items and dropping some in the process is less problematic than starting an analysis all over again for additional constructs and items. However, in the process of content analysis, such an iterative process might be required to improve the meaningfulness of the data and findings (Seuring and Gold 2012 ). A recent example of such an approach can be found in Khalid and Seuring ( 2019 ), building on the conceptual frameworks for SSCM of Carter and Rogers ( 2008 ), Seuring and Müller ( 2008 ), and Pagell and Wu ( 2009 ). This allows for an in-depth analysis of how SSCM constructs are inherently referred to in base-of-the-pyramid-related research. The core criticism and limitation of such an approach is the random and subjectively biased selection of frameworks for the purpose of analysis.

Beyond the aforementioned SLR methods, some reviews, similar to the one used here, apply a critical review approach. This is, however, nonsystematic, and not an SLR; thus, it is beyond the scope of this paper. Interested readers can nevertheless find some guidance on critical reviews in the available literature (e.g., Kraus et al. 2022 ; Snyder 2019 ).

4.3 Step 2: determining the required characteristics of primary studies

After setting the stage for the review, it is essential to determine which literature is to be reviewed in Decision 4. This topic is discussed by almost all existing guidelines and will thus only briefly be discussed here. Durach et al. ( 2017 ) elaborated in great detail on defining strict inclusion and exclusion criteria that need to be aligned with the chosen theoretical framework. The relevant units of analysis need to be specified (often a single paper, but other approaches might be possible) along with suitable research methods, particularly if exclusively empirical studies are reviewed or if other methods are applied. Beyond that, they elaborated on potential quality criteria that should be applied. The same is considered by a number of guidelines that especially draw on medical research, in which systematic reviews aim to pool prior studies to infer findings from their total population. Here, it is essential to ensure the exclusion of poor-quality evidence that would lower the quality of the review findings (Mulrow 1987 ; Tranfield et al. 2003 ). This could be ensured by, for example, only taking papers from journals listed on the Web of Science or Scopus or journals listed in quartile 1 of Scimago ( https://www.scimagojr.com/ ), a database providing citation and reference data for journals.

The selection of relevant publication years should again follow the purpose of the study defined in Step 1. As such, there might be a justified interest in the wide coverage of publication years if a historical perspective is taken. Alternatively, more contemporary developments or the analysis of very recent issues can justify the selection of very few years of publication (e.g., Kraus et al. 2022 ). Again, it is hard to specify a certain time period covered, but if developments of a field should be analyzed, a five-year period might be a typical lower threshold. On current topics, there is often a trend of rising publishing numbers. This scenario implies the rising relevance of a topic; however, this should be treated with caution. The total number of papers published per annum has increased substantially in recent years, which might account for the recently heightened number of papers on a certain topic.

4.4 Step 3: retrieving a sample of potentially relevant literature

After defining the required characteristics of the literature to be reviewed, the literature needs to be retrieved based on two decisions. Decision 5 concerns suitable literature sources and databases that need to be defined. Turning to Web of Science or Scopus would be two typical options found in many of the examples mentioned already (see also detailed guidance by Paul and Criado ( 2020 ) as well as Paul et al. ( 2021 )). These databases aggregate many management journals, and a typical argument for turning to the Web of Science database is the inclusion of impact factors, as they indicate a certain minimum quality of the journal (Sauer and Seuring 2017 ). Additionally, Google Scholar is increasingly mentioned as a usable search engine, often providing higher numbers of search results than the mentioned databases (e.g., Pearce 2018 ). These results often entail duplicates of articles from multiple sources or versions of the same article, as well as articles in predatory journals (Paul et al. 2021 ). Therefore, we concur with Paul et al. ( 2021 ) who underline the quality assurance mechanisms in Web of Science and Scopus, making them preferred databases for the literature search. From a practical perspective, it needs to be mentioned that SLRs in management mainly rely on databases that are not free to use. Against this limitation, Pearce ( 2018 ) provided a list of 20 search engines that are free of charge and elaborated on their advantages and disadvantages. Due to the individual limitations of the databases, it is advisable to use a combination of them (Kraus et al. 2020 , 2022 ) and build a consolidated sample by screening the papers found for duplicates, as regularly done in SLRs.

This decision also includes the choice of the types of literature to be analyzed. Typically, journal papers are selected, ensuring that the collected papers are peer-reviewed and have thus undergone an academic quality management process. Meanwhile, conference papers are usually avoided since they are often less mature and not checked for quality (e.g., Seuring et al. 2021 ). Nevertheless, for emerging topics, it might be too restrictive to consider only peer-reviewed journal articles and limit the literature to only a few references. Analyzing such rapidly emerging topics is relevant for timely and impact-oriented research and might justify the selection of different sources. Kraus et al. ( 2020 ) provided a discussion on the use of gray literature (i.e., nonacademic sources), and Sauer ( 2021 ) provided an example of a review of sustainability standards from a management perspective to derive implications for their application by managers on the one hand and for enhancing their applicability on the other hand.

Another popular way to limit the review sample is the restriction to a certain list of journals (Kraus et al. 2020 ; Snyder 2019 ). While this is sometimes favored by highly ranked journals, Carter and Washispack ( 2018 ), for example, found that many pertinent papers are not necessarily published in journals within the field. Webster and Watson ( 2002 ) quite tellingly cited a reviewer labeling the selection of top journals as an unjustified excuse for investigating the full body of relevant literature. Both aforementioned guidelines thus discourage the restriction to particular journals, a guidance that we fully support.

However, there is an argument to be made supporting the exclusion of certain lower-ranked journals. This can be done, for example, by using Scimago Journal quartiles ( www.scimagojr.com , last accessed 13. of April 2023) and restricting it to journals in the first quartile (e.g., Yavaprabhas et al. 2022 ). Other papers (e.g., Kraus et al. 2021 ; Rojas-Córdova et al. 2022 ) use certain journal quality lists to limit their sample. However, we argue for a careful check by the authors against the topic reviewed regarding what would be included and excluded.

Decision 6 entails the definition of search terms and a search string to be applied in the database just chosen. The search terms should reflect the aims of the review and the exclusion criteria that might be derived from the unit of analysis and the theoretical framework (Durach et al. 2017 ; Snyder 2019 ). Overall, two approaches to keywords can be observed. First, some guides suggest using synonyms of the key terms of interest (e.g., Durach et al. 2017 ; Kraus et al. 2020 ) in order to build a wide baseline sample that will be condensed in the next step. This is, of course, especially helpful if multiple terms delimitate a field together or different synonymous terms are used in parallel in different fields or journals. Empirical journals in supply chain management, for example, use the term “multiple supplier tiers ” (e.g., Tachizawa and Wong 2014 ), while modeling journals in the same field label this as “multiple supplier echelons ” (e.g., Brandenburg and Rebs 2015 ). Second, in some cases, single keywords are appropriate for capturing a central aspect or construct of a field if the single keyword has a global meaning tying this field together. This approach is especially relevant to the study of relatively broad terms, such as “social media” (Lim and Rasul 2022 ). However, this might result in very high numbers of publications found and therefore requires a purposeful combination with other search criteria, such as specific journals (Kraus et al. 2021 ; Lim et al. 2021 ), publication dates, article types, research methods, or the combination with keywords covering domains to which the search is aimed to be specified.

Since SLRs are often required to move into detail or review the intersections of relevant fields, we recommend building groups of keywords (single terms or multiple synonyms) for each field to be connected that are coupled via Boolean operators. To determine when a point of saturation for a keyword group is reached, one could monitor the increase in papers found in a database when adding another synonym. Once the increase is significantly decreasing or even zeroing, saturation is reached (Sauer and Seuring 2017 ). The keywords themselves can be derived from the list of keywords of influential publications in the field, while attention should be paid to potential synonyms in neighboring fields (Carter and Washispack 2018 ; Durach et al. 2017 ; Kraus et al. 2020 ).

4.5 Step 4: selecting the pertinent literature

The inclusion and exclusion criteria (Decision 6) are typically applied in Decision 7 in a two-stage process, first on the title, abstract, and keywords of an article before secondly applying them to the full text of the remaining articles (see also Kraus et al. 2020 ; Snyder 2019 ). Beyond this, Durach et al. ( 2017 ) underlined that the pertinence of the publication regarding units of analysis and the theoretical framework needs to be critically evaluated in this step to avoid bias in the review analysis. Moreover, Carter and Washispack ( 2018 ) requested the publication of the included and excluded sources to ensure the replicability of Steps 3 and 4. This can easily be done as an online supplement to an eventually published review article.

Nevertheless, the question remains: How many papers justify a literature review? While it is hard to specify how many papers comprise a body of literature, there might be certain thresholds for which Kraus et al. ( 2020 ) provide a useful discussion. As a rough guide, more than 50 papers would usually make a sound starting point (see also Paul and Criado 2020 ), while there are SLRs on emergent topics, such as multitier supply chain management, where 39 studies were included (Tachizawa and Wong 2014 ). An SLR on “learning from innovation failures” builds on 36 papers (Rhaiem and Amara 2021 ), which we would see as the lower threshold. However, such a low number should be an exception, and anything lower would certainly trigger the following question: Why is a review needed? Meanwhile, there are also limits on how many papers should be reviewed. While there are cases with 191 (Seuring and Müller 2008 ), 235 (Rojas-Córdova et al. 2022 ), or up to nearly 400 papers reviewed (Spens and Kovács 2006 ), these can be regarded as upper thresholds. Over time, similar topics seem to address larger datasets.

4.6 Step 5: synthesizing the literature

Before synthesizing the literature, Decision 8 considers the selection of a data extraction tool for which we found surprisingly little guidance. Some guidance is given on the use of cloud storage to enable remote team work (Clark et al. 2021 ). Beyond this, we found that SLRs have often been compiled with marked and commented PDFs or printed papers that were accompanied by tables (Kraus et al. 2020 ) or Excel sheets (see also the process tips by Clark et al. 2021 ). This sheet tabulated the single codes derived from the theoretical framework (Decision 3) and the single papers to be reviewed (Decision 7) by crossing out individual cells, signaling the representation of a particular code in a particular paper. While the frequency distribution of the codes is easily compiled from this data tool, the related content needs to be looked at in the papers in a tedious back-and-forth process. Beyond that, we would strongly recommend using data analysis software, such as MAXQDA or NVivo. Such programs enable the import of literature in PDF format and the automatic or manual coding of text passages, their comparison, and tabulation. Moreover, there is a permanent and editable reference of the coded text to a code. This enables a very quick compilation of content summaries or statistics for single codes and the identification of qualitative and quantitative links between codes and papers.

All the mentioned data extraction or data processing tools require a license and therefore are not free of cost. While many researchers may benefit from national or institutional subscriptions to these services, others may not. As a potential alternative, Pearce ( 2018 ) proposed a set of free open-source software (FOSS), including an elaboration on how they can be combined to perform an SLR. He also highlighted that both free and proprietary solutions have advantages and disadvantages that are worthwhile for those who do not have the required tools provided by their employers or other institutions they are members of. The same may apply to the literature databases used for the literature acquisition in Decision 5 (Pearce 2018 ).

Moreover, there is a link to Step 1, Decision 3, where bibliometric reviews and meta-analyses were mentioned. These methods, which are alternatives to content analysis–based approaches, have specific demands, so specific tools would be appropriate, such as the Biblioshiny software or VOSviewer. As we will point out for all decisions, there is a high degree of interdependence among the steps and decisions made.

Decision 9 looks at conducting the data analysis, such as coding against (pre-defined) constructs, in SLRs that rely, in most cases, on content analysis. Seuring and Gold ( 2012 ) elaborated in detail on its characteristics and application in SLRs. As this paper also explains the process of qualitative content analysis in detail, repetition is avoided here, but a summary is offered. Since different ways exist to conduct a content analysis, it is even more important to explain and justify, for example, the choice of an inductive or deductive approach (see Decision 2). In several cases, analytic variables are applied on the go, so there is no theory-based introduction of related constructs. However, to ensure the validity and replicability of the review (see Decision 11), it is necessary to explicitly define all the variables and codes used to analyze and synthesize the reviewed material (Durach et al. 2017 ; Seuring and Gold 2012 ). To build a valid framework as the SLR outcome, it is vital to ensure that the constructs used for the data analysis are sufficiently defined, mutually exclusive, and comprehensively exhaustive. For meta-analysis, the predefined constructs and items would demand quantitative coding so that the resulting data could be analyzed using statistical software tools such as SPSS or R (e.g., Xiao and Watson 2019 ). Pointing to bibliometric analysis again, the respective software would be used for data analysis, yielding different figures and paper clusters, which would then require interpretation (e.g., Donthu et al. 2021 ; Xiao and Watson 2019 ).

Decision 10, on conducting subsequent statistical analysis, considers follow-up analysis of the coding results. Again, this is linked to the chosen SLR method, and a bibliographic analysis will require a different statistical analysis than a content analysis–based SLR (e.g., Lim et al. 2022 ; Xiao and Watson 2019 ). Beyond the use of content analysis and the qualitative interpretation of its results, applying contingency analysis offers the opportunity to quantitatively assess the links among constructs and items. It provides insights into which items are correlated with each other without implying causality. Thus, the interpretation of the findings must explain the causality behind the correlations between the constructs and the items. This must be based on sound reasoning and linking the findings to theoretical arguments. For SLRs, there have recently been two kinds of applications of contingency analysis, differentiated by unit of analysis. De Lima et al. ( 2021 ) used the entire paper as the unit of analysis, deriving correlations on two constructs that were used together in one paper. This is, of course, subject to critique as to whether the constructs really represent correlated content. Moving a level deeper, Tröster and Hiete ( 2018 ) used single-text passages on one aspect, argument, or thought as the unit of analysis. Such an approach is immune against the critique raised before and can yield more valid statistical support for thematic analysis. Another recent methodological contribution employing the same contingency analysis–based approach was made by Siems et al. ( 2021 ). Their analysis employs constructs from SSCM and dynamic capabilities. Employing four subsets of data (i.e., two time periods each in the food and automotive industries), they showed that the method allows distinguishing among time frames as well as among industries.

However, the unit of analysis must be precisely explained so that the reader can comprehend it. Both examples use contingency analysis to identify under-researched topics and develop them into research directions whose formulation represents the particular aim of an SLR (Paul and Criado 2020 ; Snyder 2019 ). Other statistical tools might also be applied, such as cluster analysis. Interestingly, Brandenburg and Rebs ( 2015 ) applied both contingency and cluster analyses. However, the authors stated that the contingency analysis did not yield usable results, so they opted for cluster analysis. In effect, Brandenburg and Rebs ( 2015 ) added analytical depth to their analysis of model types in SSCM by clustering them against the main analytical categories of content analysis. In any case, the application of statistical tools needs to fit the study purpose (Decision 1) and the literature sample (Decision 7), just as in their more conventional applications (e.g., in empirical research processes).

Decision 11 regards the additional consideration of validity and reliability criteria and emphasizes the need for explaining and justifying the single steps of the research process (Seuring and Gold 2012 ), much in line with other examples of research (Davis and Crombie 2001 ). This is critical to underlining the quality of the review but is often neglected in many submitted manuscripts. In our review, we find rich guidance on this decision, to which we want to guide readers (see Table 3 ). In particular, Durach et al. ( 2017 ) provide an entire section of biases and what needs to be considered and reported on them. Moreover, Snyder ( 2019 ) regularly reflects on these issues in her elaborations. This rich guidance elaborates on how to ensure the quality of the individual steps of the review process, such as sampling, study inclusion and exclusion, coding, synthesizing, and more practical issues, including team composition and teamwork organization, which are discussed in some guidelines (e.g., Clark et al. 2021 ; Kraus et al. 2020 ). We only want to underline that the potential biases are, of course, to be seen in conjunction with Decisions 2, 3, 4, 5, 6, 7, 9, and 10. These decisions and the elaboration by Durach et al. ( 2017 ) should provide ample points of reflection that, however, many SLR manuscripts fail to address.

4.7 Step 6: reporting the results

In the final step, there are three decisions on which there is surprisingly little guidance, although reviews often fail in this critical part of the process (Kraus et al. 2020 ). The reviewed guidelines discuss the presentation almost exclusively, while almost no guidance is given on the overall paper structure or the key content to be reported.

Consequently, the first choice to be made in Decision 12 is regarding the paper structure. We suggest following the five-step logic of typical research papers (see also Fisch and Block 2018 ) and explaining only a few points in which a difference from other papers is seen.

(1) Introduction: While the introduction would follow a conventional logic of problem statement, research question, contribution, and outline of the paper (see also Webster and Watson 2002 ), the next parts might depend on the theoretical choices made in Decision 2.

(2) Literature review section: If deductive logic is taken, the paper usually has a conventional flow. After the introduction, the literature review section covers the theoretical background and the choice of constructs and variables for the analysis (De Lima et al. 2021 ; Dieste et al. 2022 ). To avoid confusion in this section with the literature review, its labeling can also be closer to the reviewed object.

If an inductive approach is applied, it might be challenging to present the theoretical basis up front, as the codes emerge only from analyzing the material. In this case, the theory section might be rather short, concentrating on defining the core concepts or terms used, for example, in the keyword-based search for papers. The latter approach is exemplified by the study at hand, which presents a short review of the available literature in the introduction and the first part of the findings. However, we do not perform a systematic but integrative review, which allows for more freedom and creativity (Snyder 2019 ).

(3) Method section: This section should cover the steps and follow the logic presented in this paper or any of the reviewed guidelines so that the choices made during the research process are transparently disclosed (Denyer and Tranfield 2009 ; Paul et al. 2021 ; Xiao and Watson 2019 ). In particular, the search for papers and their selection requires a sound explanation of each step taken, including the provision of reasons for the delimitation of the final paper sample. A stage that is often not covered in sufficient detail is data analysis (Seuring and Gold 2012 ). This also needs to be outlined so that the reader can comprehend how sense has been made of the material collected. Overall, the demands for SLR papers are similar to case studies, survey papers, or almost any piece of empirical research; thus, each step of the research process needs to be comprehensively described, including Decisions 4–10. This comprehensiveness must also include addressing measures for validity and reliability (see Decision 11) or other suitable measures of rigor in the research process since they are a critical issue in literature reviews (Durach et al. 2017 ). In particular, inductively conducted reviews are prone to subjective influences and thus require sound reporting of design choices and their justification.

(4) Findings: The findings typically start with a descriptive analysis of the literature covered, such as journals, distribution across years, or (empirical) methods applied (Tranfield et al. 2003 ). For modeling-related reviews, classifying papers against the approach chosen is a standard approach, but this can often also serve as an analytic category that provides detailed insights. The descriptive analysis should be kept short since a paper only presenting descriptive findings will not be of great interest to other researchers due to the missing contribution (Snyder 2019 ). Nevertheless, there are opportunities to provide interesting findings in the descriptive analysis. Beyond a mere description of the distributions of the single results, such as the distribution of methods used in the sample, authors should combine analytical categories to derive more detailed insights (see also Tranfield et al. 2003 ). The distribution of methods used might well be combined with the years of publication to identify and characterize different phases in the development of a field of research or its maturity. Moreover, there could be value in the analysis of theories applied in the review sample (e.g., Touboulic and Walker 2015 ; Zhu et al. 2022 ) and in reflecting on the interplay of different qualitative and quantitative methods in spurring the theoretical development of the reviewed field. This could yield detailed insights into methodological as well as theoretical gaps, and we would suggest explicitly linking the findings of such analyses to the research directions that an SLR typically provides. This link could help make the research directions much more tangible by giving researchers a clear indication of how to follow up on the findings, as, for example, done by Maestrini et al. ( 2017 ) or Dieste et al. ( 2022 ). In contrast to the mentioned examples of an actionable research agenda, a typical weakness of premature SLR manuscripts is that they ask rather superficially for more research in the different aspects they reviewed but remain silent about how exactly this can be achieved.

We would thus like to encourage future SLR authors to systematically investigate the potential to combine two categories of descriptive analysis to move this section of the findings to a higher level of quality, interest, and relevance. The same can, of course, be done with the thematic findings, which comprise the second part of this section.

Moving into the thematic analysis, we have already reached Decision 13 on the presentation of the refined theoretical framework and the discussion of its contents. A first step might present the frequencies of the codes or constructs applied in the analysis. This allows the reader to understand which topics are relevant. If a rather small body of literature is analyzed, tables providing evidence on which paper has been coded for which construct might be helpful in improving the transparency of the research process. Tables or other forms of visualization might help to organize the many codes soundly (see also Durach et al. 2017 ; Paul and Criado 2020 ; Webster and Watson 2002 ). These findings might then lead to interpretation, for which it is necessary to extract meaning from the body of literature and present it accordingly (Snyder 2019 ). To do so, it might seem needless to say that the researchers should refer back to Decisions 1, 2, and 3 taken in Step 1 and their justifications. These typically identify the research gap to be filled, but after the lengthy process of the SLR, the authors often fail to step back from the coding results and put them into a larger perspective against the research gap defined in Decision 1 (see also Clark et al. 2021 ). To support this, it is certainly helpful to illustrate the findings in a figure or graph presenting the links among the constructs and items and adding causal reasoning to this (Durach et al. 2017 ; Paul and Criado 2020 ), such as the three figures by Seuring and Müller ( 2008 ) or other examples by De Lima et al. ( 2021 ) or Tipu ( 2022 ). This presentation should condense arguments made in the assessed literature but should also chart the course for future research. It will be these parts of the paper that are decisive for a strong SLR paper.

Moreover, some guidelines define the most fruitful way of synthesizing the findings as concept-centric synthesis (Clark et al. 2021 ; Fisch and Block 2018 ; Webster and Watson 2002 ). As presented in the previous sentence, the presentation of the review findings is centered on the content or concept of “concept-centric synthesis.” It is accompanied by a reference to all or the most relevant literature in which the concept is evident. Contrastingly, Webster and Watson ( 2002 ) found that author-centric synthesis discusses individual papers and what they have done and found (just like this sentence here). They added that this approach fails to synthesize larger samples. We want to note that we used the latter approach in some places in this paper. However, this aims to actively refer the reader to these studies, as they stand out from our relatively small sample. Beyond this, we want to link back to Decision 3, the selection of a theoretical framework and constructs. These constructs, or the parts of a framework, can also serve to structure the findings section by using them as headlines for subsections (Seuring et al. 2021 ).

Last but not least, there might even be cases where core findings and relationships might be opposed, and alternative perspectives could be presented. This would certainly be challenging to argue for but worthwhile to do in order to drive the reviewed field forward. A related example is the paper by Zhu et al. ( 2022 ), who challenged the current debate at the intersection of blockchain applications and supply chain management and pointed to the limited use of theoretical foundations for related analysis.

(5) Discussion and Conclusion: The discussion needs to explain the contribution the paper makes to the extant literature, that is, which previous findings or hypotheses are supported or contradicted and which aspects of the findings are particularly interesting for the future development of the reviewed field. This is in line with the content required in the discussion sections of any other paper type. A typical structure might point to the contribution and put it into perspective with already existing research. Further, limitations should be addressed on both the theoretical and methodological sides. This elaboration of the limitations can be coupled with the considerations of the validity and reliability of the study in Decision 11. The implications for future research are a core aim of an SLR (Clark et al. 2021 ; Mulrow 1987 ; Snyder 2019 ) and should be addressed in a further part of the discussion section. Recently, a growing number of literature reviews have also provided research questions for future research that provide a very concrete and actionable output of the SLR (e.g. Dieste et al. 2022 ; Maestrini et al. 2017 ). Moreover, we would like to reiterate our call to clearly link the research implications to the SLR findings, which helps the authors craft more tangible research directions and helps the reader to follow the authors’ interpretation. Literature review papers are usually not strongly positioned toward managerial implications, but even these implications might be included.

As a kind of normal demand, the conclusion should provide an answer to the research question put forward in the introduction, thereby closing the cycle of arguments made in the paper.

Although all the works seem to be done when the paper is written and the contribution is fleshed out, there is still one major decision to be made. Decision 14 concerns the identification of an appropriate journal for submission. Despite the popularity of the SLR method, a rising number of journals explicitly limit the number of SLRs published by them. Moreover, there are only two guidelines elaborating on this decision, underlining the need for the following considerations.

Although it might seem most attractive to submit the paper to the highest-ranking journal for the reviewed topic, we argue for two critical and review-related decisions to be made during the research process that influence whether the paper fits a certain outlet:

The theoretical foundation of the SLR (Decision 3) usually relates to certain journals in which it is published or discussed. If a deductive approach was taken, the journals in which the foundational papers were published might be suitable since the review potentially contributes to the further validation or refinement of the frameworks. Overall, we need to keep in mind that a paper needs to be added to a discussion in the journal, and this can be based on the theoretical framework or the reviewed papers, as shown below.

Appropriate journals for publication can be derived from the analyzed journal papers (Decision 7) (see also Paul and Criado 2020 ). This allows for an easy link to the theoretical debate in the respective journal by submitting it. This choice is identifiable in most of the papers mentioned in this paper and is often illustrated in the descriptive analysis.

If the journal chosen for the submission was neither related to the theoretical foundation nor overly represented in the body of literature analyzed, an explicit justification in the paper itself might be needed. Alternatively, an explanation might be provided in the letter to the editor when submitting the paper. If such a statement is not presented, the likelihood of it being transferred into the review process and passing it is rather low. Finally, we want to refer readers interested in the specificities of the publication-related review process of SLRs to Webster and Watson ( 2002 ), who elaborated on this for Management Information Systems Quarterly.

5 Discussion and conclusion

Critically reviewing the currently available SLR guidelines in the management domain, this paper synthesizes 14 key decisions to be made and reported across the SLR research process. Guidelines are presented for each decision, including tasks that assist in making sound choices to complete the research process and make meaningful contributions. Applying these guidelines should improve the rigor and robustness of many review papers and thus enhance their contributions. Moreover, some practical hints and best-practice examples are provided on issues that unexperienced authors regularly struggle to present in a manuscript (Fisch and Block 2018 ) and thus frustrate reviewers, readers, editors, and authors alike.

Strikingly, the review of prior guidelines reported in Table 3 revealed their focus on the technical details that need to be reported in any SLR. Consequently, our discipline has come a long way in crafting search strings, inclusion, and exclusion criteria, and elaborating on the validity and reliability of an SLR. Nevertheless, we left critical areas underdeveloped, such as the identification of relevant research gaps and questions, data extraction tools, analysis of the findings, and a meaningful and interesting reporting of the results. Our study contributes to filling these gaps by providing operationalized guidance to SLR authors, especially early-stage researchers who craft SLRs at the outset of their research journeys. At the same time, we need to underline that our paper is, of course, not the only useful reference for SLR authors. Instead, the readers are invited to find more guidance on the many aspects to consider in an SLR in the references we provide within the single decisions, as well as in Tables 1 and 2 . The tables also identify the strongholds of other guidelines that our paper does not want to replace but connect and extend at selected occasions, especially in SLR Steps 5 and 6.

The findings regularly underline the interconnection of the 14 decisions identified and discussed in this paper. We thus support Tranfield et al. ( 2003 ) who requested a flexible approach to the SLR while clearly reporting all design decisions and reflecting their impacts. In line with the guidance synthesized in this review, and especially Durach et al. ( 2017 ), we also present a refined framework in Figs.  1 and 2 . It specifically refines the original six-step SLR process by Durach et al. ( 2017 ) in three ways:

Enriched six-step process including the core interrelations of the 14 decisions

First, we subdivided the six steps into 14 decisions to enhance the operationalization of the process and enable closer guidance (see Fig.  1 ). Second, we added a temporal sequence to Fig.  2 by positioning the decisions from left to right according to this temporal sequence. This is based on systematically reflecting on the need to finish one decision before the following. If this need is evident, the following decision moves to the right; if not, the decisions are positioned below each other. Turning to Fig.  2 , it becomes evident that Step 2, “determining the required characteristics of primary studies,” and Step 3, “retrieving a sample of potentially relevant literature,” including their Decisions 4–6, can be conducted in an iterative manner. While this contrasts with the strict division of the six steps by Durach et al. ( 2017 ), it supports other guidance that suggests running pilot studies to iteratively define the literature sample, its sources, and characteristics (Snyder 2019 ; Tranfield et al. 2003 ; Xiao and Watson 2019 ). While this insight might suggest merging Steps 2 and 3, we refrain from this superficial change and building yet another SLR process model. Instead, we prefer to add detail and depth to Durach et al.’s ( 2017 ) model.

(Decisions: D1: specifying the research gap and related research question, D2: opting for a theoretical approach, D3: defining the core theoretical framework and constructs, D4: specifying inclusion and exclusion criteria, D5: defining sources and databases, D6: defining search terms and crafting a search string, D7: including and excluding literature for detailed analysis and synthesis, D8: selecting data extraction tool(s), D9: coding against (pre-defined) constructs, D10: conducting a subsequent (statistical) analysis (optional), D11: ensuring validity and reliability, D12: deciding on the structure of the paper, D13: presenting a refined theoretical framework and discussing its contents, and D14: deriving an appropriate journal from the analyzed papers).

This is also done through the third refinement, which underlines which previous or later decisions need to be considered within each single decision. Such a consideration moves beyond the mere temporal sequence of steps and decisions that does not reflect the full complexity of the SLR process. Instead, its focus is on the need to align, for example, the conduct of the data analysis (Decision 9) with the theoretical approach (Decision 2) and consequently ensure that the chosen theoretical framework and the constructs (Decision 3) are sufficiently defined for the data analysis (i.e., mutually exclusive and comprehensively exhaustive). The mentioned interrelations are displayed in Fig.  2 by means of directed arrows from one decision to another. The underlying explanations can be found in the earlier paper sections by searching for the individual decisions in the text on the impacted decisions. Overall, it is unsurprising to see that the vast majority of interrelations are directed from the earlier to the later steps and decisions (displayed through arrows below the diagonal of decisions), while only a few interrelations are inverse.

Combining the first refinement of the original framework (defining the 14 decisions) and the third refinement (revealing the main interrelations among the decisions) underlines the contribution of this study in two main ways. First, the centrality of ensuring validity and reliability (Decision 11) is underlined. It becomes evident that considerations of validity and reliability are central to the overall SLR process since all steps before the writing of the paper need to be revisited in iterative cycles through Decision 11. Any lack of related considerations will most likely lead to reviewer critique, putting the SLR publication at risk. On the positive side of this centrality, we also found substantial guidance on this issue. In contrast, as evidenced in Table 3 , there is a lack of prior guidance on Decisions 1, 8, 10, 12, 13, and 14, which this study is helping to fill. At the same time, these underexplained decisions are influenced by 14 of the 44 (32%) incoming arrows in Fig.  2 and influence the other decisions in 6 of the 44 (14%) instances. These interrelations among decisions to be considered when crafting an SLR were scattered across prior guidelines, lacked in-depth elaborations, and were hardly explicitly related to each other. Thus, we hope that our study and the refined SLR process model will help enhance the quality and contribution of future SLRs.

Data availablity

The data generated during this research is summarized in Table 3 and the analyzed papers are publicly available. They are clearly identified in Table 3 and the reference list.

Aguinis H, Ramani RS, Alabduljader N (2020) Best-practice recommendations for producers, evaluators, and users of methodological literature reviews. Organ Res Methods. https://doi.org/10.1177/1094428120943281

Article   Google Scholar  

Beske P, Land A, Seuring S (2014) Sustainable supply chain management practices and dynamic capabilities in the food industry: a critical analysis of the literature. Int J Prod Econ 152:131–143. https://doi.org/10.1016/j.ijpe.2013.12.026

Brandenburg M, Rebs T (2015) Sustainable supply chain management: a modeling perspective. Ann Oper Res 229:213–252. https://doi.org/10.1007/s10479-015-1853-1

Carter CR, Rogers DS (2008) A framework of sustainable supply chain management: moving toward new theory. Int Jnl Phys Dist Logist Manage 38:360–387. https://doi.org/10.1108/09600030810882816

Carter CR, Washispack S (2018) Mapping the path forward for sustainable supply chain management: a review of reviews. J Bus Logist 39:242–247. https://doi.org/10.1111/jbl.12196

Clark WR, Clark LA, Raffo DM, Williams RI (2021) Extending fisch and block’s (2018) tips for a systematic review in management and business literature. Manag Rev Q 71:215–231. https://doi.org/10.1007/s11301-020-00184-8

Crane A, Henriques I, Husted BW, Matten D (2016) What constitutes a theoretical contribution in the business and society field? Bus Soc 55:783–791. https://doi.org/10.1177/0007650316651343

Davis J, Mengersen K, Bennett S, Mazerolle L (2014) Viewing systematic reviews and meta-analysis in social research through different lenses. Springerplus 3:511. https://doi.org/10.1186/2193-1801-3-511

Davis HTO, Crombie IK (2001) What is asystematicreview? http://vivrolfe.com/ProfDoc/Assets/Davis%20What%20is%20a%20systematic%20review.pdf . Accessed 22 February 2019

De Lima FA, Seuring S, Sauer PC (2021) A systematic literature review exploring uncertainty management and sustainability outcomes in circular supply chains. Int J Prod Res. https://doi.org/10.1080/00207543.2021.1976859

Denyer D, Tranfield D (2009) Producing a systematic review. In: Buchanan DA, Bryman A (eds) The Sage handbook of organizational research methods. Sage Publications Ltd, Thousand Oaks, CA, pp 671–689

Google Scholar  

Devece C, Ribeiro-Soriano DE, Palacios-Marqués D (2019) Coopetition as the new trend in inter-firm alliances: literature review and research patterns. Rev Manag Sci 13:207–226. https://doi.org/10.1007/s11846-017-0245-0

Dieste M, Sauer PC, Orzes G (2022) Organizational tensions in industry 4.0 implementation: a paradox theory approach. Int J Prod Econ 251:108532. https://doi.org/10.1016/j.ijpe.2022.108532

Donthu N, Kumar S, Mukherjee D, Pandey N, Lim WM (2021) How to conduct a bibliometric analysis: an overview and guidelines. J Bus Res 133:285–296. https://doi.org/10.1016/j.jbusres.2021.04.070

Durach CF, Kembro J, Wieland A (2017) A new paradigm for systematic literature reviews in supply chain management. J Supply Chain Manag 53:67–85. https://doi.org/10.1111/jscm.12145

Fink A (2010) Conducting research literature reviews: from the internet to paper, 3rd edn. SAGE, Los Angeles

Fisch C, Block J (2018) Six tips for your (systematic) literature review in business and management research. Manag Rev Q 68:103–106. https://doi.org/10.1007/s11301-018-0142-x

Fritz MMC, Silva ME (2018) Exploring supply chain sustainability research in Latin America. Int Jnl Phys Dist Logist Manag 48:818–841. https://doi.org/10.1108/IJPDLM-01-2017-0023

Garcia-Torres S, Albareda L, Rey-Garcia M, Seuring S (2019) Traceability for sustainability: literature review and conceptual framework. Supp Chain Manag 24:85–106. https://doi.org/10.1108/SCM-04-2018-0152

Hanelt A, Bohnsack R, Marz D, Antunes Marante C (2021) A systematic review of the literature on digital transformation: insights and implications for strategy and organizational change. J Manag Stud 58:1159–1197. https://doi.org/10.1111/joms.12639

Kache F, Seuring S (2014) Linking collaboration and integration to risk and performance in supply chains via a review of literature reviews. Supp Chain Mnagmnt 19:664–682. https://doi.org/10.1108/SCM-12-2013-0478

Khalid RU, Seuring S (2019) Analyzing base-of-the-pyramid research from a (sustainable) supply chain perspective. J Bus Ethics 155:663–686. https://doi.org/10.1007/s10551-017-3474-x

Koufteros X, Mackelprang A, Hazen B, Huo B (2018) Structured literature reviews on strategic issues in SCM and logistics: part 2. Int Jnl Phys Dist Logist Manage 48:742–744. https://doi.org/10.1108/IJPDLM-09-2018-363

Kraus S, Breier M, Dasí-Rodríguez S (2020) The art of crafting a systematic literature review in entrepreneurship research. Int Entrep Manag J 16:1023–1042. https://doi.org/10.1007/s11365-020-00635-4

Kraus S, Mahto RV, Walsh ST (2021) The importance of literature reviews in small business and entrepreneurship research. J Small Bus Manag. https://doi.org/10.1080/00472778.2021.1955128

Kraus S, Breier M, Lim WM, Dabić M, Kumar S, Kanbach D, Mukherjee D, Corvello V, Piñeiro-Chousa J, Liguori E, Palacios-Marqués D, Schiavone F, Ferraris A, Fernandes C, Ferreira JJ (2022) Literature reviews as independent studies: guidelines for academic practice. Rev Manag Sci 16:2577–2595. https://doi.org/10.1007/s11846-022-00588-8

Leuschner R, Rogers DS, Charvet FF (2013) A meta-analysis of supply chain integration and firm performance. J Supply Chain Manag 49:34–57. https://doi.org/10.1111/jscm.12013

Lim WM, Rasul T (2022) Customer engagement and social media: revisiting the past to inform the future. J Bus Res 148:325–342. https://doi.org/10.1016/j.jbusres.2022.04.068

Lim WM, Yap S-F, Makkar M (2021) Home sharing in marketing and tourism at a tipping point: what do we know, how do we know, and where should we be heading? J Bus Res 122:534–566. https://doi.org/10.1016/j.jbusres.2020.08.051

Lim WM, Kumar S, Ali F (2022) Advancing knowledge through literature reviews: ‘what’, ‘why’, and ‘how to contribute.’ Serv Ind J 42:481–513. https://doi.org/10.1080/02642069.2022.2047941

Lusiantoro L, Yates N, Mena C, Varga L (2018) A refined framework of information sharing in perishable product supply chains. Int J Phys Distrib Logist Manag 48:254–283. https://doi.org/10.1108/IJPDLM-08-2017-0250

Maestrini V, Luzzini D, Maccarrone P, Caniato F (2017) Supply chain performance measurement systems: a systematic review and research agenda. Int J Prod Econ 183:299–315. https://doi.org/10.1016/j.ijpe.2016.11.005

Miemczyk J, Johnsen TE, Macquet M (2012) Sustainable purchasing and supply management: a structured literature review of definitions and measures at the dyad, chain and network levels. Supp Chain Mnagmnt 17:478–496. https://doi.org/10.1108/13598541211258564

Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6:e1000097. https://doi.org/10.1371/journal.pmed.1000097

Mukherjee D, Lim WM, Kumar S, Donthu N (2022) Guidelines for advancing theory and practice through bibliometric research. J Bus Res 148:101–115. https://doi.org/10.1016/j.jbusres.2022.04.042

Mulrow CD (1987) The medical review article: state of the science. Ann Intern Med 106:485–488. https://doi.org/10.7326/0003-4819-106-3-485

Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, Chou R, Glanville J, Grimshaw JM, Hróbjartsson A, Lalu MM, Li T, Loder EW, Mayo-Wilson E, McDonald S, McGuinness LA, Stewart LA, Thomas J, Tricco AC, Welch VA, Whiting P, Moher D (2021) The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. J Clin Epidemiol 134:178–189. https://doi.org/10.1016/j.jclinepi.2021.03.001

Pagell M, Wu Z (2009) Building a more complete theory of sustainable supply chain management using case studies of 10 exemplars. J Supply Chain Manag 45:37–56. https://doi.org/10.1111/j.1745-493X.2009.03162.x

Paul J, Criado AR (2020) The art of writing literature review: What do we know and what do we need to know? Int Bus Rev 29:101717. https://doi.org/10.1016/j.ibusrev.2020.101717

Paul J, Lim WM, O’Cass A, Hao AW, Bresciani S (2021) Scientific procedures and rationales for systematic literature reviews (SPAR-4-SLR). Int J Consum Stud. https://doi.org/10.1111/ijcs.12695

Pearce JM (2018) How to perform a literature review with free and open source software. Pract Assess Res Eval 23:1–13

Rhaiem K, Amara N (2021) Learning from innovation failures: a systematic review of the literature and research agenda. Rev Manag Sci 15:189–234. https://doi.org/10.1007/s11846-019-00339-2

Rojas-Córdova C, Williamson AJ, Pertuze JA, Calvo G (2022) Why one strategy does not fit all: a systematic review on exploration–exploitation in different organizational archetypes. Rev Manag Sci. https://doi.org/10.1007/s11846-022-00577-x

Sauer PC (2021) The complementing role of sustainability standards in managing international and multi-tiered mineral supply chains. Resour Conserv Recycl 174:105747. https://doi.org/10.1016/j.resconrec.2021.105747

Sauer PC, Seuring S (2017) Sustainable supply chain management for minerals. J Clean Prod 151:235–249. https://doi.org/10.1016/j.jclepro.2017.03.049

Seuring S, Gold S (2012) Conducting content-analysis based literature reviews in supply chain management. Supp Chain Mnagmnt 17:544–555. https://doi.org/10.1108/13598541211258609

Seuring S, Müller M (2008) From a literature review to a conceptual framework for sustainable supply chain management. J Clean Prod 16:1699–1710. https://doi.org/10.1016/j.jclepro.2008.04.020

Seuring S, Yawar SA, Land A, Khalid RU, Sauer PC (2021) The application of theory in literature reviews: illustrated with examples from supply chain management. Int J Oper Prod Manag 41:1–20. https://doi.org/10.1108/IJOPM-04-2020-0247

Siems E, Land A, Seuring S (2021) Dynamic capabilities in sustainable supply chain management: an inter-temporal comparison of the food and automotive industries. Int J Prod Econ 236:108128. https://doi.org/10.1016/j.ijpe.2021.108128

Snyder H (2019) Literature review as a research methodology: an overview and guidelines. J Bus Res 104:333–339. https://doi.org/10.1016/j.jbusres.2019.07.039

Spens KM, Kovács G (2006) A content analysis of research approaches in logistics research. Int Jnl Phys Dist Logist Manage 36:374–390. https://doi.org/10.1108/09600030610676259

Tachizawa EM, Wong CY (2014) Towards a theory of multi-tier sustainable supply chains: a systematic literature review. Supp Chain Mnagmnt 19:643–663. https://doi.org/10.1108/SCM-02-2014-0070

Tipu SAA (2022) Organizational change for environmental, social, and financial sustainability: a systematic literature review. Rev Manag Sci 16:1697–1742. https://doi.org/10.1007/s11846-021-00494-5

Touboulic A, Walker H (2015) Theories in sustainable supply chain management: a structured literature review. Int Jnl Phys Dist Logist Manage 45:16–42. https://doi.org/10.1108/IJPDLM-05-2013-0106

Tranfield D, Denyer D, Smart P (2003) Towards a methodology for developing evidence-informed management knowledge by means of systematic review. Br J Manag 14:207–222. https://doi.org/10.1111/1467-8551.00375

Tröster R, Hiete M (2018) Success of voluntary sustainability certification schemes: a comprehensive review. J Clean Prod 196:1034–1043. https://doi.org/10.1016/j.jclepro.2018.05.240

Wang Y, Han JH, Beynon-Davies P (2019) Understanding blockchain technology for future supply chains: a systematic literature review and research agenda. Supp Chain Mnagmnt 24:62–84. https://doi.org/10.1108/SCM-03-2018-0148

Webster J, Watson RT (2002) Analyzing the past to prepare for the future: writing a literature review. MIS Q 26:xiii–xxiii

Wiese A, Kellner J, Lietke B, Toporowski W, Zielke S (2012) Sustainability in retailing: a summative content analysis. Int J Retail Distrib Manag 40:318–335. https://doi.org/10.1108/09590551211211792

Xiao Y, Watson M (2019) Guidance on conducting a systematic literature review. J Plan Educ Res 39:93–112. https://doi.org/10.1177/0739456X17723971

Yavaprabhas K, Pournader M, Seuring S (2022) Blockchain as the “trust-building machine” for supply chain management. Ann Oper Res. https://doi.org/10.1007/s10479-022-04868-0

Zhu Q, Bai C, Sarkis J (2022) Blockchain technology and supply chains: the paradox of the atheoretical research discourse. Transp Res Part E Logist Transp Rev 164:102824. https://doi.org/10.1016/j.tre.2022.102824

Download references

Open Access funding enabled and organized by Projekt DEAL.

Author information

Authors and affiliations.

EM Strasbourg Business School, Université de Strasbourg, HuManiS UR 7308, 67000, Strasbourg, France

Philipp C. Sauer

Chair of Supply Chain Management, Faculty of Economics and Management, The University of Kassel, Kassel, Germany

Stefan Seuring

You can also search for this author in PubMed   Google Scholar

Contributions

The article is based on the idea and extensive experience of SS. The literature search and data analysis has mainly been performed by PCS and supported by SS before the paper manuscript has been written and revised in a common effort of both authors.

Corresponding author

Correspondence to Stefan Seuring .

Ethics declarations

Conflict of interest.

The authors have no competing interests to declare that are relevant to the content of this article.

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/ .

Reprints and permissions

About this article

Sauer, P.C., Seuring, S. How to conduct systematic literature reviews in management research: a guide in 6 steps and 14 decisions. Rev Manag Sci 17 , 1899–1933 (2023). https://doi.org/10.1007/s11846-023-00668-3

Download citation

Received : 29 September 2022

Accepted : 17 April 2023

Published : 12 May 2023

Issue Date : July 2023

DOI : https://doi.org/10.1007/s11846-023-00668-3

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

• Methodology
• Replicability
• Research process
• Structured literature review
• Systematic literature review

JEL Classification

• Find a journal
• Publish with us
• Track your research

How to Do a Systematic Review: A Best Practice Guide for Conducting and Reporting Narrative Reviews, Meta-Analyses, and Meta-Syntheses

Affiliations.

• 1 Behavioural Science Centre, Stirling Management School, University of Stirling, Stirling FK9 4LA, United Kingdom; email: [email protected].
• 2 Department of Psychological and Behavioural Science, London School of Economics and Political Science, London WC2A 2AE, United Kingdom.
• 3 Department of Statistics, Northwestern University, Evanston, Illinois 60208, USA; email: [email protected].
• PMID: 30089228
• DOI: 10.1146/annurev-psych-010418-102803

Systematic reviews are characterized by a methodical and replicable methodology and presentation. They involve a comprehensive search to locate all relevant published and unpublished work on a subject; a systematic integration of search results; and a critique of the extent, nature, and quality of evidence in relation to a particular research question. The best reviews synthesize studies to draw broad theoretical conclusions about what a literature means, linking theory to evidence and evidence to theory. This guide describes how to plan, conduct, organize, and present a systematic review of quantitative (meta-analysis) or qualitative (narrative review, meta-synthesis) information. We outline core standards and principles and describe commonly encountered problems. Although this guide targets psychological scientists, its high level of abstraction makes it potentially relevant to any subject area or discipline. We argue that systematic reviews are a key methodology for clarifying whether and how research findings replicate and for explaining possible inconsistencies, and we call for researchers to conduct systematic reviews to help elucidate whether there is a replication crisis.

Keywords: evidence; guide; meta-analysis; meta-synthesis; narrative; systematic review; theory.

• Guidelines as Topic
• Meta-Analysis as Topic*
• Publication Bias
• Review Literature as Topic
• Systematic Reviews as Topic*

NGR 6638 Population Health Promotion and Management

• Getting Started
• Types of Sources
• Finding Books & Background Resources
• Finding Journal Articles
• Trauma-Informed Care
• Human Trafficking
• Steps for Conducting a Lit Review
• Types of Reviews
• Statistics and Data
• Citing Your Work

What are the types of reviews?

As you begin searching through the literature for evidence, you will come across different types of publications. Below are examples of the most common types and explanations of what they are. Although systematic reviews and meta-analysis are considered the highest quality of evidence, not every topic will have an SR or MA.

Literature Review Examples

Remember, a literature review provides an overview of a topic. There may or may not be a method for how studies are collected or interpreted. Lit reviews aren't always obviously labeled "literature review"; they may be embedded within sections such as the introduction or background. You can figure this out by reading the article. 

• Can the "Appropriate" Footwear Prevent Injury in Leisure-Time Running? Evidence Versus Beliefs. Full Citation: Malisoux, L., & Theisen, D. (2020). Can the “Appropriate” Footwear Prevent Injury in Leisure-Time Running? Evidence Versus Beliefs. Journal of Athletic Training (Allen Press), 55(12), 1215–1223.
• Weightlifting Overhead Pressing Derivatives: A Review of the Literature Full Citation: Soriano, M. A., Suchomel, T. J., & Comfort, P. (2019). Weightlifting overhead pressing derivatives: A review of the literature. Sports Medicine, 49(6), 867-885. doi:10.1007/s40279-019-01096-8

Systematic Review Examples

Systematic reviews address a clinical question.  Reviews are gathered using a specific, defined set of criteria.

• Selection criteria is defined
• The words "Systematic Review" may appear int he title or abstract
• BTW -> Cochrane Reviews aka Systematic Reviews
• Additional reviews can be found by using a systematic review limit 
• First aid cooling techniques for heat stroke and exertional hyperthermia: A systematic review and meta-analysis Full Citation: Douma, M. J., Aves, T., Allan, K. S., Bendall, J. C., Berry, D. C., Chang, W. T., ... & Lin, S. (2020). First Aid Task Force of the International Liaison Committee on Resuscitation. First aid cooling techniques for heat stroke and exertional hyperthermia: A systematic review and meta-analysis. Resuscitation, 148, 173-190.
• Treatment of exertional rhabdomyolysis in athletes Full Citation: Manspeaker, Sarah; Henderson, Kelley; Riddle, Dru Treatment of exertional rhabdomyolysis in athletes, JBI Database of Systematic Reviews and Implementation Reports: June 2016 - Volume 14 - Issue 6 - p 117-147 doi: 10.11124/JBISRIR-2016-001879
• Cochrane Library (Wiley) This link opens in a new window Over 5000 reviews of research on medical treatments, practices, and diagnostic tests are provided in this database. Cochrane Reviews is the premier resource for Evidence Based Practice.
• PubMed (NLM) This link opens in a new window PubMed comprises more than 22 million citations for biomedical literature from MEDLINE, life science journals, and online books.

Meta-Analysis Examples

Meta-analysis is a study that combines data from OTHER studies. All the studies are combined to argue whether a clinical intervention is statistically significant by combining the results from the other studies.  For example, you want to examine a specific headache intervention without running a clinical trial.  You can look at other articles that discuss your clinical intervention, combine all the participants from those articles, and run a statistical analysis to test if your results are significant. Guess what? There's a lot of math. 

• Include the words "meta-analysis" or "meta analysis" in your keywords
• Meta-analyses will always be accompanied by a systematic review, but a systematic review may not have a meta-analysis
• See if the abstract or results section mention a meta-analysis
• Use databases like Cochrane or PubMed
• A Meta-Analysis to Determine if Lower Extremity Muscle Strengthening Should Be Included in Military Knee Overuse Injury-Prevention Programs Full Citation: Roger O. Kollock, Corey Andrews, Ashlyn Johnston, Teresa Elliott, Alan E. Wilson, Kenneth E. Games, JoEllen M. Sefton; A Meta-Analysis to Determine if Lower Extremity Muscle Strengthening Should Be Included in Military Knee Overuse Injury-Prevention Programs. J Athl Train 1 November 2016; 51 (11): 919–926. doi: https://doi.org/10.4085/1062-6050-51.4.09
• Effects of training and competition on the sleep of elite athletes: a systematic review and meta-analysis Full Citation: Roberts SSH, Teo W, Warmington SAEffects of training and competition on the sleep of elite athletes: a systematic review and meta-analysisBritish Journal of Sports Medicine 2019;53:513-522.
• << Previous: Steps for Conducting a Lit Review
• Next: Statistics and Data >>
• Last Updated: May 1, 2024 9:35 AM
• URL: https://libguides.uwf.edu/phmp

Database connection failed!

• Open access
• Published: 02 May 2024

Risk factors and long-term outcomes of oropharyngeal dysphagia in persons with multiple sclerosis: a systematic review protocol

• Zahra Sadeghi 1 ,
• Mohamadreza Afshar 2 ,
• Asefeh Memarian 3 &
• Heather L. Flowers   ORCID: orcid.org/0000-0001-6742-9459 3  

Systematic Reviews volume  13 , Article number:  121 ( 2024 ) Cite this article

Metrics details

Oropharyngeal dysphagia (OPD) can be functionally debilitating in persons with multiple sclerosis (pwMS). OPD induces alterations in safety and efficiency of food and/or liquid ingestion and may incur negative sequalae such as aspiration pneumonia or malnutrition/dehydration. Early detection and timely management of OPD in pwMS could prevent such complications and reduce mortality rates. Identifying risk factors of OPD relative to its onset or repeat manifestation will enable the development of care pathways that target early assessment and sustained management. The aims of this systematic review are to compile, evaluate, and summarize the existing literature reporting potential risk factors and associated long-term outcomes (e.g., aspiration pneumonia, malnutrition, dehydration, and/or death) of OPD in pwMS.

We will undertake a systematic review to identify studies that describe patterns and complications of OPD in pwMS. Variables of interest include predictors of OPD along with long-term outcomes. We will search MEDLINE, Embase, CINAHL, AMED, the Cochrane Library, Web of Science, and Scopus. We will consider studies for inclusion if they involve at least 30 adult participants with MS and report risk factors for OPD and/or its long-term outcomes. Studies will be excluded if they refer to esophageal or oropharyngeal dysphagia induced by causes other than multiple sclerosis. Study selection and data extraction will be performed by two independent assessors for abstract and full article review. We will present study characteristics in tables and document research findings for dysphagia-related risk factors or its complications via a narrative format or meta-analysis if warranted (e.g., mean difference and/or risk ratio measurements). All included studies will undergo risk-of-bias assessment conducted independently by two authors with consensus on quality ratings.

There is a lacune for systematic reviews involving risk factors and long-term outcomes of dysphagia in pwMS to date. Our systematic review will provide the means to develop accurate and efficient management protocols for careful monitoring and evaluation of dysphagia in pwMS. The results of this systematic review will be published in a peer-reviewed journal.

Systematic review registration

PROSPERO CRD42022340625.

Peer Review reports

Oropharyngeal dysphagia (OPD) is common in multiple sclerosis (MS) [ 1 ] due to injury to the corticobulbar tracts, potentially involving the brainstem, the cerebellum [ 2 , 3 ], and the cortex [ 4 ]. There may be a differing clinical course across types of MS, classified based on disease onset and progression over time [ 5 ]. The most frequent includes relapsing–remitting MS (RRMS), usually beginning with acute exacerbation and detrimental impacts that recover fully or partially over time. The other forms of MS are all defined as progressive, including secondary progressive MS (SPMS), primary progressive MS (PPMS), and progressive relapsing MS (PRMS). Patients with RRMS develop SPMS within 10 years of the onset of RRMS. PPMS is the least frequent type of MS and is characterized by deteriorating neurological function from disease onset coupled with a lack of remittance. Nevertheless, superimposed relapses are also evidenced in this type. Overall, the course of MS is unpredictable, and depending on the severity, the diversity of anatomic impacts, and the onset of associated lesions, its clinical manifestations are also heterogeneous.

Symptoms of OPD in MS may include coughing and/or choking on saliva or other liquid and food boluses, feelings of bolus sticking in the throat, the need to swallow multiple times per bolus, difficulty initiating a swallow (accompanied by drooling), and alterations to usual eating patterns (such as viscosity or texture changes) [ 3 , 4 , 6 ]. Oropharyngeal dysphagia may incur severe and multifaceted poor outcomes, such as aspiration pneumonia, malnutrition/dehydration [ 3 , 4 , 6 ], increased psychosocial comorbidities [ 7 , 8 ], and even death during periods of medical instability [ 9 ]. Identifying risk factors for OPD in pwMS will provide the means to develop accurate and efficient management protocols for careful monitoring and evaluation by dysphagia experts. By extension, sustained management will permit timely and comprehensive care to mitigate potential serious complications.

In two recent systematic reviews, the authors provided an estimate of the pooled frequency of dysphagia in pwMS based on a range of evaluation methods, whether screening, clinical, or instrumental examination [ 1 , 10 ]. Guan et al. [ 1 ] reported a pooled frequency estimate of 36% based on subjective screenings or cursory evaluations (such as the Dysphagia in Multiple Sclerosis Questionnaire, the water swallowing test, and various dysphagia checklists from individual clinical swallowing centers) and 81% based on objective measurements (such as videofluoroscopy or fiber-optic nasoendoscopy). More specifically, the frequency of dysphagia was 46% in pwMS when Expanded Disability Status Scale (EDSS) scores were stratified as 4.5 or higher and 40% for those below 3.0. Similarly, patients with longer disease duration (over 10 years) were more likely to have dysphagia symptoms compared with shorter disease duration.

Various individual studies have also shown a higher frequency of dysphagia with greater disability [ 11 , 12 , 13 , 14 , 15 ] and/or disease duration [ 15 , 16 ]. Nevertheless, a few studies have reported that pwMS with low EDSS scores still had dysphagia [ 15 , 17 , 18 ]. To illustrate, Abraham et al. [ 18 ] reported that 43% of pwMS in their sample had dysphagia including 17% with low levels of disability (EDSS score lower than 2.5). Aghaz et al. [ 10 ] estimated the pooled frequency of dysphagia as 37% based on subjective evaluations or cursory checklists versus 47% for objective instrumental evaluations respectively. In contrast to the findings of Abraham et al. [ 18 ], they failed to demonstrate associations for the presence of dysphagia according to EDSS-based disease severity, duration of disease, or MS stage.

Taken together, reported frequencies of dysphagia hover around one-third of pwMS at a given point in time [ 1 , 10 ], whereby varied frequencies relate primarily to evaluation methods, whether screening, clinical assessment, or instrumental evaluation. The most common patient-report tool used to identify dysphagia in pwMS is the Dysphagia in Multiple Sclerosis Questionnaire (DYMUS) [ 1 , 10 , 19 ], involving 10 items with very good reliability and internal consistency. However, frequencies of reported dysphagia based on questionnaires are lower than those based on standardized tools or instrumental evaluations [ 1 ]. In general, instrumental assessment remains the gold standard for dysphagia and aspiration detection, whether by videofluoroscopy or fiber-optic nasoendoscopy rather than various types of screening tools, bedside evaluations, or patient-reported questionnaires. Some pwMS may underestimate their dysphagia severity due to altered sensory appreciation of symptoms, despite instrumental evidence to the contrary.

In addition to our poor understanding of the frequency of dysphagia in pwMS, gaps exist regarding patterns of associations between disease severity, duration, or stage. Notwithstanding, certain predictive factors may well routinely accompany the expression of dysphagia in pwMS. Elucidating such information would require a comprehensive profile of patient groups with known disease severity, duration, and stage alongside MS type, neuroanatomical impacts, and concomitant deficits or disorders. For example, dysphagia may be precipitated by coexisting psychological or cognitive impairments [ 11 , 18 , 20 ]. Therefore, continual monitoring for risk of dysphagia in pwMS who also experience negative mental health symptoms or cognitive disorders [ 4 , 20 ] is warranted. Furthermore, speech impairments (e.g., dysarthria) may provide good and readily identifiable clinical indicators for the presence of dysphagia in persons with neuromuscular diseases [ 21 ]. A systematic appraisal of the literature is required to identify the best available evidence for risk factors of dysphagia along with ensuing long-term sequelae in pwMS.

A systematic review constitutes the highest level of research evidence, especially if there is a quality evaluation and meta-analysis. Therefore, a comprehensive systematic review, aimed at establishing the predictors of dysphagia in pwMS, ideally identified with gold standard evaluation methods (such as instrumental assessment), could facilitate the development of new tools for screening or assessing dysphagia and inform practice guidelines. In addition, a close consideration of associated outcomes over the long term (e.g., pneumonia, poor social participation, death) could contribute to our understanding of prognostic indicators for particular patient groups. Consequently, our purpose is to search the existing literature to systematically identify the risk factors and associated outcomes of oropharyngeal dysphagia over the long term in persons with pwMS.

The protocol of this systematic review has been registered in PROSPERO (registration number: CRD42022340625). We have applied PRISMA-P guidelines to develop this review protocol further. It served to direct our search strategy of databases and the gray literature as well as our data extraction and compilation methods. We will document our article selection results using the PRISMA flow diagram to delineate reasons for abstract and article exclusion until the final set of articles is identified. Our investigation of risk factors is in keeping with recommendations from the Cochrane Prognostic Methods Group ( https://methods.cochrane.org/prognosis/ ) [ 22 ]. We are submitting the protocol prior to undertaking the full search or any subsequent processes such as abstract screening and full article evaluation.

Operational definitions

Oropharyngeal dysphagia is defined as body and structure impairment [ 23 ] in swallowing physiology evidenced by expert clinical or instrumental assessment of function from the anterior aspect of the lips to inferior aspect of the upper esophageal sphincter. Diagnosis of multiple sclerosis is based on accepted criteria for both definite and probable MS, according to a classification scheme that involves expert clinical and objective evaluations (such as neuroimaging) [ 24 ].

Data sources

We will conduct an electronic search in the following databases for abstracts in languages that the co-authors can read (English, French, German, Persian, Portuguese, Spanish, and Turkish). No publication date or study design restrictions will be imposed. Relevant databases will include MEDLINE, Embase, CINAHL, AMED, the Cochrane Library (CENTRAL), Web of Science, and Scopus. The MeSH and search terms used in the search strategy were developed a priori (Table  1 ). A research librarian will consult to enable valid adaptations of the MEDLINE terms into the other databases. Our MEDLNE search was conducted in OVID, revealing 189 citations (April 2023). We will also search international gray literature sources (e.g., OpenGrey and Dissertation Abstracts) and review the bibliographies and citations for all included articles in a reiterative manner until no further possible references are identified.

Eligibility criteria

Studies will be considered for inclusion if they have observational intent and involve retrospective or prospective consecutive or randomly selected sampling (either from a particular cohort or population). Study designs may include case series, cross-sectional, longitudinal, case–control, and/or other observational investigations as well as the control arm (i.e., participants who are not receiving trial-related interventions for MS or dysphagia) of randomized controlled trials. We will consider studies with at least 30 adults (18 years or older) with MS. Studies must include an aim to identify risk factors (e.g., MS subtype, disease duration, EDSS score, age, gender, smoking or alcohol use, psychological symptoms, cognitive impairments, and/or dysarthria) that may precipitate oropharyngeal dysphagia (OPD). We have chosen not to prespecify all possible risk factors as we seek to identify new potential risk factors. The body of evidence is small, and risk factors are likely underrepresented at present. Any potential new risk factors will provide a path for future researchers to investigate them in a comprehensive way and thus extend the literature and knowledge base in this respect.

Corresponding studies that include follow-up time points will contribute to our interest in long-term outcomes (e.g., detrimental medical, activity/participation, or quality-of-life outcomes). Ideally, such studies would have comparable follow-up periods (e.g., yearly) that span the course of the disease and document the outcomes relative to the absence/presence and/or severity classifications of OPD. However, we will not exclude any studies based on their follow-up points or overall time horizon.

During our review of abstracts and full articles, we will apply pre-defined exclusion criteria. That is, we will exclude studies involving convenience samples, those without extractable data (e.g., studies involving aggregate results for multiple etiologies rather than pwMS alone) for our outcomes of interest, and those reporting duplicate data. Any abstracts without corresponding full study publications will be excluded. We will also exclude articles without a clear sample of at least 30 pwMS and corresponding OPD (for at least a declared portion of the sample), identified by clinical or instrumental swallowing assessments. Finally, articles will be excluded if they do not conform to our operational definitions of OPD and MS or if they refer to oropharyngeal dysphagia induced by causes other than multiple sclerosis. We will contact authors when we cannot find full articles or when we wish to elucidate study characteristics such as method of dysphagia assessment. Our full systematic review reporting will conform to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist [ 27 ].

Data collection

Study selection from primary articles will be performed in two stages:

Initial screening and coding of titles and abstracts whereby relevant abstracts (stage 1) will undergo full article review (stage 2) (Table  2 )

Evaluation and coding of full articles for inclusion in the final sample (Table  2 )

The review process will be conducted by two independent reviewers (blind to each other’s coding) across the two stages. Any discrepancies will be resolved by consensus discussion between the two reviewers, and, when agreement is not possible, a third reviewer (also a member of the research team) will read the abstract or article independently and contribute to a decision. All references for the excluded articles will be retained for documentation purposes.

One data extractor will identify pertinent information from the final set of included articles and compile it into a table or spreadsheet. Extracted data will be verified by a second independent reviewer to ensure the accuracy of information from the following categories:

Study characteristics: First author’s name, year of publication, country in which the study was conducted, study design, and size of the sample

Study population and participant characteristics: Age, gender, MS type, disease duration, and EDSS score

Diagnostic assessments for MS and dysphagia

Risk factors for dysphagia whether related to MS (e.g., MS type, disease duration, and EDSS score), to patient characteristics (e.g., age, gender, and smoking or alcohol abuse), and/or to comorbidities (e.g., psychological symptoms, cognitive impairments, or dysarthria)

Follow-up assessments of dysphagia in terms of type and timing

Frequency and impact (e.g., severity) of detrimental medical (e.g., aspiration pneumonia, dehydration, malnutrition, institutionalization, and mortality), activity/participation (e.g., fewer social engagements around meals), or quality-of-life outcomes.

Risk of bias in individual studies

We will apply appropriate risk-of-bias evaluations [ 22 , 28 ] such as the Newcastle–Ottawa scale (NOS) [ 29 ] as a quality evaluation of included obserervational studies. To illustrate, the NOS contains grading for categories of selection (e.g., sample representativeness), comparability (e.g., evaluation of confounders), and outcome (adequacy of follow-up period). Further, if warranted, the Cochrane Collaboration’s risk-of-bias tool will be used for randomized controlled trials, based on the domains: sequence generation, blinding of participants, blinding of outcome measurement, allocation sequence concealment, missing data, selective outcome reporting, and other biases such as sources of funding and conflicts of interest [ 30 ]. Additionally, the Quality in Prognosis Studies (QUIPS) tool will facilitate assessment for the risk factor studies [ 22 ]. For any type of quality appraisal (observational study quality scale, Cochrane’s risk-of-bias tool, or QUIPS), two authors will independently review the included studies and resolve discrepancies by discussion and consensus agreement within the review team.

Data analysis

We will provide a descriptive synthesis of the findings from the included studies, structured around target population characteristics, type of assessments, and outcomes of interest. We will consider meta-analyses if there is an adequate number of studies and homogeneity of study populations and assessment methods. Otherwise, we will present a narrative synthesis of the results. We anticipate that there will be restricted scope for meta-analysis due to differing study populations and/or assessment methods along with a paucity of existing literature. Where studies have similar sample characteristics (including potential comparison groups), assessment methods, and corresponding outcomes, we will pool the results and apply various types of meta-analyses such as mean difference, standard mean difference, or Cox regression for continuous outcomes and risk ratio measurement or logistic regression for categorical outcomes depicted in forest plots along with their 95% confidence intervals (CIs). Finally, we will evaluate the overall strength of the evidence based on discussion among authors through application of a tool such as GRADE.

Assessment of heterogeneity

If there is reason to consider meta-analysis, analyses will be performed using Cochrane’s Review Manager tool (Review Manager: RevMan [computer program]. The Cochrane Collaboration, 2024). We plan to assess study features such as participant age and sex, MS subtype, time course for follow-up, and primary outcome measures as the basis for determining if data pooling for meta-analysis is warranted. Subsequently, if meta-analysis is undertaken, we will apply and interpret the I 2 statistic [ 30 ] as an indictor of heterogeneity relative to the number of studies and direction of effect using the following guide: mild (between 0 and < 25%), moderate (between 25 and < 50%), severe (between 50 and < 75%), and highly severe (between 75 and 100%). If there is moderate heterogeneity, we will present a supplementary qualitative synthesis of the findings.

Analysis of subgroups or subsets

If sufficient data are available, subgroup analyses may be conducted for different OPD assessment methods (e.g., clinical bedside evaluation, fiber-optic nasoendoscopic evaluation, and/or videofluoroscopic evaluation) relative to MS type and risk factors. Similarly, long-term outcomes based on dysphagia status or severity levels will be analyzed according to follow-up periods or comparable overall time horizon.

Assessment of publication bias

Publication bias will be evaluated using a funnel plot (i.e., plots of study results against precision) and Begg’s [ 31 ] and Egger’s [ 32 ] tests if an adequate number of studies (≥ 10) are identified. Additionally, we will incorporate Deek’s asymmetry test [ 33 ] to mitigate overestimation of effects when predictive modeling with odds ratios is applied for the determination of OPD across studies that involve low event proportions. However, if meta-analysis is not possible, publication bias will be assessed descriptively and involve documentation of direction of results across risk factors (whether significant or not) as well as potential follow-up time lags across studies.

Our search strategy is extensive compared to other recent systematic reviews in the field of multiple sclerosis [ 1 , 34 ] given the inclusion of numerous sources and comprehensive search terms. We believe that it will yield a broad capture of abstracts internationally, but that many articles will derive from western or developed countries. This may be an important limitation because many underrepresented countries, such as Iran, have a high and increasing prevalence of pwMS in certain regions [ 35 , 36 ].

Other systematic reviews have undertaken different lines of inquiry such as investigating the prevalence of dysphagia in pwMS without considering risk factors [ 1 ] or long-term outcomes [ 1 , 10 ]. Thus, we will extend the knowledge base in a new content area (involving predictors and long-term outcomes of dysphagia in pwMS). Our identification of literature in the field of MS will provide new insights into the repercussions of dysphagia and offer direction for the development of screening protocols, assessment methods, and improved therapeutic management in pwMS. In the event that our findings elucidate multiple predictors (e.g., related to MS, patient characteristics, and/or comorbidities) and varied outcomes (e.g., medical, activity/participation, or quality of life), they may warrant publication in multiple peer-reviewed papers.

We anticipate that various limitations will result during our search of the literature. First, studies may not report the timeframe between dysphagia onset, assessments, and associated outcomes. Second, dysphagia identification in specific studies might be based on cursory screenings, patient self-report (and potentially non-standardized) questionnaires, and/or subjective clinical assessments rather than on instrumental reference tests such as videofluoroscopy and fiber-optic nasoendoscopy. Finally, it may be difficult to pool results from the existing literature for some of the risk factors or outcomes if investigations restrict enrolment to particular types of pwMS, involving subsamples of larger studies, or if they fail to incorporate shared definitions and research methods in the field of multiple sclerosis [ 36 ].

Although the frequency of dysphagia in pwMS has been a topic of inquiry within the past two decades [ 1 , 10 ], a poor understanding of associations to disease-related risk factors and negative outcomes remains. Our proposed systematic review will address such a gap in the literature, as we will attempt to elucidate risk factors of dysphagia and long-term outcomes from observational studies reporting frequencies of dysphagia over the long term. Where relevant, we will pool results across studies or extract individual-level data that may permit us to model risk factors of dysphagia and/or its associated long-term outcomes in pwMS. Our inquiry will offer the means to inform best practices in the early detection of dysphagia and provide information that can be incorporated into guidelines and clinical practice initiatives for the management of dysphagia in pwMS.

Availability of data and materials

Not applicable.

Abbreviations

• Oropharyngeal dysphagia

Persons with multiple sclerosis

• Multiple sclerosis

Expanded Disability Status Scale

Dysphagia in Multiple Sclerosis Questionnaire

Strengthening the Reporting of Observational Studies in Epidemiology

Newcastle-Ottawa scale

Guan X-L, Wang H, Huang H-S, Meng L. Prevalence of dysphagia in multiple sclerosis: a systematic review and meta-analysis. Neurol Sci. 2015;36:671–81.

Article   PubMed   Google Scholar  

Tassorelli C, Bergamaschi R, Buscone S, Bartolo M, Furnari A, Crivelli P, et al. Dysphagia in multiple sclerosis: from pathogenesis to diagnosis. Neurol Sci. 2008;29:360–3.

Article   Google Scholar  

Marchese-Ragona R, Restivo D, Marioni G, Ottaviano G, Masiero S, Staffieri A. Evaluation of swallowing disorders in multiple sclerosis. Neurol Sci. 2006;27:s335–7.

Calcagno P, Ruoppolo G, Grasso M, De Vincentiis M, Paolucci S. Dysphagia in multiple sclerosis–prevalence and prognostic factors. Acta Neurol Scand. 2002;105(1):40–3.

Article   CAS   PubMed   Google Scholar  

Katz Sand IB, Lublin FD. Diagnosis and differential diagnosis of multiple sclerosis. Continuum. 2013;19(4 Multiple Sclerosis):922-43. https://doi.org/10.1212/01.CON.0000433290.15468.21 .

Prosiegel M, Schelling A, Wagner-Sonntag E. Dysphagia and multiple sclerosis. Int MS J. 2004;11:22–31.

CAS   PubMed   Google Scholar  

Ekberg O, Hamdy S, Woisard V, Wuttge-Hannig A, Ortega P. Social and psychological burden of dysphagia: its impact on diagnosis and treatment. Dysphagia. 2002;17:139–46.

Verdonschot RJ, Baijens LW, Vanbelle S, van de Kolk I, Kremer B, Leue C. Affective symptoms in patients with oropharyngeal dysphagia: a systematic review. J Psychosom Res. 2017;97:102–10.

Sadovnick A, Eisen K, Ebers GC, Paty DW. Cause of death in patients attending multiple sclerosis clinics. Neurology. 1991;41(8):1193.

Aghaz A, Alidad A, Hemmati E, Jadidi H, Ghelichi L. Prevalence of dysphagia in multiple sclerosis and its related factors: systematic review and meta-analysis. Iran J Neurol. 2018;17(4):180.

PubMed   PubMed Central   Google Scholar  

Thomas FJ, Wiles C. Dysphagia and nutritional status in multiple sclerosis. J Neurol. 1999;246:677–82.

Hartelius L, Svensson P. Speech and swallowing symptoms associated with Parkinson’s disease and multiple sclerosis: a survey. Folia Phoniatr Logop. 1994;46(1):9–17.

De Pauw A, Dejaeger E, D’hooghe B, Carton H. Dysphagia in multiple sclerosis. Clin Neurol Neurosurg. 2002;104(4):345–51.

Herrera W, Zeligman BE, Gruber J, Jones MC, Pautler R, Wriston R, et al. Dysphagia in multiple sclerosis: clinical and videofluoroscopic correlations. J Neurol Rehabil. 1990;4(1):1–8.

Google Scholar  

Sadeghi Z, Afshar M, Ebadi A, Baghban K, Qureshi ZS. Swallowing disorder in multiple sclerosis: modified version of the screening tool. Arch Rehabil. 2020;21(2):236–55.

Tarameshlu M, Azimi AR, Ghelichi L, Ansari NN. Prevalence and predictors of dysphagia in Iranian patients with multiple sclerosis. Med J Islam Repub Iran. 2017;31:133.

Article   PubMed   PubMed Central   Google Scholar  

Pajouh SD, Moradi N, Yazdi MJS, Latifi SM, Mehravar M, Majdinasab N, et al. Diagnostic evaluation of dysphagia in multiple sclerosis patients using a Persian version of DYMUS questionnaire. Mult Scler Relat Disord. 2017;17:240–3.

Abraham S, Scheinberg LC, Smith CR, LaRocca NG. Neurologic impairment and disability status in outpatients with multiple sclerosis reporting dysphagia symptomatology. J Neurol Rehabil. 1997;11(1):7–13.

Bergamaschi R, Crivelli P, Rezzani C, Patti F, Solaro C, Rossi P, et al. The DYMUS questionnaire for the assessment of dysphagia in multiple sclerosis. J Neurol Sci. 2008;269(1–2):49–53.

Sadeghi Z, Ghoreishi ZS, Flowers H. et al. Depression, anxiety, and stress relative to swallowing impairment in persons with multiple sclerosis. Dysphagia. 2021;36:902–9. https://doi.org/10.1007/s00455-020-10207-x .

Wang BJ, Carter FL, Altman KW. Relationship between dysarthria and oral-oropharyngeal dysphagia: the present evidence. Ear Nose Throat J. 2020:145561320951647. https://doi.org/10.1177/0145561320951647 .

Riley RD, Ridley G, Williams K, Altman DG, Hayden J, de Vet HC. Prognosis research: toward evidence-based results and a Cochrane methods group. J Clin Epidemiol. 2007;60(8):863-5.

World Health Organization. International classification of functioning, disability and health: ICF. World Health Organization; 2001. https://iris.who.int/handle/10665/42407.

Poser CM, Paty DW, Scheinberg L, McDonald WI, Davis FA, Ebers GC, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol. 1983;13(3):227–31.

Flowers H, Bérubé D, Ebrahimipour M, Perrier M-F, Moloci S, Skoretz S. Swallowing behaviours and feeding environment in relation to communication development from early infancy to 6 years of age: a scoping review protocol. BMJ Open. 2019;9(8):e028850.

Farinotti M, Vacchi L, Simi S, Di Pietrantonj C, Brait L, Filippini G. Dietary interventions for multiple sclerosis. Cochrane Database Syst Rev. 2012;12:CD004192.

PubMed   Google Scholar  

Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7.

Ma L-L, Wang Y-Y, Yang Z-H, Huang D, Weng H, Zeng X-T. Methodological quality (risk of bias) assessment tools for primary and secondary medical studies: what are they and which is better? Mil Med Res. 2020;7:1–11.

Wells G, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses. 2011. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp .

Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928.

Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50(4):1088–101. https://doi.org/10.2307/2533446 .

Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–34. https://doi.org/10.1136/bmj.315.7109.629 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Deeks JJ, Macaskill P, Irwig L. The performance of tests of publication bias and other sample size effects in systematic reviews of diagnostic test accuracy was assessed. J Clin Epidemiol. 2005;58(9):882–93. https://doi.org/10.1016/j.jclinepi.2005.01.016 . PMID: 16085191.

Mirmosayyeb O, Ebrahimi N, Shekarian A, Afshari-Safavi A, Shaygannejad V, Barzegar M, Bagherieh S. Prevalence of dysphagia in patients with multiple sclerosis: a systematic review and meta-analysis. J Clin Neurosci. 2023;108:84–94.

Bagherieh S, Mirmosayyeb O, Vaheb S, Afshari-Safavi A, Barzegar M, Ashtari F, Shaygannejad V. Prevalence and incidence of multiple sclerosis in Isfahan Province, Iran: a 25-year population-based study. Mult Scler Relat Disord. 2023;71:104310.

Cohen JA, Trojano M, Mowry EM, Uitdehaag BM, Reingold SC, Marrie RA. Leveraging real-world data to investigate multiple sclerosis disease behavior, prognosis, and treatment. Mult Scler J. 2020;26(1):23–37.

Download references

Acknowledgements

Author information, authors and affiliations.

Department of Speech Therapy, University of Social Welfare and Rehabilitation Sciences, Tehran, Iran

Zahra Sadeghi

Department of Speech Therapy, School of Rehabilitation, Tehran University of Medical Sciences, Tehran, Iran

Mohamadreza Afshar

School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa, 200 Lees Avenue, Ottawa, ON, K1S 5S9, Canada

Asefeh Memarian & Heather L. Flowers

You can also search for this author in PubMed   Google Scholar

Contributions

ZS conceptualized, designed, and wrote much manuscript. HF conceptualized, designed, wrote, and edited the manuscript. MA conceptualized, designed, and wrote parts of the manuscript. AM developed the search terms and wrote parts of the manuscript.

Corresponding author

Correspondence to Heather L. Flowers .

Ethics declarations

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

The authors declare that they have no competing interests.

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.

Sadeghi, Z., Afshar, M., Memarian, A. et al. Risk factors and long-term outcomes of oropharyngeal dysphagia in persons with multiple sclerosis: a systematic review protocol. Syst Rev 13 , 121 (2024). https://doi.org/10.1186/s13643-024-02530-3

Download citation

Received : 13 September 2023

Accepted : 11 April 2024

Published : 02 May 2024

DOI : https://doi.org/10.1186/s13643-024-02530-3

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

• Systematic review

Systematic Reviews

ISSN: 2046-4053

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

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
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 25 April 2024

Surgery is associated with better long-term outcomes than pharmacological treatment for obesity: a systematic review and meta-analysis

• Leonardo Zumerkorn Pipek 1 ,
• Walter Augusto Fabio Moraes 2 ,
• Rodrigo Massato Nobetani 2 ,
• Vitor Santos Cortez 2 ,
• Alberto Santos Condi 2 ,
• João Victor Taba 2 ,
• Rafaela Farias Vidigal Nascimento 3 ,
• Milena Oliveira Suzuki 2 ,
• Fernanda Sayuri do Nascimento 2 ,
• Vitoria Carneiro de Mattos 2 ,
• Leandro Ryuchi Iuamoto 4 ,
• Wu Tu Hsing 4 ,
• Luiz Augusto Carneiro-D’Albuquerque 5 ,
• Alberto Meyer 5 &
• Wellington Andraus 5  

Scientific Reports volume  14 , Article number:  9521 ( 2024 ) Cite this article

316 Accesses

1 Altmetric

Metrics details

• Endocrine system and metabolic diseases
• Gastrointestinal diseases

Obesity is a highly prevalent disease with numerous complications. Both intensive medical treatment with the use of pharmacological drugs and bariatric surgery are current options. The objective of this meta-analysis was to compare, in the long-term, intensive medical treatment and surgery based on twelve parameters related to weight loss, cardiovascular and endocrine changes. A review of the literature was conducted in accordance with the PRISMA guidelines (PROSPERO: CRD42021265637). The literature screening was done from inception to October 2023 through PubMed, EMBASE and Web of Science databases. We included randomized clinical trials that had separate groups for medical treatment and bariatric surgery as an intervention for obesity. The risk of bias was assessed through RoB2. A meta-analysis was performed with measures of heterogeneity and publication bias. Subgroup analysis for each surgery type was performed. Data is presented as forest-plots. Reviewers independently identified 6719 articles and 6 papers with a total 427 patients were included. All studies were randomized controlled trials, three had a follow up of 5 years and two had a follow up of 10 years. Both groups demonstrated statistical significance for most parameters studied. Surgery was superior for weight loss (− 22.05 kg [− 28.86; − 15.23), total cholesterol (− 0.88 [− 1.59; − 0.17]), triglycerides (− 0.70 [− 0.82; − 0.59]), HDL (0.12 [0.02; 0.23]), systolic pressure (− 4.49 [− 7.65; − 1.33]), diastolic pressure (− 2.28 [− 4.25; − 0.31]), Hb glycated (− 0.97 [− 1.31; − 0.62]), HOMA IR (− 2.94; [− 3.52; − 2.35]) and cardiovascular risk (− 0.08; [− 0.10; − 0.05]). Patient in the surgical treatment group had better long term outcomes when compared to the non-surgical group for most clinical parameters.

Similar content being viewed by others

Two-year effects of semaglutide in adults with overweight or obesity: the STEP 5 trial

Fasting-mimicking diet causes hepatic and blood markers changes indicating reduced biological age and disease risk

Development and validation of a new algorithm for improved cardiovascular risk prediction

Introduction.

Obesity has been a known condition for over 2000 years 1 but that has become much more prevalent in recent decades. Despite great efforts to prevent this disease, the prevalence in adults in the United States has increased in recent decades and reached 42.4% in 2018. The GBD Obesity Study 2 Collaborators 2015 showed that this increasing trend occurred in more than 70 countries and is highly expressive in adolescents.

The classification of obesity is defined by a body mass index (BMI) greater than 30 kg/m 2 . The psychological damage that many of these patients suffer in a society governed by aesthetic standards is just one of the most visible and immediate consequences of obesity. Mortality from cardiovascular causes and its relationship with BMI has already been widely studied 3 , showing that the risk increases progressively with the increase of the index. Similarly, obesity was associated with a higher incidence of cancer 4 , respiratory 5 and metabolic 6 diseases.

In this context, the importance of effective treatment of this condition is clear, reducing mortality and improving the quality of life of these patients. While some benefits are evident with a loss of just 5% 6 of their weight, many patients require a more expressive loss to reduce the risks associated with obesity.

There are several treatments available for weight loss. Lifestyle changes, low calorie diet and increasing physical activity are the mainstay treatment for all patients 7 , 8 . Specific weight loss diets and exercise programs have also been developed for this purpose, yielding varying results. Finally, pharmacological, and surgical treatment has gained more attention in recent years for selected patients in whom other measures were insufficient.

Several studies have demonstrated the effectiveness of bariatric surgery in the short and medium term for the treatment of obesity. More recent studies have also shown that new drugs developed for weight loss may be a viable option for the treatment of this disease 8 , 9 . Comparison of these new drugs with surgical treatment is scarce in the literature and aimed only at evaluating changes related to weight loss in a short period of time.

This systematic review evaluated the hypothesis whether surgical treatment is superior than non-surgical treatment for patients with obesity. We evaluated the long-term effect of these treatments on anthropometric measures (weight, waist circumference, BMI) and on obesity related pathologies (triglycerides, LDL, HDL, total cholesterol, cardiovascular risk, systolic and diastolic blood pressure, HOMA and glycated hemoglobin).

Materials and methods

This systematic review was carried out in accordance with the items of Preferred Reports for Systematic Reviews and Protocol Meta-Analysis (PRISMA-P) 10 and assessing the methodological quality of systematic reviews (AMSTAR-2) guidelines 11 . This study was registered by the Prospective Register of Systematic Reviews (PROSPERO, 258667) before the research was carried out.

Drafting of the research question was based on the PICO strategy 12 , considering: P (Patients with obesity with indication for bariatric surgery based on BMI); I (Bariatric Surgery); C (Pharmacological treatment); O (Long term morbidity/mortality—at least 5 years of follow up).

Eligibility criteria

Inclusion criteria.

Types of studies: Randomized clinical trials.

Types of participants: Patients eligible for bariatric surgery, according to the American Society for Metabolic and Bariatric Surgery (ASMBS).

Types of intervention: Bariatric surgery or medical treatment.

Exclusion criteria

Studies were excluded if they: (1) did not have one group for each type of intervention (surgery or pharmacologic treatment); (2) had a heterogeneous population; (3) did not use a standard assessment method for the entire duration of the study, or did not have pre-assessment; (4) were not related to the question in the review; (5) were in a language other than English, Portuguese or Spanish; (6) were incomplete, unpublished or inaccessible to the authors.

Types of variables/parameters analyzed

Data was collected and arranged in tables, including the authors name, date and country of publication, number of participants included in the final analysis, sex, age, and body mass index.

Literature revision

The survey was from inception to October 10, 2023, without language restrictions, in the Medline database (via PubMed), EMBASE and Web of Science.

Using the search tool, we selected MeSH terms from the most relevant publications to conduct a new search to obtain articles that could be included in this systematic review. In addition, a manual search of theses, meetings, references, study records and contact with experts in the field was carried out.

Search strategy

The same keywords were used in all databases, according to each database input format.

The search strategy was:

(Bariatric Surgery) AND ((nonsurgical) OR (Orlistat) OR (phentermine) OR (topiramate) OR (lorcaserin) OR (naltrexone) OR (bupropion) OR (liraglutide) OR (conservative) OR (conventional) OR (Anti-Obesity Agents) OR (Intensive medical)) AND (obesity) → 3024.

(Bariatric Surgery) AND ((nonsurgical) OR (conservative) OR (Anti-Obesity Agents) OR (Intensive medical)) AND (obesity) → 4732.

Web of Science:

(Bariatric Surgery) AND ((nonsurgical) OR (conservative) OR (Anti-Obesity Agents) OR (Intensive medical)) AND (obesity) → 1772.

Data extraction

The data for each study was extracted independently by two authors. Disagreements were resolved by consensus. If no consensus was reached, a third author was consulted. Data extraction was carried out using the Rayyan tool— https://rayyan.qcri.org/ 13 .

All studies were analyzed by their titles and abstracts, according to inclusion and exclusion criteria. If the eligibility criteria was met, the full text would be extracted. All studies eligible for qualitative analysis are described in the “Results” section.

Missing data was clarified by contacting the authors directly.

Data validation

The risk of bias for intervention-type studies was analyzed using the guidelines of the Cochrane Back Review Group (CBRG) 14 .

Statistical analysis

As several studies of sufficient quality were available, a meta-analysis was carried out with measures of heterogeneity and publication bias. The data was presented through forest-plots, according to their statistical relevance.

Characteristics of study participants are presented as means, minimum and maximum values for quantitative variables, and as frequencies and percentages for qualitative variables. The prevalence values and 95% confidence intervals was calculated using the Wilson method To assess the global heterogeneity between the studies, Cochran's Q test was calculated, as well as the I2 (percentage of variation). The results of the studies' association measures and their respective 95% confidence intervals are presented in forest-plots.

Statistical analysis were performed using the Stata/MP 14.0 software for Windows.

Study selection

The electronic search found 9528 results for the keywords used. After removing 2809 duplicates and screening through abstract, we considered 55 potentially eligible studies for full-text analysis. Of these, 49 did not respect the exclusion criteria. Only 6 studies were considered eligible for qualitative analysis and 6 articles were eligible for meta-analysis [Fig.  1 ].

PRISMA 2020 flow diagram for new systematic reviews.

Many studies were excluded due to lack of description for the intervention in the non-surgical group.

Study characteristics

The following articles were included in the systematic review and meta-analysis 15 , 16 , 17 , 18 , 19 , 20 . In total, there were 427 participants. All studies were RCT. Four had a follow up of five years 15 , 16 , 19 and two had a follow up of 10 years 17 , 18 . Of the six eligible studies, two were undertaken in the United States of America 15 , 16 , two in Italy 17 , 19 , one in Australia 18 , and one in Singapore 20 . Study characteristics and detailed demographics can be found in Tables 1 and 2 . All studies included a group treated exclusively with intensive medical treatment (IMT). The definition of IMT differed between them but were considered if the patients had frequent follow up visits and were instructed on health habits including exercise and diet, with or without the use of pharmacological treatment.

There were four modalities of surgery used for weight loss: Roux-en-Y Gastric Bypass (RYGB) 15 , 17 , 18 , 19 , 20 ; Biliopancreatic diversion (BPD) 17 , 19 ; Laparoscopic Sleeve Gastrectomy (LSG) 15 , 16 ; Laparoscopic Adjustable Gastric Band (LAGB) 18 . The subgroup analysis for outcomes separated studies in RYGB, LSG and other types of surgery. The non-surgical treatment for obesity included one or the combination of the following medications: Orlistat, Phentermine, Naltrexone, Bupropion, Liraglutide, Lorcaserin, Sibutramine.

Risk of bias

After reading the articles included in the systematic review, the following elements were analyzed to determine the level of evidence: study design and selection, detection, loss, reporting and information bias. The summary of the risk of bias analysis for each of the included articles is presented in Fig.  2

Risk of bias analysis.

All studies had a low risk of bias for most criteria. In three of the studies, assessors were aware of the intervention received by study participants or the information was not available 16 , 17 , 20 . Three other studies 15 , 18 , 19 had bias regarding deviations from intended interventions due to the fact that an appropriate analysis to estimate the effects of assignment to intervention was not performed 15 ; patients assigned to the control group crossed over to the intervention group, and no measures were reportedly taken to balance that deviation 19 ; there was a significant loss of follow-up for all groups 20 .

All six studies had data on weight loss after treatment. Mean difference values and their respective 95% confidence intervals (95% CI) were calculated. In Fig.  3 A, the forest plot is shown. All publications found that surgical procedures were more efficient for long term weight loss. The global MD value was − 22.1 kg (95% CI [− 28.9; − 15.2). The measure of heterogeneity I2 (Higgins heterogeneity measure) was 77.8%, a value considered as high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).The subgroup analysis showed that there was not a significant difference between the types of surgery ( p  = 0.30).

(A ) Weight outcomes; ( B ) Waist circumference outcomes; ( C ) BMI outcomes.

Waist circumference

Four studies had data on waist circumference 16 , 17 , 19 , 20 . In Fig.  3 B, the forest plot is shown. Patients treated with surgery had a mean difference of − 12.3 (95% CI [− 15.0; − 9.6]) compared to IMT. The measure of heterogeneity I2 (Higgins heterogeneity measure) was 0%, a value considered as low heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did not allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.99).

The subgroup analysis showed that there was not a significant difference between the types of surgery ( p  = 0.99).

Five studies had data on BMI 16 , 17 , 18 , 19 , 20 . In Fig.  3 C, the forest plot is shown. Patients treated with surgery had a mean difference of − 8.0 (95% CI [− 10.5; − 5.5]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 84%, a value considered high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).

The subgroup analysis showed that there was a significant difference between the types of surgery ( p  = 0.01). The group with LAGB and BPD surgery had the highest decrease in BMI, with a mean of − 10.0.

Triglycerides

Three studies had data on tryglycerides 17 , 19 , 20 . In Fig.  4 A, the forest plot is shown. Patients treated with surgery had a mean difference of − 0.7 (95% CI [− 0.8; − 0.6]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 50.4%, a value considered high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did not allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.08).

(A ) Triglycerides outcomes; ( B ) LDL outcomes; ( C ) HDL outcome; ( D ) Cholesterol outcomes.

The subgroup analysis showed that there was a significant difference between the types of surgery ( p  = 0.01), with a worse outcome for RYGB.

Four studies had data on LDL 16 , 17 , 19 , 20 . In Fig.  4 B, the forest plot is shown. Patients treated with surgery had a mean difference of − 0.5 (95% CI [− 1.0; 0.0]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 92.7%, a value considered high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).

The subgroup analysis showed that there was a significant difference between the types of surgery ( p  = 0.01). There was an increase of 0.5 in LDL for the LSG group. The group with LAGB and BPD surgery had the highest decrease in LDL, with a mean of − 1.3.

Four studies had data on HDL 16 , 17 , 19 , 20 . In Fig.  4 C, the forest plot is shown. Patients treated with surgery had a mean difference of 0.1 (95% CI [0.0; 0.2]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 90.5%, a value considered high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).

The subgroup analysis showed that there was a significant difference between the types of surgery ( p  = 0.01). The group with RYGB surgery had the highest significant increase in HDL, with a mean of 0.2.

Cholesterol

Three studies had data on cholesterol 17 , 19 , 20 . In Fig.  4 D, the forest plot is shown. Patients treated with surgery had a mean difference of − 0.9 (95% CI [− 1.6; − 0.2]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 94.8%, a value considered as high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).

The subgroup analysis showed that there was a significant difference between the types of surgery ( p  = 0.01). The group with LAGB and BPD surgery had the highest decrease in cholesterol, with a mean of − 1.7.

Cardiovascular risk

Two studies had data on cardiovascular risk 17 , 19 . In Fig.  5 A, the forest plot is shown. Patients treated with surgery had a mean difference of − 0.08 (95% CI [− 0.10; − 0.05]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 0%, a value considered as low heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did not allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.44).

(A ) Cardiovascular risk outcomes; ( B ) Systolic blood pressure outcomes; ( C ) Diastolic blood pressure outcomes; ( D ) HOMA outcomes; ( E ) Glycated Hemoglobin outcomes.

The subgroup analysis showed that there was no significant difference between the types of surgery ( p  = 0.36).

Systolic blood pressure

Four studies had data on systolic blood pressure 16 , 17 , 19 , 20 . In Fig.  5 B, the forest plot is shown. Patients treated with surgery had a mean difference of − 4.49 (95% CI [− 7.65; − 1.33]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 71%, a value considered as high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).

The subgroup analysis showed that there was not a significant difference between the types of surgery ( p  = 0.79).

Diastolic blood pressure

Four studies had data on diastolic blood pressure 16 , 17 , 19 , 20 . In Fig.  5 C, the forest plot is shown. Patients treated with surgery had a mean difference of − 2.28 (95% CI [− 4.25; − 0.31]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 60.5%, a value considered as high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).

The subgroup analysis showed that there was not a significant difference between the types of surgery ( p  = 0.66).

Three studies had data on HOMA 15 , 17 , 19 . In Fig.  5 D, the forest plot is shown. Patients treated with surgery had a mean difference of − 2.94 (95% CI [− 3.52; − 2.35]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 14%, a value considered as low heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did not allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.32).

The subgroup analysis showed that there was no significant difference between the types of surgery ( p  = 0.33).

Glycated Hemoglobin

Five studies had data on glycated haemoglobin 15 , 16 , 17 , 19 , 20 . In Fig.  5 E, the forest plot is shown. Patients treated with surgery had a mean difference of − 1.0(95% CI [− 1.3; − 0.6]) compared to IMT. The measure of heterogeneity I2 (Higgins’s heterogeneity measure) was 79.8%, a value considered as high heterogeneity. According to Cochran’s Q heterogeneity test, the sample evidence did allow us to reject the null hypothesis of non-heterogeneity ( p  = 0.01).

The subgroup analysis showed that there was no significant difference between the types of surgery ( p  = 0.98).

Obesity is defined as a BMI greater than or equal to 30 by the CDC and is currently among the most prevalent diseases in the world, in addition to being an important risk factor for many other diseases. It has high rates of morbidity and mortality 21 , 22 and, in this context, weight loss can bring countless positive impacts to the individual. Currently, there are several treatments for obesity, and we can divide them into non-surgical or surgical.

Non-surgical treatments include non-drug and drug treatments. Among the non-medicated, we can highlight the change in eating habits, regular physical exercise, and cognitive behavioral therapy 8 . Ideally, these measures should be implemented for all patients living with obesity, even for those who will undergo drug or surgical treatment. Recently, in addition to lifestyle change, neuromodulation with deep transcranial stimulation has also been studied and has shown effectiveness in weight loss reduction 23 .

A systematic review carried out in 2021, which analyzed 64 articles concluded that among the most effective non-surgical interventions are low-carbohydrate or low-fat diets and combined therapies. This study also showed that non-drug interventions, such as physical exercise, when used alone, are not very effective in reducing the weight of these patients Therefore, a combination of two or more therapies should be chosen 24 .

Pharmacological treatment must be chosen together with the patient. One or more drugs can be used, the main ones used being: Liraglutide, Semaglutide, Tirzepatide, Orlistat, Phentermine and Sibutramine 25 .

Liraglutide was recently approved for the treatment of obesity and is now one of the most widely used drugs. It acts as a GLP-1 receptor agonist 26 , 27 , 28 , enhancing its effects. This group of drugs is already known in the treatment of Type 2 Diabetes Mellitus, a condition that can often be associated with obesity 29 , 30 , since its pathophysiology involves increased insulin resistance. The main actions of this drug are: increased satiety due to a reduction in the speed of gastric emptying, increased insulin release and decreased glucagon release. Semaglutide is a drug with a similar mechanism of action who demonstrated not only a substantial weight loss 31 , but was also associated with a lower 10-year T2D risk in people with overweight or obesity after 2 years of follow up 32 . More recently, a new drug that combines GLP-1 and GIP receptor agonist, Tirzepatide, has shown even better results in the short term 33 .

Orlistat, in turn, reversibly inhibits the lipase enzyme 34 , which has the function of breaking down fat from food for its absorption, as well as inhibiting the absorption of ingested triglycerides. Thus, there is elimination of fat in the feces 35 . The main adverse effects are gastrointestinal symptoms, however this can be beneficial as it leads to a change in behavior, for example causing a lower consumption of foods rich in fat 36 .

Phentermine, an amphetamine analogue, can be used in conjunction with topiramate for the treatment of obesity. The mechanism of action of the drugs is not yet known, however, significant weight loss has already been observed, in addition to a reduction in the consumption of hypercaloric foods and a decrease in the speed of gastric emptying with the use of this combination of drugs 37 , 38 .

Sibutramine, widely used in the 1990s, acts to inhibit the reuptake of serotonin, norepinephrine, and dopamine 34 . Serotonin, in turn, activates POMC system neurons and inhibits NPY neurons, thereby promoting reduced appetite and increased satiety. Despite generating weight reduction 39 , some data show increased cardiovascular risk 40 , and therefore, it is no longer used as a first-line drug.

Among the possible surgeries, the most performed today are: Roux-en-Y Gastric Bypass (RYGB), Biliopancreatic diversion (BPD), Laparoscopic Sleeve Gastrectomy (LSG) and Laparoscopic Adjustable Gastric Band (LAGB). According to the NIH and the American Bariatric Society 41 , 42 , some indications for performing bariatric surgery are adults with BMI greater than or equal to 40 and adults with BMI greater than 35 accompanied by some comorbidity such as type 2 diabetes mellitus, obstructive sleep apnea or hypertension.

RYGB is one of the best-known procedures and its complications vary according to the surgical technique used. Some complications include gastric distention, ulcers, cholelithiasis, hernias, dumping syndrome, and hyperammonaemia encephalopathy.

BPD presents long-term nutritional complications, such as anemia, bone diseases and fat-soluble vitamin deficiency. This technique has high mortality rates, mainly due to the complexity of the technique.

Among the procedures described, LSG is the one with the fewest complications, being described in the literature bleeding or stenosis of the stoma. An alternative technique using endoscopy for sleeve gastroplasty has shown to be safe and efficient for weight loss after 104 weeks, with important improvements in metabolic comorbidities 43 .

The procedure with the lowest mortality rate is the LAGB 44 . Despite this, it can present complications such as obstruction, band erosion, band slippage and gastric prolapse, esophagitis, hernia, in addition to having a high rate of reoperation, reaching 50% of patients who underwent this surgery 45 .

In this article, we compare data on weight loss through intensive drug treatment, which includes changes in eating habits, physical exercise, and medications, and through surgical treatment. Both treatments showed that weight loss caused an improvement in the lipid panel, with a reduction in total cholesterol, triglycerides and LDL, an increase in HDL, improvement in systolic and diastolic blood pressure, decrease in glycated hemoglobin and insulin resistance (accessed through HOMA), in addition to reducing the risk for cardiovascular diseases.

Our systematic review confirmed the findings of individual studies that bariatric surgery has a greater potential for weight reduction, BMI and waist circumference, as already described in individual articles and widely in the literature. It should be noted that even in the long term, this difference remained. Similarly, a 2014 Cochrane systematic review 46 comparing RCT with more than 1 year of follow-up showed that all 7 articles included demonstrated an advantage of the surgical group. An article 47 on the use of pharmacological treatment for obesity showed that even recent drugs approved, including GLP 1 agonists, are not able to reduce weight to levels similar to those of bariatric surgery to date, despite the emergence of new drugs still in initial phase 48 . It is worth mentioning that in these studies the comparison time is relatively short (12 months) and that we do not have data on the long-term impact. Thus, in relation to long term weight loss, bariatric surgery is still the best option.

Most articles were not able to individually demonstrate that surgical treatment is superior to non-surgical in terms of pressure reduction. However, the result of the meta-analysis showed a superiority of the surgical group in relation to both systolic and diastolic pressure, more pronounced in the BPD group. Wang 49 performed a systematic review focused on the impact on pressure and demonstrated that there was a reduction in systolic and diastolic values, but the subgroup analysis showed that this occurs only in the RYGB groups for systolic pressure. Similarly, Schiavon also demonstrated a significant reduction in the need of blood pressure medication after 3 years in the RYGB group when compared intensive medical treatment for obesity 50 . This difference found in only one subtype of surgery seems to be just a reflection of the sample size, which can be interpreted that surgical treatment in general tends to reduce pressure to a greater extent than non-surgical treatment. The fact that different types of surgery are significant may reflect the studies selected in our meta-analysis, which have longer follow-ups.

In relation to both HOMA-IR and glycated Hb, there was a more significant improvement in the group that underwent surgery. The way in which the data on diabetes remission was reported in the articles did not allow a meta-analysis to be carried out with these data and, therefore, it was not included. However, individual data from the Mingrone 2015, Mingrone 2021 and Schauer articles showed that the surgery group had better results. A network meta-analysis from 2021 51 comparing the different types of metabolic surgery for the treatment of obesity and diabetes showed that RYGB was 20% more likely to result in remission of type 2 diabetes compared to SG. There was no significant difference between the other groups. Moreover, the effects of bariatric surgery on diabetes is not exclusive for patients with obesity, as shown by a study with patients with a BMI of 27–32 kg/m 2 that had a better glycemic control when treated with RYGB 20 . Regarding the lipid profile, Schauer's study was not able to demonstrate superiority in relation to LDL and HDL parameters. However, by combining the data from Mingrone's articles, it is possible to demonstrate that surgical treatment is superior. Regarding cholesterol reduction, Mingrone's studies showed that although RYGB and BDP were better in relation to non-surgical treatment, the BDP technique had a statistically greater reduction in relation to RYGB. This can be explained by the greater intestinal exclusion in BDP and, therefore, having a greater impact on lipid absorption. Despite Sayeed's study 52 et al. was not included in this meta-analysis due to the inadequate way of separating the groups for analysis, the results regarding the lipid profile showed that the group that received both interventions was superior to the exclusive non-surgical treatment. It is important to point out that despite a statistically significant difference between the groups, the effect size of this difference is probably not clinically significant.

The choice of treatment for obesity can also have an impact on several other patient comorbidities. Hossain et al. 53 performed a systematic review with 26 studies that showed that bariatric surgery appears to be more effective in the treatment of asthma. Similarly, a study by Crawford et al. 15 showed that there is a greater increase in bone turnover in groups undergoing bariatric surgery in relation to pharmacological treatment. Other than that, bariatric surgery is also demonstrated to be superior in the treatment of other obesity related pathologies, such as Non-Alcoholic Steatohepatitis (NASH), and in the treatment of obesity in adolescents 54 , 55 .

The effect of major cardiovascular adverse events (MACE) and mortality 56 have also been promising for bariatric surgery. A recent cohort comparing bariatric surgery in patients with obesity and use of GLP1-agonists inpatients with diabetes showed a lower risk of MACE in the surgical group 57 . The surgical treatment has also shown superiority when compared to medical treatment regarding the prevention of diabetic kidney disease in 5 years for patients with diabetes and obesity 58 . Boyers et al. evaluated the cost-effectiveness of surgical and pharmacological treatment in the treatment of obesity and found that RYGB should be the treatment of choice only if the optimization of health system costs is considered 59 .

Another important consideration is the fact that pharmacological and surgical treatment for obesity are not mutually exclusive. Most clinicians choose to combine both treatment modalities in practice to improve results. Weight gain after bariatric surgery is a known possibility, and for those patients, two-thirds of the weight regain can be safely lost with GLP1 agonist, providing clinicians with a therapeutic option for this clinical challenge.

Methodologies and limitations of the studies

Despite the large number of articles in the literature on the treatment of obesity, there are few RCTs comparing non-surgical and surgical treatment, and most of them only follow up in the short term. In addition, many articles do not adequately describe the strategy used in non-surgical treatment. This lack of data and standardization in this type of treatment can lead to bias and possibly the formation of extremely heterogeneous groups for analysis.

Most of the studies included in our systematic review have diabetes as an inclusion criteria. In this circumstance, our findings may not be generalized to patients with obesity without diabetes.

Another important limitation of our systematic review refers to pharmacological treatment in the non-surgical group. The use of GLP 1 agonists has great potential in the treatment of obesity, but they have only started to be used recently. As the purpose of our article is to assess the long-term impact, there are still few articles available that used this drug. The use of the most recent medications, such as Tirzepatide, could not be evaluated in our study, once there are no RCTs in the literature presenting its long-term effects. Those drugs proved to be very efficient and might have similar effect in the long term. Future systematic reviews may reveal a different results when including the new generation of weight loss medication.

Finally, choosing the most appropriate treatment often involves individual characteristics of each patient, and the impact on quality of life can be extremely subjective and difficult to assess.

Obesity is a disease that increases the morbidity and mortality of patients, contributing to several secondary diseases. This systematic review evaluated the impact on the main variables related to obesity in the long term. The findings indicated that both treatment modalities are efficacious in managing obesity; however, the surgical group demonstrated superior outcomes in comparison to the non-surgical group across most variables. Nonetheless, the advent of novel pharmacological treatments has shown promising potential. Further studies focusing on the long-term impacts of these new drug treatments should be undertaken to allow for a comprehensive comparison with non-surgical treatment methods.

Data availability

Data is provided within the manuscript or supplementary information files.

Bray, G. The Battle of the Bulge: A History of Obesity Research (Dorrance Pub., 2007).

Collaborators GBD 2015 O, Afshin, A., Forouzanfar, M. H., Reitsma, M. B., Sur, P., Estep, K. et al. Health effects of overweight and obesity in 195 countries over 25 years. N. Engl. J. Med. 377 , 13–27 (2017).

Whitlock, G. et al. Body-mass index and cause-specific mortality in 900 000 adults: Collaborative analyses of 57 prospective studies. Lancet 373 , 1083–1096 (2009).

Article   PubMed   Google Scholar  

Steele, C. B. et al. Vital Signs: Trends in incidence of cancers associated with overweight and obesity—United States, 2005–2014. MMWR Morb. Mortal. Wkly. Rep. 66 , 1052–1058 (2017).

Article   PubMed   PubMed Central   Google Scholar  

Goldhaber, S. Z. et al. Risk factors for pulmonary embolism. The Framingham Study. Am. J. Med. 74 , 1023–1028 (1983).

Article   CAS   PubMed   Google Scholar  

Knowler, W. C. et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N. Engl. J. Med. 346 , 393–403 (2002).

Bray, G. A., Frühbeck, G., Ryan, D. H. & Wilding, J. P. H. Management of obesity. The Lancet 387 , 1947–1956 (2016).

Article   Google Scholar  

Perdomo, C. M., Cohen, R. V., Sumithran, P., Clément, K. & Frühbeck, G. Contemporary medical, device, and surgical therapies for obesity in adults. The Lancet 401 , 1116–1130 (2023).

Updike, W. H. et al. Is it time to expand glucagon-like peptide-1 receptor agonist use for weight loss in patients without diabetes?. Drugs 81 , 881–893 (2021).

Moher, D. et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst. Rev. 4 , 1 (2015).

Shea, B. J. et al. AMSTAR 2: A critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ 358 , j4008 (2017).

Brown, D. A review of the PubMed PICO Tool: using evidence-based practice in health education. Health Promot. Pract. 21 , 496–498 (2020).

Ouzzani, M., Hammady, H., Fedorowicz, Z. & Elmagarmid, A. Rayyan—A web and mobile app for systematic reviews. Syst. Rev. 5 , 210 (2016).

Leeflang, M. M. G., Deeks, J. J., Takwoingi, Y. & Macaskill, P. Cochrane diagnostic test accuracy reviews. Syst. Rev. 2 , 82 (2013).

Crawford, M. R. et al. Increased bone turnover in type 2 diabetes patients randomized to bariatric surgery versus medical therapy at 5 years. Endocr. Pract. 24 , 256–264 (2018).

Schauer, P. R. et al. Bariatric surgery versus intensive medical therapy for diabetes—5-Year outcomes. N. Engl. J. Med. 376 , 641–651 (2017).

Mingrone, G. et al. Metabolic surgery versus conventional medical therapy in patients with type 2 diabetes: 10-Year follow-up of an open-label, single-centre, randomised controlled trial. The Lancet 397 , 293–304 (2021).

O’Brien, P. E., Brennan, L., Laurie, C. & Brown, W. Intensive medical weight loss or laparoscopic adjustable gastric banding in the treatment of mild to moderate obesity: Long-term follow-up of a prospective randomised trial. Obes. Surg. 23 , 1345–1353 (2013).

Mingrone, G. et al. Bariatric-metabolic surgery versus conventional medical treatment in obese patients with type 2 diabetes: 5 Year follow-up of an open-label, single-centre, randomised controlled trial. The Lancet 386 , 964–973 (2015).

Cheng, A. et al. Roux-en-Y gastric bypass versus best medical treatment for type 2 diabetes mellitus in adults with body mass index between 27 and 32 kg/m 2 : A 5-year randomized controlled trial. Diabetes Res. Clin. Pract. 188 , 109900 (2022).

Chooi, Y. C., Ding, C. & Magkos, F. The epidemiology of obesity. Metabolism 92 , 6–10 (2019).

Christensen, S. Recognizing obesity as a disease. J. Am. Assoc. Nurse Pract. 32 , 497–503 (2020).

Ferrulli, A. et al. Weight loss induced by deep transcranial magnetic stimulation in obesity: A randomized, double-blind, sham-controlled study. Diabetes Obes. Metab. 21 , 1849–1860 (2019).

Twells, L. K. et al. Nonsurgical weight loss interventions: A systematic review of systematic reviews and meta-analyses. Obes. Rev. 22 , e13320 (2021).

Rosa-Gonçalves, P. & Majerowicz, D. Pharmacotherapy of obesity: Limits and perspectives. Am. J. Cardiovasc. Drugs 19 , 349–364 (2019).

Pi-Sunyer, X. et al. A randomized, controlled trial of 3.0 mg of liraglutide in weight management. N. Engl. J. Med. 373 , 11–22 (2015).

Knudsen, L. B. & Lau, J. The discovery and development of liraglutide and semaglutide. Front. Endocrinol. (Lausanne) 10 , 155 (2019).

de Oca alejandra PZMTS, PelliTero S, PUig-DoMingo M. obesity and glP-1. Minerva Endocrinology 46 , 168–176 (2021).

Kahn, S. E., Hull, R. L. & Utzschneider, K. M. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature 444 , 840–846 (2006).

Article   ADS   CAS   PubMed   Google Scholar  

Rubio-Almanza, M., Cámara-Gómez, R. & Merino-Torres, J. F. Obesity and type 2 diabetes: Also linked in therapeutic options. Endocrinol. Diabetes Nutr. 66 , 140–149 (2019).

Wharton, S. et al. Two-year effect of semaglutide 2.4 mg on control of eating in adults with overweight/obesity: STEP 5. Obesity 31 , 703–715 (2023).

Wilkinson, L. et al. Effect of semaglutide 2.4 mg once weekly on 10-year type 2 diabetes risk in adults with overweight or obesity. Obesity 31 , 2249–2259 (2023).

Frías, J. P. et al. Tirzepatide versus semaglutide once weekly in patients with type 2 diabetes. N. Engl. J. Med. 385 , 503–515 (2021).

Son, J. W. & Kim, S. Comprehensive review of current and upcoming anti-obesity drugs. Diabetes Metab. J. 44 , 802–818 (2020).

Ballinger, A. & Peikin, S. R. Orlistat: Its current status as an anti-obesity drug. Eur. J. Pharmacol. 440 , 109–117 (2002).

Zhou, Y. H. et al. Effect of anti-obesity drug on cardiovascular risk factors: A systematic review and meta-analysis of randomized controlled trials. PLoS One 7 , e39062 (2012).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Cosentino, G., Conrad, A. O. & Uwaifo, G. I. Phentermine and topiramate for the management of obesity: A review. Drug Des. Dev. Ther. 7 , 267–278 (2013).

CAS   Google Scholar  

Smith, S. M., Meyer, M. & Trinkley, K. E. Fentermina/topiramato (qsymia) para el tratamiento de obesidad. Ann. Pharmacother. 47 , 340–349 (2013).

Sharma, B. & Henderson, D. C. Sibutramine: Current status as an anti-obesity drug and its future perspectives. Expert Opin. Pharmacother. 9 , 2161–2173 (2008).

Tziomalos, K., Krassas, G. E. & Tzotzas, T. The use of sibutramine in the management of obesity and related disorders: An update. Vasc. Health Risk Manag. 5 , 441–452 (2009).

CAS   PubMed   PubMed Central   Google Scholar  

Burguera, B. et al. Critical assessment of the current guidelines for the management and treatment of morbidly obese patients. J. Endocrinol. Invest. 30 , 844–852 (2007).

Grundy, S. M. et al. Gastrointestinal surgery for severe obesity. Ann. Intern. Med. 115 , 956–961 (1991).

Abu Dayyeh, B. K. et al. Endoscopic sleeve gastroplasty for treatment of class 1 and 2 obesity (MERIT): A prospective, multicentre, randomised trial. The Lancet 400 , 441–451 (2022).

Chapman, A. E. et al. Laparoscopic adjustable gastric banding in the treatment of obesity: A systematic literature review. Surgery 135 , 326–351 (2004).

Himpens, J. et al. Long-term outcomes of laparoscopic adjustable gastric banding. Arch. Surg. 146 , 802–807 (2011).

Colquitt, J. L., Pickett, K., Loveman, E. & Frampton, G. K. Surgery for weight loss in adults. Cochrane Database Syst. Rev. 2014 , CD003641 (2014).

Cotugno, M. et al. Clinical efficacy of bariatric surgery versus liraglutide in patients with type 2 diabetes and severe obesity: A 12-month retrospective evaluation. Acta Diabetol. 52 , 331–336 (2014).

Tan, Q. et al. Recent advances in incretin-based pharmacotherapies for the treatment of obesity and diabetes. Front. Endocrinol. (Lausanne) https://doi.org/10.3389/fendo.2022.838410 (2022).

Wang, L. et al. The impact of bariatric surgery versus non-surgical treatment on blood pressure: Systematic review and meta-analysis. Obes. Surg. 31 , 4970–4984 (2021).

Schiavon, C. A. et al. Three-year outcomes of bariatric surgery in patients with obesity and hypertension. Ann. Intern. Med. 173 , 685–693 (2020).

Currie, A. C., Askari, A., Fangueiro, A. & Mahawar, K. Network meta-analysis of metabolic surgery procedures for the treatment of obesity and diabetes. Obes. Surg. 31 , 4528–4541 (2021).

Ikramuddin, S. et al. Lifestyle intervention and medical management with vs without roux-en-y gastric bypass and control of hemoglobin a1c, ldl cholesterol, and systolic blood pressure at 5 years in the diabetes surgery study. JAMA J. Am. Med. Assoc. 319 , 266–278 (2018).

Hossain, N., Arhi, C. & Borg, C. M. Is bariatric surgery better than nonsurgical weight loss for improving asthma control? A systematic review. Obes. Surg. 31 , 1810–1832 (2021).

Järvholm, K. et al. Metabolic and bariatric surgery versus intensive non-surgical treatment for adolescents with severe obesity (AMOS2): A multicentre, randomised, controlled trial in Sweden. Lancet Child Adolesc. Health 7 , 249–260 (2023).

Verrastro, O. et al. Bariatric–metabolic surgery versus lifestyle intervention plus best medical care in non-alcoholic steatohepatitis (BRAVES): A multicentre, open-label, randomised trial. The Lancet 401 , 1786–1797 (2023).

Courcoulas, A. P. et al. Reduction in long-term mortality after sleeve gastrectomy and gastric bypass compared to nonsurgical patients with severe obesity. Ann. Surg. 277 , 442–448 (2023).

Stenberg, E. & Näslund, E. Major adverse cardiovascular events among patients with type-2 diabetes, a nationwide cohort study comparing primary metabolic and bariatric surgery to GLP-1 receptor agonist treatment. Int. J. Obes. 47 , 251–256 (2023).

Article   CAS   Google Scholar  

Bjornstad, P. et al. Effect of surgical versus medical therapy on diabetic kidney disease over 5 years in severely obese adolescents with type 2 diabetes. Diabetes Care 43 , 187–195 (2020).

Boyers, D. et al. Cost-effectiveness of bariatric surgery and non-surgical weight management programmes for adults with severe obesity: A decision analysis model. Int. J. Obes. 45 , 2179–2190 (2021).

Download references

Acknowledgements

The authors are thankful to Justin Axel-Berg for the English corrections and Rossana V. Mendoza López for the statistical analysis.

Author information

Authors and affiliations.

Department of Neurology, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de Sao Paulo, São Paulo, SP, Brazil

Leonardo Zumerkorn Pipek

Faculty of Medicine FMUSP, University of São Paulo, São Paulo, Brazil

Walter Augusto Fabio Moraes, Rodrigo Massato Nobetani, Vitor Santos Cortez, Alberto Santos Condi, João Victor Taba, Milena Oliveira Suzuki, Fernanda Sayuri do Nascimento & Vitoria Carneiro de Mattos

Centro Universitário FMABC, Santo André, São Paulo, Brazil

Rafaela Farias Vidigal Nascimento

Center of Acupuncture, Department of Orthopaedics and Traumatology, University of São Paulo, São Paulo, Brazil

Leandro Ryuchi Iuamoto & Wu Tu Hsing

Department of Gastroenterology, Hospital das Clínicas, HCFMUSP, Avenida Doutor Arnaldo, 455, São Paulo, Brazil

Luiz Augusto Carneiro-D’Albuquerque, Alberto Meyer & Wellington Andraus

You can also search for this author in PubMed   Google Scholar

Contributions

Conceptualization: L.Z.P., A.M. Methodology: L.Z.P., L.R.I., A.M. Formal analysis: L.Z.P., R.F.V.N., A.M. Data Curation: L.Z.P., W.A.F.B., R.M.N., V.S.C., A.S.C., J.V.T., R.F.V.N., M.O.S., F.S.N., V.C.M. Writing—Original Draft: L.Z.P., W.A.F.B., R.M.N., V.S.C., A.S.C., J.V.T., R.F.V.N., M.O.S., F.S.N., V.C.M. Writing—Review and Editing: L.Z.P., R.F.V.N., L.R.I., W.T.H., L.A.C., A.M., W.A. Visualization: L.Z.P., R.F.V.N., A.M. Supervision: L.R.I., W.T.H., L.A.C., A.M., W.A. Project administration: L.R.I., W.T.H., L.A.C., A.M., W.A.

Corresponding author

Correspondence to Alberto Meyer .

Ethics declarations

Competing interests.

The authors declare no competing interests.

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/ .

Reprints and permissions

About this article

Cite this article.

Pipek, L.Z., Moraes, W.A.F., Nobetani, R.M. et al. Surgery is associated with better long-term outcomes than pharmacological treatment for obesity: a systematic review and meta-analysis. Sci Rep 14 , 9521 (2024). https://doi.org/10.1038/s41598-024-57724-5

Download citation

Received : 19 November 2023

Accepted : 21 March 2024

Published : 25 April 2024

DOI : https://doi.org/10.1038/s41598-024-57724-5

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

• Long term outcome
• Pharmacological treatment

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

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

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

To read this content please select one of the options below:

Please note you do not have access to teaching notes, reflections on solar energy adoption research: a semi-systematic review.

International Journal of Energy Sector Management

ISSN : 1750-6220

Article publication date: 3 May 2024

Research on solar energy adoption offers a multidimensional scope and warrants exploration from multiple perspectives, including political, economic, management, behavioral, policy and innovation aspects. The aim of this paper is to comprehensively consolidate major research findings on the premise of solar energy adoption and to disclose gaps in the existing literature.

Design/methodology/approach

A bibliometric analysis of the vast literature is conducted on 1,009 meticulously shortlisted articles following the semi-systematic literature review methodology. A text analytics tool named BibExcel is used for synthesizing the literature, and the results are visualized using Gephi, Pajek and a spreadsheet application.

This paper reports the evolution of research in the selected domain. It is noted that research in this domain was primarily concentrated on four broad themes, namely, peer effects and spatial patterns, public perceptions, policies and economics and technological evolution. The analysis further reveals the merging of two of these themes as a result of transdisciplinary research and also projects future research trends emphasizing political interventions in technological evolution and diffusion.

Originality/value

Research trends and future research scope are identified and discussed in detail. The information revealed from the analysis, along with the research implications, will assist policymakers in noting the flaws in the current doctrines and practices, entrepreneurs in understanding potential enablers and barriers influencing solar energy adoption and budding scholars in comprehending the current research status and framing promising research objectives to close the existing research gaps.

• Solar energy adoption
• Semi-systematic literature review
• Bibliometric analysis

Acknowledgements

The authors gratefully acknowledge the Editor, Associate Editor and anonymous reviewers for their constructive suggestions for enriching the quality of the paper. They also thank PGDM students of XIME Kochi from the Operations Management specialization for assisting in retrieving relevant data from the WoS platform.

Declarations .

Funding: No funding was received to assist with the preparation of this manuscript.

Competing interests: The authors have no competing interests to declare that are relevant to the content of this article.

Data availability: Data will be made available on request.

C.R., V. and John, J. (2024), "Reflections on solar energy adoption research: a semi-systematic review", International Journal of Energy Sector Management , Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/IJESM-01-2024-0028

Emerald Publishing Limited

Copyright © 2024, Emerald Publishing Limited

Related articles

We’re listening — tell us what you think, something didn’t work….

Report bugs here

All feedback is valuable

Please share your general feedback

Join us on our journey

Platform update page.

Visit emeraldpublishing.com/platformupdate to discover the latest news and updates

Questions & More Information

Answers to the most commonly asked questions here