case study for class 3

• Our Mission

Making Learning Relevant With Case Studies

The open-ended problems presented in case studies give students work that feels connected to their lives.

To prepare students for jobs that haven’t been created yet, we need to teach them how to be great problem solvers so that they’ll be ready for anything. One way to do this is by teaching content and skills using real-world case studies, a learning model that’s focused on reflection during the problem-solving process. It’s similar to project-based learning, but PBL is more focused on students creating a product.

Case studies have been used for years by businesses, law and medical schools, physicians on rounds, and artists critiquing work. Like other forms of problem-based learning, case studies can be accessible for every age group, both in one subject and in interdisciplinary work.

You can get started with case studies by tackling relatable questions like these with your students:

• How can we limit food waste in the cafeteria?
• How can we get our school to recycle and compost waste? (Or, if you want to be more complex, how can our school reduce its carbon footprint?)
• How can we improve school attendance?
• How can we reduce the number of people who get sick at school during cold and flu season?

Addressing questions like these leads students to identify topics they need to learn more about. In researching the first question, for example, students may see that they need to research food chains and nutrition. Students often ask, reasonably, why they need to learn something, or when they’ll use their knowledge in the future. Learning is most successful for students when the content and skills they’re studying are relevant, and case studies offer one way to create that sense of relevance.

Teaching With Case Studies

Ultimately, a case study is simply an interesting problem with many correct answers. What does case study work look like in classrooms? Teachers generally start by having students read the case or watch a video that summarizes the case. Students then work in small groups or individually to solve the case study. Teachers set milestones defining what students should accomplish to help them manage their time.

During the case study learning process, student assessment of learning should be focused on reflection. Arthur L. Costa and Bena Kallick’s Learning and Leading With Habits of Mind gives several examples of what this reflection can look like in a classroom: 

Journaling: At the end of each work period, have students write an entry summarizing what they worked on, what worked well, what didn’t, and why. Sentence starters and clear rubrics or guidelines will help students be successful. At the end of a case study project, as Costa and Kallick write, it’s helpful to have students “select significant learnings, envision how they could apply these learnings to future situations, and commit to an action plan to consciously modify their behaviors.”

Interviews: While working on a case study, students can interview each other about their progress and learning. Teachers can interview students individually or in small groups to assess their learning process and their progress.

Student discussion: Discussions can be unstructured—students can talk about what they worked on that day in a think-pair-share or as a full class—or structured, using Socratic seminars or fishbowl discussions. If your class is tackling a case study in small groups, create a second set of small groups with a representative from each of the case study groups so that the groups can share their learning.

4 Tips for Setting Up a Case Study

1. Identify a problem to investigate: This should be something accessible and relevant to students’ lives. The problem should also be challenging and complex enough to yield multiple solutions with many layers.

2. Give context: Think of this step as a movie preview or book summary. Hook the learners to help them understand just enough about the problem to want to learn more.

3. Have a clear rubric: Giving structure to your definition of quality group work and products will lead to stronger end products. You may be able to have your learners help build these definitions.

4. Provide structures for presenting solutions: The amount of scaffolding you build in depends on your students’ skill level and development. A case study product can be something like several pieces of evidence of students collaborating to solve the case study, and ultimately presenting their solution with a detailed slide deck or an essay—you can scaffold this by providing specified headings for the sections of the essay.

Problem-Based Teaching Resources

There are many high-quality, peer-reviewed resources that are open source and easily accessible online.

• The National Center for Case Study Teaching in Science at the University at Buffalo built an online collection of more than 800 cases that cover topics ranging from biochemistry to economics. There are resources for middle and high school students.
• Models of Excellence , a project maintained by EL Education and the Harvard Graduate School of Education, has examples of great problem- and project-based tasks—and corresponding exemplary student work—for grades pre-K to 12.
• The Interdisciplinary Journal of Problem-Based Learning at Purdue University is an open-source journal that publishes examples of problem-based learning in K–12 and post-secondary classrooms.
• The Tech Edvocate has a list of websites and tools related to problem-based learning.

In their book Problems as Possibilities , Linda Torp and Sara Sage write that at the elementary school level, students particularly appreciate how they feel that they are taken seriously when solving case studies. At the middle school level, “researchers stress the importance of relating middle school curriculum to issues of student concern and interest.” And high schoolers, they write, find the case study method “beneficial in preparing them for their future.”

If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

To log in and use all the features of Khan Academy, please enable JavaScript in your browser.

Center for Teaching

Case studies.

Print Version

Case studies are stories that are used as a teaching tool to show the application of a theory or concept to real situations. Dependent on the goal they are meant to fulfill, cases can be fact-driven and deductive where there is a correct answer, or they can be context driven where multiple solutions are possible. Various disciplines have employed case studies, including humanities, social sciences, sciences, engineering, law, business, and medicine. Good cases generally have the following features: they tell a good story, are recent, include dialogue, create empathy with the main characters, are relevant to the reader, serve a teaching function, require a dilemma to be solved, and have generality.

Instructors can create their own cases or can find cases that already exist. The following are some things to keep in mind when creating a case:

• What do you want students to learn from the discussion of the case?
• What do they already know that applies to the case?
• What are the issues that may be raised in discussion?
• How will the case and discussion be introduced?
• What preparation is expected of students? (Do they need to read the case ahead of time? Do research? Write anything?)
• What directions do you need to provide students regarding what they are supposed to do and accomplish?
• Do you need to divide students into groups or will they discuss as the whole class?
• Are you going to use role-playing or facilitators or record keepers? If so, how?
• What are the opening questions?
• How much time is needed for students to discuss the case?
• What concepts are to be applied/extracted during the discussion?
• How will you evaluate students?

To find other cases that already exist, try the following websites:

• The National Center for Case Study Teaching in Science , University of Buffalo. SUNY-Buffalo maintains this set of links to other case studies on the web in disciplines ranging from engineering and ethics to sociology and business
• A Journal of Teaching Cases in Public Administration and Public Policy , University of Washington

For more information:

• World Association for Case Method Research and Application

Book Review :  Teaching and the Case Method , 3rd ed., vols. 1 and 2, by Louis Barnes, C. Roland (Chris) Christensen, and Abby Hansen. Harvard Business School Press, 1994; 333 pp. (vol 1), 412 pp. (vol 2).

Teaching Guides

• Online Course Development Resources
• Principles & Frameworks
• Pedagogies & Strategies
• Reflecting & Assessing
• Challenges & Opportunities
• Populations & Contexts

Quick Links

• Services for Departments and Schools
• Examples of Online Instructional Modules

Do Your Students Know How to Analyze a Case—Really?

Explore more.

• Case Teaching
• Student Engagement

J ust as actors, athletes, and musicians spend thousands of hours practicing their craft, business students benefit from practicing their critical-thinking and decision-making skills. Students, however, often have limited exposure to real-world problem-solving scenarios; they need more opportunities to practice tackling tough business problems and deciding on—and executing—the best solutions.

To ensure students have ample opportunity to develop these critical-thinking and decision-making skills, we believe business faculty should shift from teaching mostly principles and ideas to mostly applications and practices. And in doing so, they should emphasize the case method, which simulates real-world management challenges and opportunities for students.

To help educators facilitate this shift and help students get the most out of case-based learning, we have developed a framework for analyzing cases. We call it PACADI (Problem, Alternatives, Criteria, Analysis, Decision, Implementation); it can improve learning outcomes by helping students better solve and analyze business problems, make decisions, and develop and implement strategy. Here, we’ll explain why we developed this framework, how it works, and what makes it an effective learning tool.

The Case for Cases: Helping Students Think Critically

Business students must develop critical-thinking and analytical skills, which are essential to their ability to make good decisions in functional areas such as marketing, finance, operations, and information technology, as well as to understand the relationships among these functions. For example, the decisions a marketing manager must make include strategic planning (segments, products, and channels); execution (digital messaging, media, branding, budgets, and pricing); and operations (integrated communications and technologies), as well as how to implement decisions across functional areas.

Faculty can use many types of cases to help students develop these skills. These include the prototypical “paper cases”; live cases , which feature guest lecturers such as entrepreneurs or corporate leaders and on-site visits; and multimedia cases , which immerse students into real situations. Most cases feature an explicit or implicit decision that a protagonist—whether it is an individual, a group, or an organization—must make.

For students new to learning by the case method—and even for those with case experience—some common issues can emerge; these issues can sometimes be a barrier for educators looking to ensure the best possible outcomes in their case classrooms. Unsure of how to dig into case analysis on their own, students may turn to the internet or rely on former students for “answers” to assigned cases. Or, when assigned to provide answers to assignment questions in teams, students might take a divide-and-conquer approach but not take the time to regroup and provide answers that are consistent with one other.

To help address these issues, which we commonly experienced in our classes, we wanted to provide our students with a more structured approach for how they analyze cases—and to really think about making decisions from the protagonists’ point of view. We developed the PACADI framework to address this need.

PACADI: A Six-Step Decision-Making Approach

The PACADI framework is a six-step decision-making approach that can be used in lieu of traditional end-of-case questions. It offers a structured, integrated, and iterative process that requires students to analyze case information, apply business concepts to derive valuable insights, and develop recommendations based on these insights.

Prior to beginning a PACADI assessment, which we’ll outline here, students should first prepare a two-paragraph summary—a situation analysis—that highlights the key case facts. Then, we task students with providing a five-page PACADI case analysis (excluding appendices) based on the following six steps.

Step 1: Problem definition. What is the major challenge, problem, opportunity, or decision that has to be made? If there is more than one problem, choose the most important one. Often when solving the key problem, other issues will surface and be addressed. The problem statement may be framed as a question; for example, How can brand X improve market share among millennials in Canada? Usually the problem statement has to be re-written several times during the analysis of a case as students peel back the layers of symptoms or causation.

Step 2: Alternatives. Identify in detail the strategic alternatives to address the problem; three to five options generally work best. Alternatives should be mutually exclusive, realistic, creative, and feasible given the constraints of the situation. Doing nothing or delaying the decision to a later date are not considered acceptable alternatives.

Step 3: Criteria. What are the key decision criteria that will guide decision-making? In a marketing course, for example, these may include relevant marketing criteria such as segmentation, positioning, advertising and sales, distribution, and pricing. Financial criteria useful in evaluating the alternatives should be included—for example, income statement variables, customer lifetime value, payback, etc. Students must discuss their rationale for selecting the decision criteria and the weights and importance for each factor.

Step 4: Analysis. Provide an in-depth analysis of each alternative based on the criteria chosen in step three. Decision tables using criteria as columns and alternatives as rows can be helpful. The pros and cons of the various choices as well as the short- and long-term implications of each may be evaluated. Best, worst, and most likely scenarios can also be insightful.

Step 5: Decision. Students propose their solution to the problem. This decision is justified based on an in-depth analysis. Explain why the recommendation made is the best fit for the criteria.

Step 6: Implementation plan. Sound business decisions may fail due to poor execution. To enhance the likeliness of a successful project outcome, students describe the key steps (activities) to implement the recommendation, timetable, projected costs, expected competitive reaction, success metrics, and risks in the plan.

“Students note that using the PACADI framework yields ‘aha moments’—they learned something surprising in the case that led them to think differently about the problem and their proposed solution.”

PACADI’s Benefits: Meaningfully and Thoughtfully Applying Business Concepts

The PACADI framework covers all of the major elements of business decision-making, including implementation, which is often overlooked. By stepping through the whole framework, students apply relevant business concepts and solve management problems via a systematic, comprehensive approach; they’re far less likely to surface piecemeal responses.

As students explore each part of the framework, they may realize that they need to make changes to a previous step. For instance, when working on implementation, students may realize that the alternative they selected cannot be executed or will not be profitable, and thus need to rethink their decision. Or, they may discover that the criteria need to be revised since the list of decision factors they identified is incomplete (for example, the factors may explain key marketing concerns but fail to address relevant financial considerations) or is unrealistic (for example, they suggest a 25 percent increase in revenues without proposing an increased promotional budget).

In addition, the PACADI framework can be used alongside quantitative assignments, in-class exercises, and business and management simulations. The structured, multi-step decision framework encourages careful and sequential analysis to solve business problems. Incorporating PACADI as an overarching decision-making method across different projects will ultimately help students achieve desired learning outcomes. As a practical “beyond-the-classroom” tool, the PACADI framework is not a contrived course assignment; it reflects the decision-making approach that managers, executives, and entrepreneurs exercise daily. Case analysis introduces students to the real-world process of making business decisions quickly and correctly, often with limited information. This framework supplies an organized and disciplined process that students can readily defend in writing and in class discussions.

PACADI in Action: An Example

Here’s an example of how students used the PACADI framework for a recent case analysis on CVS, a large North American drugstore chain.

The CVS Prescription for Customer Value*

PACADI Stage

Summary Response

How should CVS Health evolve from the “drugstore of your neighborhood” to the “drugstore of your future”?

Alternatives

A1. Kaizen (continuous improvement)

A2. Product development

A3. Market development

A4. Personalization (micro-targeting)

Criteria (include weights)

C1. Customer value: service, quality, image, and price (40%)

C2. Customer obsession (20%)

C3. Growth through related businesses (20%)

C4. Customer retention and customer lifetime value (20%)

Each alternative was analyzed by each criterion using a Customer Value Assessment Tool

Alternative 4 (A4): Personalization was selected. This is operationalized via: segmentation—move toward segment-of-1 marketing; geodemographics and lifestyle emphasis; predictive data analysis; relationship marketing; people, principles, and supply chain management; and exceptional customer service.

Implementation

Partner with leading medical school

Curbside pick-up

Pet pharmacy

E-newsletter for customers and employees

Employee incentive program

CVS beauty days

Expand to Latin America and Caribbean

Healthier/happier corner

Holiday toy drives/community outreach

*Source: A. Weinstein, Y. Rodriguez, K. Sims, R. Vergara, “The CVS Prescription for Superior Customer Value—A Case Study,” Back to the Future: Revisiting the Foundations of Marketing from Society for Marketing Advances, West Palm Beach, FL (November 2, 2018).

Results of Using the PACADI Framework

When faculty members at our respective institutions at Nova Southeastern University (NSU) and the University of North Carolina Wilmington have used the PACADI framework, our classes have been more structured and engaging. Students vigorously debate each element of their decision and note that this framework yields an “aha moment”—they learned something surprising in the case that led them to think differently about the problem and their proposed solution.

These lively discussions enhance individual and collective learning. As one external metric of this improvement, we have observed a 2.5 percent increase in student case grade performance at NSU since this framework was introduced.

Tips to Get Started

The PACADI approach works well in in-person, online, and hybrid courses. This is particularly important as more universities have moved to remote learning options. Because students have varied educational and cultural backgrounds, work experience, and familiarity with case analysis, we recommend that faculty members have students work on their first case using this new framework in small teams (two or three students). Additional analyses should then be solo efforts.

To use PACADI effectively in your classroom, we suggest the following:

Advise your students that your course will stress critical thinking and decision-making skills, not just course concepts and theory.

Use a varied mix of case studies. As marketing professors, we often address consumer and business markets; goods, services, and digital commerce; domestic and global business; and small and large companies in a single MBA course.

As a starting point, provide a short explanation (about 20 to 30 minutes) of the PACADI framework with a focus on the conceptual elements. You can deliver this face to face or through videoconferencing.

Give students an opportunity to practice the case analysis methodology via an ungraded sample case study. Designate groups of five to seven students to discuss the case and the six steps in breakout sessions (in class or via Zoom).

Ensure case analyses are weighted heavily as a grading component. We suggest 30–50 percent of the overall course grade.

Once cases are graded, debrief with the class on what they did right and areas needing improvement (30- to 40-minute in-person or Zoom session).

Encourage faculty teams that teach common courses to build appropriate instructional materials, grading rubrics, videos, sample cases, and teaching notes.

When selecting case studies, we have found that the best ones for PACADI analyses are about 15 pages long and revolve around a focal management decision. This length provides adequate depth yet is not protracted. Some of our tested and favorite marketing cases include Brand W , Hubspot , Kraft Foods Canada , TRSB(A) , and Whiskey & Cheddar .

Art Weinstein , Ph.D., is a professor of marketing at Nova Southeastern University, Fort Lauderdale, Florida. He has published more than 80 scholarly articles and papers and eight books on customer-focused marketing strategy. His latest book is Superior Customer Value—Finding and Keeping Customers in the Now Economy . Dr. Weinstein has consulted for many leading technology and service companies.

Herbert V. Brotspies , D.B.A., is an adjunct professor of marketing at Nova Southeastern University. He has over 30 years’ experience as a vice president in marketing, strategic planning, and acquisitions for Fortune 50 consumer products companies working in the United States and internationally. His research interests include return on marketing investment, consumer behavior, business-to-business strategy, and strategic planning.

John T. Gironda , Ph.D., is an assistant professor of marketing at the University of North Carolina Wilmington. His research has been published in Industrial Marketing Management, Psychology & Marketing , and Journal of Marketing Management . He has also presented at major marketing conferences including the American Marketing Association, Academy of Marketing Science, and Society for Marketing Advances.

Related Articles

We use cookies to understand how you use our site and to improve your experience, including personalizing content. Learn More . By continuing to use our site, you accept our use of cookies and revised Privacy Policy .

Organizing Your Social Sciences Research Assignments

• Annotated Bibliography
• Analyzing a Scholarly Journal Article
• Group Presentations
• Dealing with Nervousness
• Using Visual Aids
• Grading Someone Else's Paper
• Types of Structured Group Activities
• Group Project Survival Skills
• Leading a Class Discussion
• Multiple Book Review Essay
• Reviewing Collected Works
• Writing a Case Analysis Paper
• Writing a Case Study
• About Informed Consent
• Writing Field Notes
• Writing a Policy Memo
• Writing a Reflective Paper
• Writing a Research Proposal
• Generative AI and Writing
• Acknowledgments

A case study research paper examines a person, place, event, condition, phenomenon, or other type of subject of analysis in order to extrapolate  key themes and results that help predict future trends, illuminate previously hidden issues that can be applied to practice, and/or provide a means for understanding an important research problem with greater clarity. A case study research paper usually examines a single subject of analysis, but case study papers can also be designed as a comparative investigation that shows relationships between two or more subjects. The methods used to study a case can rest within a quantitative, qualitative, or mixed-method investigative paradigm.

Case Studies. Writing@CSU. Colorado State University; Mills, Albert J. , Gabrielle Durepos, and Eiden Wiebe, editors. Encyclopedia of Case Study Research . Thousand Oaks, CA: SAGE Publications, 2010 ; “What is a Case Study?” In Swanborn, Peter G. Case Study Research: What, Why and How? London: SAGE, 2010.

How to Approach Writing a Case Study Research Paper

General information about how to choose a topic to investigate can be found under the " Choosing a Research Problem " tab in the Organizing Your Social Sciences Research Paper writing guide. Review this page because it may help you identify a subject of analysis that can be investigated using a case study design.

However, identifying a case to investigate involves more than choosing the research problem . A case study encompasses a problem contextualized around the application of in-depth analysis, interpretation, and discussion, often resulting in specific recommendations for action or for improving existing conditions. As Seawright and Gerring note, practical considerations such as time and access to information can influence case selection, but these issues should not be the sole factors used in describing the methodological justification for identifying a particular case to study. Given this, selecting a case includes considering the following:

• The case represents an unusual or atypical example of a research problem that requires more in-depth analysis? Cases often represent a topic that rests on the fringes of prior investigations because the case may provide new ways of understanding the research problem. For example, if the research problem is to identify strategies to improve policies that support girl's access to secondary education in predominantly Muslim nations, you could consider using Azerbaijan as a case study rather than selecting a more obvious nation in the Middle East. Doing so may reveal important new insights into recommending how governments in other predominantly Muslim nations can formulate policies that support improved access to education for girls.
• The case provides important insight or illuminate a previously hidden problem? In-depth analysis of a case can be based on the hypothesis that the case study will reveal trends or issues that have not been exposed in prior research or will reveal new and important implications for practice. For example, anecdotal evidence may suggest drug use among homeless veterans is related to their patterns of travel throughout the day. Assuming prior studies have not looked at individual travel choices as a way to study access to illicit drug use, a case study that observes a homeless veteran could reveal how issues of personal mobility choices facilitate regular access to illicit drugs. Note that it is important to conduct a thorough literature review to ensure that your assumption about the need to reveal new insights or previously hidden problems is valid and evidence-based.
• The case challenges and offers a counter-point to prevailing assumptions? Over time, research on any given topic can fall into a trap of developing assumptions based on outdated studies that are still applied to new or changing conditions or the idea that something should simply be accepted as "common sense," even though the issue has not been thoroughly tested in current practice. A case study analysis may offer an opportunity to gather evidence that challenges prevailing assumptions about a research problem and provide a new set of recommendations applied to practice that have not been tested previously. For example, perhaps there has been a long practice among scholars to apply a particular theory in explaining the relationship between two subjects of analysis. Your case could challenge this assumption by applying an innovative theoretical framework [perhaps borrowed from another discipline] to explore whether this approach offers new ways of understanding the research problem. Taking a contrarian stance is one of the most important ways that new knowledge and understanding develops from existing literature.
• The case provides an opportunity to pursue action leading to the resolution of a problem? Another way to think about choosing a case to study is to consider how the results from investigating a particular case may result in findings that reveal ways in which to resolve an existing or emerging problem. For example, studying the case of an unforeseen incident, such as a fatal accident at a railroad crossing, can reveal hidden issues that could be applied to preventative measures that contribute to reducing the chance of accidents in the future. In this example, a case study investigating the accident could lead to a better understanding of where to strategically locate additional signals at other railroad crossings so as to better warn drivers of an approaching train, particularly when visibility is hindered by heavy rain, fog, or at night.
• The case offers a new direction in future research? A case study can be used as a tool for an exploratory investigation that highlights the need for further research about the problem. A case can be used when there are few studies that help predict an outcome or that establish a clear understanding about how best to proceed in addressing a problem. For example, after conducting a thorough literature review [very important!], you discover that little research exists showing the ways in which women contribute to promoting water conservation in rural communities of east central Africa. A case study of how women contribute to saving water in a rural village of Uganda can lay the foundation for understanding the need for more thorough research that documents how women in their roles as cooks and family caregivers think about water as a valuable resource within their community. This example of a case study could also point to the need for scholars to build new theoretical frameworks around the topic [e.g., applying feminist theories of work and family to the issue of water conservation].

Eisenhardt, Kathleen M. “Building Theories from Case Study Research.” Academy of Management Review 14 (October 1989): 532-550; Emmel, Nick. Sampling and Choosing Cases in Qualitative Research: A Realist Approach . Thousand Oaks, CA: SAGE Publications, 2013; Gerring, John. “What Is a Case Study and What Is It Good for?” American Political Science Review 98 (May 2004): 341-354; Mills, Albert J. , Gabrielle Durepos, and Eiden Wiebe, editors. Encyclopedia of Case Study Research . Thousand Oaks, CA: SAGE Publications, 2010; Seawright, Jason and John Gerring. "Case Selection Techniques in Case Study Research." Political Research Quarterly 61 (June 2008): 294-308.

Structure and Writing Style

The purpose of a paper in the social sciences designed around a case study is to thoroughly investigate a subject of analysis in order to reveal a new understanding about the research problem and, in so doing, contributing new knowledge to what is already known from previous studies. In applied social sciences disciplines [e.g., education, social work, public administration, etc.], case studies may also be used to reveal best practices, highlight key programs, or investigate interesting aspects of professional work.

In general, the structure of a case study research paper is not all that different from a standard college-level research paper. However, there are subtle differences you should be aware of. Here are the key elements to organizing and writing a case study research paper.

I.  Introduction

As with any research paper, your introduction should serve as a roadmap for your readers to ascertain the scope and purpose of your study . The introduction to a case study research paper, however, should not only describe the research problem and its significance, but you should also succinctly describe why the case is being used and how it relates to addressing the problem. The two elements should be linked. With this in mind, a good introduction answers these four questions:

• What is being studied? Describe the research problem and describe the subject of analysis [the case] you have chosen to address the problem. Explain how they are linked and what elements of the case will help to expand knowledge and understanding about the problem.
• Why is this topic important to investigate? Describe the significance of the research problem and state why a case study design and the subject of analysis that the paper is designed around is appropriate in addressing the problem.
• What did we know about this topic before I did this study? Provide background that helps lead the reader into the more in-depth literature review to follow. If applicable, summarize prior case study research applied to the research problem and why it fails to adequately address the problem. Describe why your case will be useful. If no prior case studies have been used to address the research problem, explain why you have selected this subject of analysis.
• How will this study advance new knowledge or new ways of understanding? Explain why your case study will be suitable in helping to expand knowledge and understanding about the research problem.

Each of these questions should be addressed in no more than a few paragraphs. Exceptions to this can be when you are addressing a complex research problem or subject of analysis that requires more in-depth background information.

II.  Literature Review

The literature review for a case study research paper is generally structured the same as it is for any college-level research paper. The difference, however, is that the literature review is focused on providing background information and  enabling historical interpretation of the subject of analysis in relation to the research problem the case is intended to address . This includes synthesizing studies that help to:

• Place relevant works in the context of their contribution to understanding the case study being investigated . This would involve summarizing studies that have used a similar subject of analysis to investigate the research problem. If there is literature using the same or a very similar case to study, you need to explain why duplicating past research is important [e.g., conditions have changed; prior studies were conducted long ago, etc.].
• Describe the relationship each work has to the others under consideration that informs the reader why this case is applicable . Your literature review should include a description of any works that support using the case to investigate the research problem and the underlying research questions.
• Identify new ways to interpret prior research using the case study . If applicable, review any research that has examined the research problem using a different research design. Explain how your use of a case study design may reveal new knowledge or a new perspective or that can redirect research in an important new direction.
• Resolve conflicts amongst seemingly contradictory previous studies . This refers to synthesizing any literature that points to unresolved issues of concern about the research problem and describing how the subject of analysis that forms the case study can help resolve these existing contradictions.
• Point the way in fulfilling a need for additional research . Your review should examine any literature that lays a foundation for understanding why your case study design and the subject of analysis around which you have designed your study may reveal a new way of approaching the research problem or offer a perspective that points to the need for additional research.
• Expose any gaps that exist in the literature that the case study could help to fill . Summarize any literature that not only shows how your subject of analysis contributes to understanding the research problem, but how your case contributes to a new way of understanding the problem that prior research has failed to do.
• Locate your own research within the context of existing literature [very important!] . Collectively, your literature review should always place your case study within the larger domain of prior research about the problem. The overarching purpose of reviewing pertinent literature in a case study paper is to demonstrate that you have thoroughly identified and synthesized prior studies in relation to explaining the relevance of the case in addressing the research problem.

III.  Method

In this section, you explain why you selected a particular case [i.e., subject of analysis] and the strategy you used to identify and ultimately decide that your case was appropriate in addressing the research problem. The way you describe the methods used varies depending on the type of subject of analysis that constitutes your case study.

If your subject of analysis is an incident or event . In the social and behavioral sciences, the event or incident that represents the case to be studied is usually bounded by time and place, with a clear beginning and end and with an identifiable location or position relative to its surroundings. The subject of analysis can be a rare or critical event or it can focus on a typical or regular event. The purpose of studying a rare event is to illuminate new ways of thinking about the broader research problem or to test a hypothesis. Critical incident case studies must describe the method by which you identified the event and explain the process by which you determined the validity of this case to inform broader perspectives about the research problem or to reveal new findings. However, the event does not have to be a rare or uniquely significant to support new thinking about the research problem or to challenge an existing hypothesis. For example, Walo, Bull, and Breen conducted a case study to identify and evaluate the direct and indirect economic benefits and costs of a local sports event in the City of Lismore, New South Wales, Australia. The purpose of their study was to provide new insights from measuring the impact of a typical local sports event that prior studies could not measure well because they focused on large "mega-events." Whether the event is rare or not, the methods section should include an explanation of the following characteristics of the event: a) when did it take place; b) what were the underlying circumstances leading to the event; and, c) what were the consequences of the event in relation to the research problem.

If your subject of analysis is a person. Explain why you selected this particular individual to be studied and describe what experiences they have had that provide an opportunity to advance new understandings about the research problem. Mention any background about this person which might help the reader understand the significance of their experiences that make them worthy of study. This includes describing the relationships this person has had with other people, institutions, and/or events that support using them as the subject for a case study research paper. It is particularly important to differentiate the person as the subject of analysis from others and to succinctly explain how the person relates to examining the research problem [e.g., why is one politician in a particular local election used to show an increase in voter turnout from any other candidate running in the election]. Note that these issues apply to a specific group of people used as a case study unit of analysis [e.g., a classroom of students].

If your subject of analysis is a place. In general, a case study that investigates a place suggests a subject of analysis that is unique or special in some way and that this uniqueness can be used to build new understanding or knowledge about the research problem. A case study of a place must not only describe its various attributes relevant to the research problem [e.g., physical, social, historical, cultural, economic, political], but you must state the method by which you determined that this place will illuminate new understandings about the research problem. It is also important to articulate why a particular place as the case for study is being used if similar places also exist [i.e., if you are studying patterns of homeless encampments of veterans in open spaces, explain why you are studying Echo Park in Los Angeles rather than Griffith Park?]. If applicable, describe what type of human activity involving this place makes it a good choice to study [e.g., prior research suggests Echo Park has more homeless veterans].

If your subject of analysis is a phenomenon. A phenomenon refers to a fact, occurrence, or circumstance that can be studied or observed but with the cause or explanation to be in question. In this sense, a phenomenon that forms your subject of analysis can encompass anything that can be observed or presumed to exist but is not fully understood. In the social and behavioral sciences, the case usually focuses on human interaction within a complex physical, social, economic, cultural, or political system. For example, the phenomenon could be the observation that many vehicles used by ISIS fighters are small trucks with English language advertisements on them. The research problem could be that ISIS fighters are difficult to combat because they are highly mobile. The research questions could be how and by what means are these vehicles used by ISIS being supplied to the militants and how might supply lines to these vehicles be cut off? How might knowing the suppliers of these trucks reveal larger networks of collaborators and financial support? A case study of a phenomenon most often encompasses an in-depth analysis of a cause and effect that is grounded in an interactive relationship between people and their environment in some way.

NOTE:   The choice of the case or set of cases to study cannot appear random. Evidence that supports the method by which you identified and chose your subject of analysis should clearly support investigation of the research problem and linked to key findings from your literature review. Be sure to cite any studies that helped you determine that the case you chose was appropriate for examining the problem.

IV.  Discussion

The main elements of your discussion section are generally the same as any research paper, but centered around interpreting and drawing conclusions about the key findings from your analysis of the case study. Note that a general social sciences research paper may contain a separate section to report findings. However, in a paper designed around a case study, it is common to combine a description of the results with the discussion about their implications. The objectives of your discussion section should include the following:

Reiterate the Research Problem/State the Major Findings Briefly reiterate the research problem you are investigating and explain why the subject of analysis around which you designed the case study were used. You should then describe the findings revealed from your study of the case using direct, declarative, and succinct proclamation of the study results. Highlight any findings that were unexpected or especially profound.

Explain the Meaning of the Findings and Why They are Important Systematically explain the meaning of your case study findings and why you believe they are important. Begin this part of the section by repeating what you consider to be your most important or surprising finding first, then systematically review each finding. Be sure to thoroughly extrapolate what your analysis of the case can tell the reader about situations or conditions beyond the actual case that was studied while, at the same time, being careful not to misconstrue or conflate a finding that undermines the external validity of your conclusions.

Relate the Findings to Similar Studies No study in the social sciences is so novel or possesses such a restricted focus that it has absolutely no relation to previously published research. The discussion section should relate your case study results to those found in other studies, particularly if questions raised from prior studies served as the motivation for choosing your subject of analysis. This is important because comparing and contrasting the findings of other studies helps support the overall importance of your results and it highlights how and in what ways your case study design and the subject of analysis differs from prior research about the topic.

Consider Alternative Explanations of the Findings Remember that the purpose of social science research is to discover and not to prove. When writing the discussion section, you should carefully consider all possible explanations revealed by the case study results, rather than just those that fit your hypothesis or prior assumptions and biases. Be alert to what the in-depth analysis of the case may reveal about the research problem, including offering a contrarian perspective to what scholars have stated in prior research if that is how the findings can be interpreted from your case.

Acknowledge the Study's Limitations You can state the study's limitations in the conclusion section of your paper but describing the limitations of your subject of analysis in the discussion section provides an opportunity to identify the limitations and explain why they are not significant. This part of the discussion section should also note any unanswered questions or issues your case study could not address. More detailed information about how to document any limitations to your research can be found here .

Suggest Areas for Further Research Although your case study may offer important insights about the research problem, there are likely additional questions related to the problem that remain unanswered or findings that unexpectedly revealed themselves as a result of your in-depth analysis of the case. Be sure that the recommendations for further research are linked to the research problem and that you explain why your recommendations are valid in other contexts and based on the original assumptions of your study.

V.  Conclusion

As with any research paper, you should summarize your conclusion in clear, simple language; emphasize how the findings from your case study differs from or supports prior research and why. Do not simply reiterate the discussion section. Provide a synthesis of key findings presented in the paper to show how these converge to address the research problem. If you haven't already done so in the discussion section, be sure to document the limitations of your case study and any need for further research.

The function of your paper's conclusion is to: 1) reiterate the main argument supported by the findings from your case study; 2) state clearly the context, background, and necessity of pursuing the research problem using a case study design in relation to an issue, controversy, or a gap found from reviewing the literature; and, 3) provide a place to persuasively and succinctly restate the significance of your research problem, given that the reader has now been presented with in-depth information about the topic.

Consider the following points to help ensure your conclusion is appropriate:

• If the argument or purpose of your paper is complex, you may need to summarize these points for your reader.
• If prior to your conclusion, you have not yet explained the significance of your findings or if you are proceeding inductively, use the conclusion of your paper to describe your main points and explain their significance.
• Move from a detailed to a general level of consideration of the case study's findings that returns the topic to the context provided by the introduction or within a new context that emerges from your case study findings.

Note that, depending on the discipline you are writing in or the preferences of your professor, the concluding paragraph may contain your final reflections on the evidence presented as it applies to practice or on the essay's central research problem. However, the nature of being introspective about the subject of analysis you have investigated will depend on whether you are explicitly asked to express your observations in this way.

Problems to Avoid

Overgeneralization One of the goals of a case study is to lay a foundation for understanding broader trends and issues applied to similar circumstances. However, be careful when drawing conclusions from your case study. They must be evidence-based and grounded in the results of the study; otherwise, it is merely speculation. Looking at a prior example, it would be incorrect to state that a factor in improving girls access to education in Azerbaijan and the policy implications this may have for improving access in other Muslim nations is due to girls access to social media if there is no documentary evidence from your case study to indicate this. There may be anecdotal evidence that retention rates were better for girls who were engaged with social media, but this observation would only point to the need for further research and would not be a definitive finding if this was not a part of your original research agenda.

Failure to Document Limitations No case is going to reveal all that needs to be understood about a research problem. Therefore, just as you have to clearly state the limitations of a general research study , you must describe the specific limitations inherent in the subject of analysis. For example, the case of studying how women conceptualize the need for water conservation in a village in Uganda could have limited application in other cultural contexts or in areas where fresh water from rivers or lakes is plentiful and, therefore, conservation is understood more in terms of managing access rather than preserving access to a scarce resource.

Failure to Extrapolate All Possible Implications Just as you don't want to over-generalize from your case study findings, you also have to be thorough in the consideration of all possible outcomes or recommendations derived from your findings. If you do not, your reader may question the validity of your analysis, particularly if you failed to document an obvious outcome from your case study research. For example, in the case of studying the accident at the railroad crossing to evaluate where and what types of warning signals should be located, you failed to take into consideration speed limit signage as well as warning signals. When designing your case study, be sure you have thoroughly addressed all aspects of the problem and do not leave gaps in your analysis that leave the reader questioning the results.

Case Studies. Writing@CSU. Colorado State University; Gerring, John. Case Study Research: Principles and Practices . New York: Cambridge University Press, 2007; Merriam, Sharan B. Qualitative Research and Case Study Applications in Education . Rev. ed. San Francisco, CA: Jossey-Bass, 1998; Miller, Lisa L. “The Use of Case Studies in Law and Social Science Research.” Annual Review of Law and Social Science 14 (2018): TBD; Mills, Albert J., Gabrielle Durepos, and Eiden Wiebe, editors. Encyclopedia of Case Study Research . Thousand Oaks, CA: SAGE Publications, 2010; Putney, LeAnn Grogan. "Case Study." In Encyclopedia of Research Design , Neil J. Salkind, editor. (Thousand Oaks, CA: SAGE Publications, 2010), pp. 116-120; Simons, Helen. Case Study Research in Practice . London: SAGE Publications, 2009;  Kratochwill,  Thomas R. and Joel R. Levin, editors. Single-Case Research Design and Analysis: New Development for Psychology and Education .  Hilldsale, NJ: Lawrence Erlbaum Associates, 1992; Swanborn, Peter G. Case Study Research: What, Why and How? London : SAGE, 2010; Yin, Robert K. Case Study Research: Design and Methods . 6th edition. Los Angeles, CA, SAGE Publications, 2014; Walo, Maree, Adrian Bull, and Helen Breen. “Achieving Economic Benefits at Local Events: A Case Study of a Local Sports Event.” Festival Management and Event Tourism 4 (1996): 95-106.

Writing Tip

At Least Five Misconceptions about Case Study Research

Social science case studies are often perceived as limited in their ability to create new knowledge because they are not randomly selected and findings cannot be generalized to larger populations. Flyvbjerg examines five misunderstandings about case study research and systematically "corrects" each one. To quote, these are:

Misunderstanding 1 :  General, theoretical [context-independent] knowledge is more valuable than concrete, practical [context-dependent] knowledge. Misunderstanding 2 :  One cannot generalize on the basis of an individual case; therefore, the case study cannot contribute to scientific development. Misunderstanding 3 :  The case study is most useful for generating hypotheses; that is, in the first stage of a total research process, whereas other methods are more suitable for hypotheses testing and theory building. Misunderstanding 4 :  The case study contains a bias toward verification, that is, a tendency to confirm the researcher’s preconceived notions. Misunderstanding 5 :  It is often difficult to summarize and develop general propositions and theories on the basis of specific case studies [p. 221].

While writing your paper, think introspectively about how you addressed these misconceptions because to do so can help you strengthen the validity and reliability of your research by clarifying issues of case selection, the testing and challenging of existing assumptions, the interpretation of key findings, and the summation of case outcomes. Think of a case study research paper as a complete, in-depth narrative about the specific properties and key characteristics of your subject of analysis applied to the research problem.

Flyvbjerg, Bent. “Five Misunderstandings About Case-Study Research.” Qualitative Inquiry 12 (April 2006): 219-245.

• << Previous: Writing a Case Analysis Paper
• Next: Writing a Field Report >>
• Last Updated: May 7, 2024 9:45 AM
• URL: https://libguides.usc.edu/writingguide/assignments

• Bihar Board

RBSE Result 2024

Srm university.

• Goa Board Result 2024
• Maharashtra HSC Result
• Maharashtra SSC Result
• RBSE 10th Result 2024
• RBSE 12th Result 2024
• CBSE Board Result 2024
• Shiv Khera Special
• Education News
• Web Stories
• Current Affairs
• नए भारत का नया उत्तर प्रदेश
• School & Boards
• College Admission
• Govt Jobs Alert & Prep
• GK & Aptitude
• CBSE Class 10 Study Material

CBSE Class 10 Maths Case Study Questions for Chapter 3 - Pair of Linear Equations in Two Variables (Published by CBSE)

Cbse's question bank on case study for class 10 maths chapter 3 is available here. these questions will be very helpful to prepare for the cbse class 10 maths exam 2022..

Case study questions are going to be new for CBSE Class 10 students. These are the competency-based questions that are completely new to class 10 students. To help students understand the format of the questions, CBSE has released a question bank on case study for class 10 Maths. Students must practice with these questions to get familiarised with the concepts and logic used in the case study and understand how to answers them correctly. You may check below the case study questions for CBSE Class 10 Maths Chapter 3 - Pair of Linear Equations in Two Variables. You can also check the right answer at the end of each question.

Check Case Study Questions for Class 10 Maths Chapter 3 - Pair of Linear Equations in Two Variables

CASE STUDY-1:

1. If answer to all questions he attempted by guessing were wrong, then how many questions did he answer correctly?

2. How many questions did he guess?

3. If answer to all questions he attempted by guessing were wrong and answered 80 correctly, then how many marks he got?

4. If answer to all questions he attempted by guessing were wrong, then how many questions answered correctly to score 95 marks?

Let the no of questions whose answer is known to the student x and questions attempted by cheating be y

x – 1/4y =90

solving these two

x = 96 and y = 24

1. He answered 96 questions correctly.

2. He attempted 24 questions by guessing.

3. Marks = 80- ¼ 0f 40 =70

4. x – 1/4 of (120 – x) = 95

5x = 500, x = 100

CASE STUDY-2:

Amit is planning to buy a house and the layout is given below. The design and the measurement has been made such that areas of two bedrooms and kitchen together is 95 sq.m.

Based on the above information, answer the following questions:

1. Form the pair of linear equations in two variables from this situation.

2. Find the length of the outer boundary of the layout.

3. Find the area of each bedroom and kitchen in the layout.

4. Find the area of living room in the layout.

5. Find the cost of laying tiles in kitchen at the rate of Rs. 50 per sq.m.

1. Area of two bedrooms= 10x sq.m

Area of kitchen = 5y sq.m

10x + 5y = 95

Also, x + 2+ y = 15

2. Length of outer boundary = 12 + 15 + 12 + 15 = 54m

3. On solving two equation part(i)

x = 6m and y = 7m

area of bedroom = 5 x 6 = 30m

area of kitchen = 5 x 7 = 35m

4. Area of living room = (15 x 7) – 30 = 105 – 30 = 75 sq.m

5. Total cost of laying tiles in the kitchen = Rs50 x 35 = Rs1750

Case study-3 :

It is common that Governments revise travel fares from time to time based on various factors such as inflation ( a general increase in prices and fall in the purchasing value of money) on different types of vehicles like auto, Rickshaws, taxis, Radio cab etc. The auto charges in a city comprise of a fixed charge together with the charge for the distance covered. Study the following situations:

Situation 1: In city A, for a journey of 10 km, the charge paid is Rs 75 and for a journey of 15 km, the charge paid is Rs 110.

Situation 2: In a city B, for a journey of 8km, the charge paid is Rs91 and for a journey of 14km, the charge paid is Rs 145.

Refer situation 1

1. If the fixed charges of auto rickshaw be Rs x and the running charges be Rs y km/hr, the pair of linear equations representing the situation is

a) x + 10y =110, x + 15y = 75

b) x + 10y = 75, x + 15y = 110

c) 10x + y = 110, 15x + y = 75

d) 10x + y = 75, 15x + y = 110

Answer: b) x + 10y = 75, x + 15y = 110

2. A person travels a distance of 50km. The amount he has to pay is

Answer: c) Rs.355

Refer situation 2

3. What will a person have to pay for travelling a distance of 30km?

Answer: b) Rs.289

4. The graph of lines representing the conditions are: (situation 2)

Answer: (iii)

Also Check:

CBSE Case Study Questions for Class 10 Maths - All Chapters

Tips to Solve Case Study Based Questions Accurately

Get here latest School , CBSE and Govt Jobs notification in English and Hindi for Sarkari Naukari and Sarkari Result . Download the Jagran Josh Sarkari Naukri App . Check  Board Result 2024  for Class 10 and Class 12 like  CBSE Board Result ,  UP Board Result ,  Bihar Board Result ,  MP Board Result ,  Rajasthan Board Result  and Other States Boards.

• JAC 9th, 11th रिजल्ट 2024
• JAC 9th रिजल्ट 2024
• JAC Result 2024
• jac.jharkhand.gov.in 11th Result 2024
• JAC 11वीं रिजल्ट 2024 रोल नंबर
• GBSHSE SSC Result 2024
• SSC Result 2024 Goa
• CDS 2 Notification 2024
• results.gbshsegoa.net Result 2024
• CBSE 10th Result 2024
• CBSE Study Material
• CBSE Class 10

Latest Education News

BTEUP Result 2024 on bteup.ac.in: Check Odd Semester Scrutiny/Reevaluation Result Updates

[रिजल्ट लिंक] JAC 9th Result 2024 OUT: कोडरमा जिला रहा अव्वल, jac.jharkhand.gov.in पर इस Link से तुरंत डाउनलोड करें अपनी Marksheet

PSEB Class 11 Computer Science Syllabus 2024-2025: Download in PDF

PSEB 12th Economics Syllabus 2025: Download Punjab Board Class 12 Economics Syllabus 2024-2025 PDF Here

CBSE Class 12 2024 Marks Verification Application Begins, Check Important Dates And Fees Here

South Africa T20 World Cup Squad 2024: Complete List of Team Players and Name

Afghanistan T20 World Cup Squad 2024: Complete List of Team Players and Name

ICC T20 World Cup 2024: T20 वर्ल्ड कप का शेड्यूल जारी, कब और किससे है भारत का मैच देखें यहां

Bangladesh T20 World Cup Squad 2024: Complete List of Team Players and Name

KEAM 2024 Application Correction Date Extended To May 18, Check Steps To Edit And Important Dates Here

Pakistan T20 World Cup Squad 2024: पाक की विश्व कप टीम में किसे मिला मौका कौन हुआ बाहर देखें यहां

UPSC ESIC Nursing Officer Previous Year Question Paper: PDF Download

HBSE Class 12th English Syllabus 2024-25: Download PDF for Board Exam

JEE Advanced 2024: Most Sought BTech Courses in IITs

Odisha High Court Recruitment 2024: Applications Invited For Junior Stenographer Posts, Check Eligibility And Application Process

[लिंक एक्टिव] JAC 9th, 11th Result 2024 OUT: झारखंड बोर्ड के नतीजे 9वीं और 11वीं बोर्ड परीक्षा के नतीजे jacresults.com पर जारी

SSC GD 2024 Result Live Update: Official Website to Download the Constable Results ssc.gov.in, Check Merit List Date, Latest Updates

UPSC NDA 2 Eligibility Criteria 2024: Age Limit, Qualification and Nationality

CUET UG Answer Key May 17, 2024: Download Set Wise Answer Sheet PDF

CUET UG Question Paper 2024, May 17: Download Question Paper PDF (SET A, B, C, D)

The Site is down as we are performing important server maintenance, during which time the server will be unavailable for approximately 24 hours. Please hold off on any critical actions until we are finished. As always your feedback is appreciated.

• Study Packages
• NCERT Solutions
• Sample Papers
• Online Test

• Questions Bank
• Mathematics
• Test Series
• Ncert Solutions
• Solved Papers
• Current Affairs
• General Knowledge
• Mental Ability
• Computers Science

Question Bank for 3rd Class Mathematics

Select chapter, recent question banks, data handling, practice now, geometrical shapes, measurement, time and calendar, multiplication and division, addition and subtraction, number sense and numeration, logical reasoning, length, weight, quantity, time and temperature, computation operations, number sense, pictographs.

Reset Password.

OTP has been sent to your mobile number and is valid for one hour

Mobile Number Verified

Your mobile number is verified.

• Bipolar Disorder
• Therapy Center
• When To See a Therapist
• Types of Therapy
• Best Online Therapy
• Best Couples Therapy
• Best Family Therapy
• Managing Stress
• Sleep and Dreaming
• Understanding Emotions
• Self-Improvement
• Healthy Relationships
• Student Resources
• Personality Types
• Guided Meditations
• Verywell Mind Insights
• 2024 Verywell Mind 25
• Mental Health in the Classroom
• Editorial Process
• Meet Our Review Board
• Crisis Support

What Is a Case Study?

Weighing the pros and cons of this method of research

Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

Cara Lustik is a fact-checker and copywriter.

Verywell / Colleen Tighe

• Pros and Cons

What Types of Case Studies Are Out There?

Where do you find data for a case study, how do i write a psychology case study.

A case study is an in-depth study of one person, group, or event. In a case study, nearly every aspect of the subject's life and history is analyzed to seek patterns and causes of behavior. Case studies can be used in many different fields, including psychology, medicine, education, anthropology, political science, and social work.

The point of a case study is to learn as much as possible about an individual or group so that the information can be generalized to many others. Unfortunately, case studies tend to be highly subjective, and it is sometimes difficult to generalize results to a larger population.

While case studies focus on a single individual or group, they follow a format similar to other types of psychology writing. If you are writing a case study, we got you—here are some rules of APA format to reference.  

At a Glance

A case study, or an in-depth study of a person, group, or event, can be a useful research tool when used wisely. In many cases, case studies are best used in situations where it would be difficult or impossible for you to conduct an experiment. They are helpful for looking at unique situations and allow researchers to gather a lot of˜ information about a specific individual or group of people. However, it's important to be cautious of any bias we draw from them as they are highly subjective.

What Are the Benefits and Limitations of Case Studies?

A case study can have its strengths and weaknesses. Researchers must consider these pros and cons before deciding if this type of study is appropriate for their needs.

One of the greatest advantages of a case study is that it allows researchers to investigate things that are often difficult or impossible to replicate in a lab. Some other benefits of a case study:

• Allows researchers to capture information on the 'how,' 'what,' and 'why,' of something that's implemented
• Gives researchers the chance to collect information on why one strategy might be chosen over another
• Permits researchers to develop hypotheses that can be explored in experimental research

On the other hand, a case study can have some drawbacks:

• It cannot necessarily be generalized to the larger population
• Cannot demonstrate cause and effect
• It may not be scientifically rigorous
• It can lead to bias

Researchers may choose to perform a case study if they want to explore a unique or recently discovered phenomenon. Through their insights, researchers develop additional ideas and study questions that might be explored in future studies.

It's important to remember that the insights from case studies cannot be used to determine cause-and-effect relationships between variables. However, case studies may be used to develop hypotheses that can then be addressed in experimental research.

Case Study Examples

There have been a number of notable case studies in the history of psychology. Much of  Freud's work and theories were developed through individual case studies. Some great examples of case studies in psychology include:

• Anna O : Anna O. was a pseudonym of a woman named Bertha Pappenheim, a patient of a physician named Josef Breuer. While she was never a patient of Freud's, Freud and Breuer discussed her case extensively. The woman was experiencing symptoms of a condition that was then known as hysteria and found that talking about her problems helped relieve her symptoms. Her case played an important part in the development of talk therapy as an approach to mental health treatment.
• Phineas Gage : Phineas Gage was a railroad employee who experienced a terrible accident in which an explosion sent a metal rod through his skull, damaging important portions of his brain. Gage recovered from his accident but was left with serious changes in both personality and behavior.
• Genie : Genie was a young girl subjected to horrific abuse and isolation. The case study of Genie allowed researchers to study whether language learning was possible, even after missing critical periods for language development. Her case also served as an example of how scientific research may interfere with treatment and lead to further abuse of vulnerable individuals.

Such cases demonstrate how case research can be used to study things that researchers could not replicate in experimental settings. In Genie's case, her horrific abuse denied her the opportunity to learn a language at critical points in her development.

This is clearly not something researchers could ethically replicate, but conducting a case study on Genie allowed researchers to study phenomena that are otherwise impossible to reproduce.

There are a few different types of case studies that psychologists and other researchers might use:

• Collective case studies : These involve studying a group of individuals. Researchers might study a group of people in a certain setting or look at an entire community. For example, psychologists might explore how access to resources in a community has affected the collective mental well-being of those who live there.
• Descriptive case studies : These involve starting with a descriptive theory. The subjects are then observed, and the information gathered is compared to the pre-existing theory.
• Explanatory case studies : These   are often used to do causal investigations. In other words, researchers are interested in looking at factors that may have caused certain things to occur.
• Exploratory case studies : These are sometimes used as a prelude to further, more in-depth research. This allows researchers to gather more information before developing their research questions and hypotheses .
• Instrumental case studies : These occur when the individual or group allows researchers to understand more than what is initially obvious to observers.
• Intrinsic case studies : This type of case study is when the researcher has a personal interest in the case. Jean Piaget's observations of his own children are good examples of how an intrinsic case study can contribute to the development of a psychological theory.

The three main case study types often used are intrinsic, instrumental, and collective. Intrinsic case studies are useful for learning about unique cases. Instrumental case studies help look at an individual to learn more about a broader issue. A collective case study can be useful for looking at several cases simultaneously.

The type of case study that psychology researchers use depends on the unique characteristics of the situation and the case itself.

There are a number of different sources and methods that researchers can use to gather information about an individual or group. Six major sources that have been identified by researchers are:

• Archival records : Census records, survey records, and name lists are examples of archival records.
• Direct observation : This strategy involves observing the subject, often in a natural setting . While an individual observer is sometimes used, it is more common to utilize a group of observers.
• Documents : Letters, newspaper articles, administrative records, etc., are the types of documents often used as sources.
• Interviews : Interviews are one of the most important methods for gathering information in case studies. An interview can involve structured survey questions or more open-ended questions.
• Participant observation : When the researcher serves as a participant in events and observes the actions and outcomes, it is called participant observation.
• Physical artifacts : Tools, objects, instruments, and other artifacts are often observed during a direct observation of the subject.

If you have been directed to write a case study for a psychology course, be sure to check with your instructor for any specific guidelines you need to follow. If you are writing your case study for a professional publication, check with the publisher for their specific guidelines for submitting a case study.

Here is a general outline of what should be included in a case study.

Section 1: A Case History

This section will have the following structure and content:

Background information : The first section of your paper will present your client's background. Include factors such as age, gender, work, health status, family mental health history, family and social relationships, drug and alcohol history, life difficulties, goals, and coping skills and weaknesses.

Description of the presenting problem : In the next section of your case study, you will describe the problem or symptoms that the client presented with.

Describe any physical, emotional, or sensory symptoms reported by the client. Thoughts, feelings, and perceptions related to the symptoms should also be noted. Any screening or diagnostic assessments that are used should also be described in detail and all scores reported.

Your diagnosis : Provide your diagnosis and give the appropriate Diagnostic and Statistical Manual code. Explain how you reached your diagnosis, how the client's symptoms fit the diagnostic criteria for the disorder(s), or any possible difficulties in reaching a diagnosis.

Section 2: Treatment Plan

This portion of the paper will address the chosen treatment for the condition. This might also include the theoretical basis for the chosen treatment or any other evidence that might exist to support why this approach was chosen.

• Cognitive behavioral approach : Explain how a cognitive behavioral therapist would approach treatment. Offer background information on cognitive behavioral therapy and describe the treatment sessions, client response, and outcome of this type of treatment. Make note of any difficulties or successes encountered by your client during treatment.
• Humanistic approach : Describe a humanistic approach that could be used to treat your client, such as client-centered therapy . Provide information on the type of treatment you chose, the client's reaction to the treatment, and the end result of this approach. Explain why the treatment was successful or unsuccessful.
• Psychoanalytic approach : Describe how a psychoanalytic therapist would view the client's problem. Provide some background on the psychoanalytic approach and cite relevant references. Explain how psychoanalytic therapy would be used to treat the client, how the client would respond to therapy, and the effectiveness of this treatment approach.
• Pharmacological approach : If treatment primarily involves the use of medications, explain which medications were used and why. Provide background on the effectiveness of these medications and how monotherapy may compare with an approach that combines medications with therapy or other treatments.

This section of a case study should also include information about the treatment goals, process, and outcomes.

When you are writing a case study, you should also include a section where you discuss the case study itself, including the strengths and limitiations of the study. You should note how the findings of your case study might support previous research. 

In your discussion section, you should also describe some of the implications of your case study. What ideas or findings might require further exploration? How might researchers go about exploring some of these questions in additional studies?

Need More Tips?

Here are a few additional pointers to keep in mind when formatting your case study:

• Never refer to the subject of your case study as "the client." Instead, use their name or a pseudonym.
• Read examples of case studies to gain an idea about the style and format.
• Remember to use APA format when citing references .

Crowe S, Cresswell K, Robertson A, Huby G, Avery A, Sheikh A. The case study approach .  BMC Med Res Methodol . 2011;11:100.

Crowe S, Cresswell K, Robertson A, Huby G, Avery A, Sheikh A. The case study approach . BMC Med Res Methodol . 2011 Jun 27;11:100. doi:10.1186/1471-2288-11-100

Gagnon, Yves-Chantal.  The Case Study as Research Method: A Practical Handbook . Canada, Chicago Review Press Incorporated DBA Independent Pub Group, 2010.

Yin, Robert K. Case Study Research and Applications: Design and Methods . United States, SAGE Publications, 2017.

By Kendra Cherry, MSEd Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

Gurukul of Excellence

Classes for Physics, Chemistry and Mathematics by IITians

Join our Telegram Channel for Free PDF Download

Case Study Questions for Class 12 Chemistry Chapter 3 Electrochemistry

• Last modified on: 3 years ago
• Reading Time: 5 Minutes

There is Case Study Questions in class 12 Chemistry in session 2020-21. For the first time, the board has introduced the case study questions in the board exam. The first two questions in the board exam question paper will be based on Case Study and Assertion & Reason. The first question will have 5 MCQs out of which students will have to attempt any 4 questions. The second question will carry 5 Assertion & Reason type questions with the choice to attempt any four. Here are the questions based on case study.

Case Study Question 1:

Read the passage given below and answer the following questions:

All chemical reactions involve interaction of atoms and molecules. A large number of atoms/molecules are present in a few gram of any chemical compound varying with their atomic/molecular masses. To handle such large number conveniently, the mole concept was introduced. All electrochemical cell reactions are also based on mole concept. For example, a 4.0 molar aqueous solution of NaCl is prepared and 500 mL of this solution is electrolysed. This leads to the evolution of chlorine gas at one of the electrode. The amount of products formed can be calculated by using mole concept.

The following questions are multiple choice questions. Choose the most appropriate answer:

(i) The total number of moles of chlorine gas evolved is (a) 0.5 (b) 1.0 (c) 1.5 (d) 1.9

(ii) If cathode is a Hg electrode, then the maximum weight of amalgam formed from this solution is (a) 300g (b) 446 g (c) 396 g (d) 256 g

The total charge (coulomb) required for complete electrolysis is (a) 186000 (b) 24125 (c) 48296 (d) 193000

(iii) In the electrolytes, the number of moles of electrons involved are (a) 2 (b) 1 (c) 3 (d) 4

(iv) In electrolysis of aqueous NaCl solution when Pt electrode is taken, then which gas is liberated at cathode? (a) H 2 gas (b) Cl 2 gas (c) O 2 gas (d) None of these

Download CBSE Books

Exam Special Series:

• Sample Question Paper for CBSE Class 10 Science (for 2024)
• Sample Question Paper for CBSE Class 10 Maths (for 2024)
• CBSE Most Repeated Questions for Class 10 Science Board Exams
• CBSE Important Diagram Based Questions Class 10 Physics Board Exams
• CBSE Important Numericals Class 10 Physics Board Exams
• CBSE Practical Based Questions for Class 10 Science Board Exams
• CBSE Important “Differentiate Between” Based Questions Class 10 Social Science
• Sample Question Papers for CBSE Class 12 Physics (for 2024)
• Sample Question Papers for CBSE Class 12 Chemistry (for 2024)
• Sample Question Papers for CBSE Class 12 Maths (for 2024)
• Sample Question Papers for CBSE Class 12 Biology (for 2024)
• CBSE Important Diagrams & Graphs Asked in Board Exams Class 12 Physics
• Master Organic Conversions CBSE Class 12 Chemistry Board Exams
• CBSE Important Numericals Class 12 Physics Board Exams
• CBSE Important Definitions Class 12 Physics Board Exams
• CBSE Important Laws & Principles Class 12 Physics Board Exams
• 10 Years CBSE Class 12 Chemistry Previous Year-Wise Solved Papers (2023-2024)
• 10 Years CBSE Class 12 Physics Previous Year-Wise Solved Papers (2023-2024)
• 10 Years CBSE Class 12 Maths Previous Year-Wise Solved Papers (2023-2024)
• 10 Years CBSE Class 12 Biology Previous Year-Wise Solved Papers (2023-2024)
• ICSE Important Numericals Class 10 Physics BOARD Exams (215 Numericals)
• ICSE Important Figure Based Questions Class 10 Physics BOARD Exams (230 Questions)
• ICSE Mole Concept and Stoichiometry Numericals Class 10 Chemistry (65 Numericals)
• ICSE Reasoning Based Questions Class 10 Chemistry BOARD Exams (150 Qs)
• ICSE Important Functions and Locations Based Questions Class 10 Biology
• ICSE Reasoning Based Questions Class 10 Biology BOARD Exams (100 Qs)

✨ Join our Online JEE Test Series for 499/- Only (Web + App) for 1 Year

✨ Join our Online NEET Test Series for 499/- Only for 1 Year

1 thought on “ Case Study Questions for Class 12 Chemistry Chapter 3 Electrochemistry ”

Leave a Reply Cancel reply

Join our Online Test Series for CBSE, ICSE, JEE, NEET and Other Exams

Editable Study Materials for Your Institute - CBSE, ICSE, State Boards (Maharashtra & Karnataka), JEE, NEET, FOUNDATION, OLYMPIADS, PPTs

Discover more from Gurukul of Excellence

Subscribe now to keep reading and get access to the full archive.

Type your email…

Continue reading

30,000+ students realised their study abroad dream with us. Take the first step today

Meet top uk universities from the comfort of your home, here’s your new year gift, one app for all your, study abroad needs, start your journey, track your progress, grow with the community and so much more.

Verification Code

An OTP has been sent to your registered mobile no. Please verify

Thanks for your comment !

Our team will review it before it's shown to our readers.

• School Education /

7 Interesting and Fun English Stories For Class 3 

• Updated on  
• May 15, 2024

Stories that are interesting and fun make learning great for kids! Along with enjoying themselves and having fun, they also learn. English stories are a great way to ensure the holistic development of children. They also pick up skills like recitation, pronunciation, confidence, etc. Hence, in this blog, we bring you 7 Interesting and Fun English Stories For Class 3. Here, you will find some great stories for kids. Let’s explore them now! 

Table of Contents

• 1 1. The Bear and the Two Friends
• 2 2. The Miser and His Gold
• 3 3. The Dog At the Well
• 4 4. Controlling Anger
• 5 5. The Leap at Rhodes
• 6 6. The Wolf and the Sheep 
• 7 7. The Tortoise And The Bird 
• 8 FAQs 

1. The Bear and the Two Friends

Also Read : Mindfulness for Kids: With 11 Exercises, Importance  

2. The Miser and His Gold

Also Read : 5 Tips to Improve Kid’s Vocabulary   

3. The Dog At the Well

Moral : Heed the warnings of those who are wiser. 

4. Controlling Anger

Also Read : 5 Best Fine Motor Activities for Preschoolers    

5. The Leap at Rhodes

6. the wolf and the sheep , 7. the tortoise and the bird .

Ans: Some of the best stories are mentioned below:  – The Golden Egg  – The Shepherd Boy and the Wolf  – Having a Best Friend: Friendship Moral Stories in English  – The King’s Painting  – The Pig and the Sheep 

Ans: “Cinderella” is one of the most famous stories for kids.

Ans: A short story is simply written material consisting of a plot that may be fictional and can be read in a short time span of, like, 5-10 minutes. 

Related Reads : 

Hope you like the English stories for class 3 that we have provided in this blog. For more such kids’ learning material, check out School Education and follow Leverage Edu!! 

Deepansh Gautam

Leave a Reply Cancel reply

Save my name, email, and website in this browser for the next time I comment.

Contact no. *

Connect With Us

30,000+ students realised their study abroad dream with us. Take the first step today.

Resend OTP in

Need help with?

Study abroad.

UK, Canada, US & More

IELTS, GRE, GMAT & More

Scholarship, Loans & Forex

Country Preference

New Zealand

Which English test are you planning to take?

Which academic test are you planning to take.

Not Sure yet

When are you planning to take the exam?

Already booked my exam slot

Within 2 Months

Want to learn about the test

Which Degree do you wish to pursue?

When do you want to start studying abroad.

January 2024

September 2024

What is your budget to study abroad?

How would you describe this article ?

Please rate this article

We would like to hear more.

Have something on your mind?

Make your study abroad dream a reality in January 2022 with

India's Biggest Virtual University Fair

Essex Direct Admission Day

Why attend .

Don't Miss Out

Class 9 Geography Case Study Questions of Chapter 3 Drainage

• Post author: studyrate
• Post published:
• Post category: class 9th
• Post comments: 0 Comments

Case study Questions in Class 9 Social Science Chapter 3  are very important to solve for your exam. Class 9 Social Science Chapter 10 Case Study Questions have been prepared for the latest exam pattern. You can check your knowledge by solving  case study-based questions for Class 9 Geography Case Study Questions Chapter 3 Drainage

Join our Telegram Channel, there you will get various e-books for CBSE 2024 Boards exams for Class 9th, 10th, 11th, and 12th.

In CBSE Class 9 Social Science Paper, Students will have to answer some questions based on Assertion and Reason. There will be a few questions based on case studies and passage-based as well. In that, a paragraph will be given, and then the MCQ questions based on it will be asked.

Drainage Case Study Questions With Answers

Here, we have provided case-based/passage-based questions for Class 9 Social Science Chapter 3 Drainage

Case Study 1: The drainage system of India is mainly controlled by the broad relief features of the subcontinent. Accordingly, the Indian rivers are divided into two major groups: the Himalayan rivers; and the Peninsular Rivers. Apart from originating from the two major physiographic regions of India, the Himalayan and the Peninsular rivers are different from each other in many ways. Most of the Himalayan rivers are perennial. It means that they have water throughout the year. These rivers receive water from rain as well as from melted snow from the loft mountains. The two major Himalayan rivers, the Indus and the Brahmaputra originate from the North of the mountain ranges. They have cut through the mountains making gorges. The Himalayan rivers have long courses from their source to the sea. They perform an intensive erosional activity in their upper courses from their source to the sea. They perform an intensive erosional activity in their upper courses and carry huge loads of silt and sand. In the middle and lower courses, these rivers form meanders, oxbow lakes, and many other depositional features in their flood plains. They also have well-developed deltas.

Which of the following is not the Himalayan river? (a) Godavari (b) Indus (c) Ganga (d) Brahmaputra

Answer: (a) Godavari

Which of the following is not the feature of the Himalayan rivers? (a) Some Himalayan rivers originate from the North of the mountain range. (b) The Himalayan rivers have shorter and shallower courses. (c) Brahmaputra is the example of Himalayan rivers. (d) None of the above

Answer: (b) The Himalayan rivers have shorter and shallower courses.

Why do some Himalayan rivers perform the intensive erosional activity? Identify the best suitable options. (a) Because they originates from high altitude. (b) These are small rivers. (c) These rivers are flows from West to East. (d) These river are non-perennial.

Answer: (a) Because they originates from high altitude.

Himalayan rivers has well developed deltas. Which among the following is the prominent cause? (a) Himalayan rivers have long courses from their source to sea. (b) They flows from mountain and carry huge loads of silt and sand. (c) They flow with high density of water. (d) All of the above

Answer: (d) All of the above

Which of the following is not the characteristics of Himalayan rivers? (a) This rivers formed deltas at their mouth. (b) The Himalayan rivers are short in length. (c) These rivers are seasonal. (d) All of the above

Answer: (a) This rivers formed deltas at their mouth.

Two statements are given in the question below as Assertion (A) and Reason (R). Read the statements and choose the appropriate option. Assertion (A) Peninsular river are perennial river. Reason (R) Perennial rivers receives water from rain as well as from melted snow from the lofty mountains. Codes (a) Both A and R are true and R is the correct explanation of A (b) Both A and R are true, but R is not the correct explanation of A (c) A is true, but R is false (d) A is false, but R is true

Answer: (d) A is false, but R is true

Case Study 2: The drainage system plays a crucial role in shaping the landscape and determining the flow of water in an area. In India, the drainage patterns are diverse and influenced by various factors such as topography, climate, and geological formations. The major drainage basins in India are the Indus, Ganga, and Brahmaputra. The Himalayan rivers have a snow-fed perennial source, resulting in the formation of large river systems. The Peninsular rivers, on the other hand, have a rain-fed source and exhibit seasonal variations in their water flow. The rivers in India not only provide water for irrigation, drinking, and industrial purposes but also serve as important transportation routes. However, the improper management of drainage systems can lead to issues such as floods, soil erosion, and water pollution.

What factors influence the drainage patterns in India? a) Political boundaries and population density b) Topography, climate, and geological formations c) Religious diversity and cultural practices d) Economic development and industrialization

Answer: b) Topography, climate, and geological formations

Which are the major drainage basins in India? a) Yamuna, Godavari, and Krishna b) Narmada, Mahanadi, and Tapti c) Indus, Ganga, and Brahmaputra d) Cauvery, Tungabhadra, and Pennar

Answer: c) Indus, Ganga, and Brahmaputra

What is the source of Himalayan rivers in India? a) Underground springs b) Lakes and reservoirs c) Rainfall and monsoons d) Snow-fed perennial glaciers

Answer: d) Snow-fed perennial glaciers

How do Peninsular rivers in India differ from Himalayan rivers? a) Peninsular rivers have a snow-fed source. b) Peninsular rivers are rain-fed and exhibit seasonal variations. c) Peninsular rivers flow through the Himalayan region. d) Peninsular rivers have a larger water volume.

Answer: b) Peninsular rivers are rain-fed and exhibit seasonal variations.

Besides water supply, what other functions do rivers serve in India? a) Formation of deltas and estuaries b) Generation of electricity through hydropower projects c) Creation of tourist attractions and scenic spots d) Development of recreational activities like boating and fishing

Answer: b) Generation of electricity through hydropower projects

Hope the information shed above regarding Case Study and Passage Based Questions for Class 9 Social Science Chapter 3 Drainage with Answers Pdf free download has been useful to an extent. If you have any other queries about CBSE Class 9 Social Science Drainage Case Study and Passage Based Questions with Answers, feel free to comment below so that we can revert back to us at the earliest possible By Team Study Rate

You Might Also Like

Cbse class 9 science term 1 mcq questions with answers pdf download, mcq questions of class 9 social science geography chapter 5 natural vegetation and wildlife with answers, class 9 science case study questions chapter 10 gravitation, leave a reply cancel reply.

Save my name, email, and website in this browser for the next time I comment.

• Open access
• Published: 14 May 2024

Quantitative evaluation of vertical control in orthodontic camouflage treatment for skeletal class II with hyperdivergent facial type

• Yan-Ning Guo 1 , 2   na1 ,
• Sheng-Jie Cui 1   na1 ,
• Jie-Ni Zhang 1 ,
• Yan-Heng Zhou 1 &
• Xue-Dong Wang 1  

Head & Face Medicine volume  20 , Article number:  31 ( 2024 ) Cite this article

Metrics details

In this study, we sought to quantify the influence of vertical control assisted by a temporary anchorage device (TAD) on orthodontic treatment efficacy for skeletal class II patients with a hyperdivergent facial type and probe into the critical factors of profile improvement.

A total of 36 adult patients with skeletal class II and a hyperdivergent facial type were included in this retrospective case–control study. To exclude the effect of sagittal anchorage reinforcement, the patients were divided into two groups: a maxillary maximum anchorage (MMA) group ( N  = 17), in which TADs were only used to help with anterior tooth retraction, and the MMA with vertical control (MMA + VC) group ( N  = 19), for which TADs were also used to intrude the maxillary molars and incisors. The treatment outcome was evaluated using dental, skeletal, and soft-tissue-related parameters via a cephalometric analysis and cast superimposition.

A significant decrease in ANB ( P  < 0.05 for both groups), the retraction and uprighting of the maxillary and mandibular incisors, and the retraction of protruded upper and lower lips were observed in both groups. Moreover, a significant intrusion of the maxillary molars was observed via the cephalometric analysis (− 1.56 ± 1.52 mm, P  < 0.05) and cast superimposition (− 2.25 ± 1.03 mm, P  < 0.05) of the MMA + VC group but not the MMA group, which resulted in a remarkable decrease in the mandibular plane angle (− 1.82 ± 1.38°, P  < 0.05). The Z angle (15.25 ± 5.30°, P  < 0.05) and Chin thickness (− 0.97 ± 0.45°, P  < 0.05) also improved dramatically in the MMA + VC group, indicating a better profile and a relaxed mentalis. Multivariate regression showed that the improvement in the soft tissue was closely related to the counterclockwise rotation of the mandible plane ( P  < 0.05).

Conclusions

TAD-assisted vertical control can achieve intrusion of approximately 2 mm for the upper first molars and induce mandibular counterclockwise rotation of approximately 1.8°. Moreover, it is especially important for patients without sufficient retraction of the upper incisors or a satisfactory chin shape.

Peer Review reports

For adult patients with severe class II malocclusion accompanied by a hyperdivergent growth pattern, orthognathic surgery is usually the optimal therapy to improve facial aesthetics and masticatory function [ 1 , 2 ]. Nevertheless, some patients refuse surgery due to its possible risks and high cost. Orthodontic camouflage treatment provides an alternative for such patients [ 3 , 4 ].

To improve the profile of this kind of patient, both sagittal retraction and vertical control are important. Several studies have found and confirmed the importance of vertical control in orthodontic treatment for skeletal class II malocclusion [ 5 , 6 , 7 ]. However, varying treatment methods are used. For adolescent patients, the most effective approach is often to utilize their vertical growth potential to guide their facial development in the desired direction. Jamilian et al. applied a modified functional orthodontic appliance to induce sagittal and vertical changes in the mandible, achieving significant facial improvement for a patient with severe skeletal class II [ 8 ].

On the other hand, for adult patients lacking growth potential, active intrusion of posterior teeth is required to intervene vertically. Early on, high-pull headgear was the most common vertical control method, but this approach relied heavily on patient compliance, and it involved the application of intermittent force, making it relatively unreliable [ 9 , 10 , 11 ].

TADs’ emergence has greatly improved the convenience and efficiency of treatment [ 12 , 13 ]. Compared to headgear, TAD-assisted vertical control can provide more dental intrusion and counterclockwise rotation of the mandibular plane, which contributes to further improvement of profile [ 14 , 15 ]. Additionally, when active intrusion was applied, we typically utilize a sustained light force (approximately 50 g), which is more favorable for the remodeling of periodontal tissues compared to the intermittent heavy force exerted by headgear.

However, the mini-implants placed in the maxilla’s posterior region can also provide strong sagittal anchorage. Several studies have shown that maximum anchorage itself can achieve a better treatment outcome and improve the profile [ 16 , 17 , 18 ]. These findings have prompted the following questions: If sagittal retraction can already lead to sufficient facial aesthetics, is vertical tooth movement still necessary? To what extent can vertical movements benefit the facial profile?

Our research group has paid close attention to the efficacy of TAD-assisted vertical control in orthodontic camouflage treatments for patients with skeletal class II malocclusion. We have published several case reports and long-term follow-up studies showing that vertical control significantly improved the profiles of patients with skeletal class II malocclusion and a hyperdivergent facial type, achieving good long-term stability [ 19 , 20 , 21 , 22 , 23 ]. We believe that specifying how the active intrusion of upper dentition contributes to these craniofacial improvements will provide more information about the ability and limits of TAD-assisted vertical control and broaden the understanding of orthodontic camouflage treatment. Therefore, we included a control group whose TADs were used only to reinforce maxillary sagittal anchorage in order to exclude the influence of sagittal retraction.

With this retrospective case–control study, we aimed to quantify the effectiveness of TAD-assisted vertical control in the improvement of dentoalveolar malformation and soft tissue profiles in adult patients with a severe skeletal class II hyperdivergent pattern, and justified the necessity of active intrusion. We believe that this article provides specific references for orthodontists and general dentists concerning the camouflage treatment of patients with skeletal class II malocclusion.

This study was based on retrospective data obtained from orthodontic records at the Peking University School and Hospital of Stomatology, and it was approved by the institution’s biomedical ethics committee (approval number: PKUSSIRB-201630096, retrospectively registered). The patients included in this study accepted orthodontic treatment between 2006 and 2018.

The study’s sample selection was based on the following inclusion criteria: good-quality orthodontic records, the presence of permanent dentition, age > 18 years, a convex profile, skeletal class II (ANB > 5°), and a hyperdivergent skeletal pattern (FMA > 28°) [ 24 ]. The exclusion criteria included the following: dental anomalies in size, number, shape, or structure; permanent tooth loss; orthodontic–orthognathic combined surgery treatment; and Botox injection or prosthesis implantation before or during orthodontic treatment.

Treatment protocols

All the participants underwent systematic periodontal and endodontic assessments and therapies before orthodontic interventions. A straight-wire MBT technique was utilized after the extraction of four premolars from all patients. Braces and archwires were obtained from TP Orthodontics (La Porte, IN, USA). The alignment and leveling phases involved initial bracket-bonding followed by a certain procedure utilizing 0.014 in. NiTi, 0.016 in. NiTi, 0.016 in. × 0.022 in. NiTi, and 0.019 in. × 0.025 in. NiTi archwires sequentially. During the space-closing phase, a 0.019 × 0.025 in. stainless steel archwire was applied using a conventional sliding mechanism. This phase was terminated upon the complete closure of the premolar spaces. The patients’ dentition was finely adjusted before debonding. Miniscrews (diameter: 1.5 mm; length: 7 mm; Zhongbang Medical Treatment Appliance, Xi’an, China) were surgically inserted into the alveolar ridge.

The patients were divided into two groups: (1) the maxillary maximum anchorage (MMA) group, in which TADs were implanted only at the bilateral buccal side of the alveolar bone, between the roots of the upper premolar and the upper first molar or between the upper first molar and the upper second molar; and (2) the maxillary maximum anchorage with vertical control (MMA + VC) group, in which TADs were implanted into the bilateral buccal and lingual sides of the alveolar bone, between the roots of the upper first molar and the upper second molar, to intrude the upper molars with or without the TADs implanted in the anterior segment for incisor intrusion (Fig.  1 ).

Representative image of intraoral devices. A . TAD-assisted intrusion of the upper anterior teeth. B . Buccal view of the posterior intrusion devices. C . Palatal view of the posterior intrusion devices

Sample size calculation

In this study, the effect size of the primary outcome was expected to be 2.32. This number was the difference in mandibular counterclockwise rotation (the decrease in the FMA value) between the two groups calculated in our preliminary study. The sample size was calculated using online software ( http://hedwig.mgh.harvard.edu/sample_size/ ) by assuming 5% type I errors and 20% type II errors. The sample calculation indicated that at least 10 patients were needed in each group.

In total, 36 patients were selected for the current study. The MMA group comprised 17 patients (14 females, 3 males) with a mean age of 24.18 ± 3.83 years and a mean treatment duration of 34.4 ± 12.8 months. The MMA + VC group consisted of 19 patients (16 females, 3 males) aged 25.00 ± 4.99 years, whose mean treatment duration was 34.7 ± 6.8 months. No significant difference in the patients’ gender, age, or treatment duration was observed between the groups (Additional Table  1 ).

Cephalometric analysis

Pre-treatment and post-treatment lateral cephalograms were collected, digitized, and superimposed using the Dolphin 11.0 software (Dolphin Imaging, Chatsworth, CA). An investigator who was not informed about the study’s groups obtained the measurements, which a second blinded investigator checked for accuracy. Any disagreements between these investigators were resolved through a weighted reevaluation until they were both satisfied. The variables used in the cephalometric analysis included skeletal, dental, and soft-tissue-related measurements. In total, 29 such variables were used (8 skeletal, 12 dental, and 9 soft-tissue-related). Figure  2 depicts the landmarks and important variables used in this study, while Additional Table  2 provides definitions.

Tracing of a pretreatment cephalometric radiograph

Dental cast analysis

Pre-treatment and post-treatment dental casts were scanned using a 3Shape scanner (3Shape D, Kopenhagen, Dänemark) and measured in a double-blinded manner by a trained orthodontist using the Geomagic 13.0 software (Geomagic Qualify, Durham, NC, US). As Fig.  3 shows, the superimposition of the dental casts was based on the palate’s stable structure. A coordinate system was built, based on the definition of the anatomical occlusal plane and the midline of the palate. The tooth movements were analyzed in two dimensions, anterior or posterior (X) and intrusion or extrusion (Z). Additionally, posterior and extrusive movement was defined as positive.

Superimposition of the dental casts. A . The pre-treatment maxillary model. B . The post-treatment maxillary model. C . Superimposition based on the stable structure of the palate. D . Transfer of corresponding landmarks

To evaluate the method’s error, 10 post-treatment lateral cephalograms and digital casts were randomly selected and remeasured by the same examiners two weeks after the first measurement was obtained. The intraclass correlation coefficient (ICC) was used to assess intra-examiner reliability and the reproducibility of all linear and angular measurements.

Statistical analysis

The intraclass correlation efficient (ICC) was evaluated using a two-way random model. Descriptive statistics for the dental casts and radiographic measurements were calculated for both the first and second measurements. Comparisons were performed and correlations were identified using Student’s t test in accordance with the results of Shapiro–Wilk normality tests. The pre-treatment skeletal, dental, and soft-tissue-related variables were compared between the groups using independent-sample t tests. The same variables were also compared from pre-treatment to post-treatment using paired t tests. The differences in treatment changes (concerning both the lateral cephalograms and the dental casts) between the MMA and MMA + VC groups were evaluated using independent-sample t tests. Multiple linear regression analysis was used to test the correlation between the independent variables of craniofacial structures and the dependent variable, the Z angle. Both groups’ differences in treatment changes were normalized to the mean variance. Then, a backward method was used to screen the independent variables. The entry probability of F was 0.05, and the removal criterion was 0.1. The statistical tests were performed with SPSS 18.0 software (IBM Corp., Armonk, NY). The results were considered statistically significant at P  < 0.05.

The groups were similar in age at the beginning of the orthodontic treatment (Additional Table  1 ). ICC was calculated with good reproducibility of the measurements (0.810–0.997), as Additional Table  3 shows.

The two groups showed similar mandibular retrognathia and hyperdivergent skeletal patterns. However, differences were observed in several variables, such as the Z angle, ANB, and L1-NB (mm). These differences indicated that the patients in the MMA + VC group had a more convex profile and more severe malocclusion (Table  1 ).

TAD-assisted vertical control better improved patients’ profiles

TADs’ efficacy in improving therapeutic outcomes is certain. However, whether and to what extent TAD-assisted vertical control can help patients with skeletal class II achieve better results from camouflage orthodontic treatments compared to the simple reinforcement of the maxillary anchorage is unclear.

For most of the patients whose results we recorded, a convex profile was the main complaint. Therefore, we first analyzed the improvements in soft-tissue-related variables for both groups (Tables  2 and 3 ). We discovered a similar trend of lip retraction (the UL-SnV angle and distance and the LL-SnV distance) and soft tissue relaxation (UL thickness and LL thickness). However, the change in the Z angle and Chin thickness showed that patients in the MMA + VC group experienced more improvement in their profiles and mentalis relaxation. (Figures  4 and 5 show the representative cases of the two groups, respectively.) Through these results, we have shown that TAD-assisted vertical control further improved the patients’ profiles, but how this advantage was achieved remained unclear.

A representative case from the MMA group. The upper anterior incisors were restored using a ceramic veneer

A representative case from the MMA + VC group

TAD-assisted vertical control contributed to maxillary retraction and mandibular counterclockwise rotation

Remarkable decreases in SNA and ANB were discovered in both groups. Furthermore, the decrease in ANB in the MMA + VC group was significantly greater compared to MMA group, showing effective maxillary retraction, which partly explained the dramatic change in the soft tissue (Tables  4 and 5 ).

Additionally, no significant differences in the mandibular plane angle in the MMA group pre- and post-treatment were observed. Indeed, the lower facial height (ANS-Me) even increased slightly. Meanwhile, the MP-SN and FMA values significantly decreased in the MMA + VC group, suggesting that TAD-assisted vertical control effectively achieved mandibular counterclockwise rotation. The decrease in the mandibular plane angle showed a significant difference in the MP-SN and FMA values between the MMA and MMA + VC groups (Table  5 ). An emphatic change was also observed in the improvement of PFH/AFH, indicating an improvement in the hyperdivergent facial type. Thus, the application of TAD-assisted vertical control achieved a certain extent of mandibular counterclockwise rotation, which also helped improve patients’ profiles (Fig.  6 ).

Schematic graph of TAD-assisted vertical control during orthodontic camouflage treatment for patients with skeletal class II and a hyperdivergent facial type

TADs achieved substantial vertical control via the intrusion of maxillary dentition

Despite the gratifying sagittal retraction of the incisors in both groups (Table  6 ), the study’s cephalometric analysis showed significant intrusion of the upper molar on the P-P plane (U6-PP) in the MMA + VC group but not in the MMA group. Similarly, the upper incisor showed more intrusion (U1-PP) in the MMA + VC group, though no significance was observed (Table  7 ). These results were confirmed via dental cast superimposition (Table  8 ). Compared to the MMA group, the MMA + VC group experienced significant intrusion of the upper dentition. However, our cephalometric analysis also revealed a significantly lower molar extrusion (L6-MP) on the mandibular plane in both groups during orthodontic treatment. Thus, the tooth movement in the vertical dimension manifested the intrusion of the upper dentition for the MMA + VC group and the extrusion of the lower molars for both groups.

Multivariate regression analysis revealed the key factors of profile improvement

Since the changes occurred at the same time, assessing which factors played the most important role in altering the patients’ soft tissue profile was difficult. Therefore, we selected the Z angle—one of the most representative and remarkably changed profile indicators—as the dependent variable for our analysis, and we conducted multiple linear regression of the standardized bone, tooth, and soft tissue measurements.

Considering the interference of collinearity, we selected the following representative indicators: ANB, MP-SN, PFH/AFH, U1/SN, IMPA, U1-PP, U6-PP, L1-MP, L6-MP, Pog-NB, UL thickness, LL thickness, and Chin thickness.

The results showed that Y  = 0.000576 − 0.416 a  − 0.340 b  + 0.403 c (where Y denotes the Z angle and a , b , and c represent the MP-SN, U1-SN, and Pog-NB, respectively; Table  9 ). This finding indicated that the change in the Z angle was negatively correlated with the MP-SN and U1-SN variables and positively correlated with Pog-NB.

Thus, the gratifying profile improvement of patients with skeletal class II and the hyperdivergent facial type relied on the massive retraction of the upper incisors, the shape of the chin, and the mandibular plane’s counterclockwise rotation.

The efficacy of TAD-assisted vertical control

In this retrospective study, we endeavored to quantify the efficacy of TAD-facilitated vertical control in managing maxillary dental intrusion and consequent mandibular counterclockwise rotation. Subsequently, we elucidated their pivotal roles in enhancing soft tissue profiles according to the baseline of MMA group.

Evaluation of hard tissue showed that following en-masse retraction with mini-implants anchorage, the MMA group exhibited slight upper molar intrusion (U6: -0.86 ± 0.89 mm) and mandibular counterclockwise rotation (MP-SN: -0.16 ± 1.05°). This result is consistent with the randomized controlled trial conducted by Al-Sibale et al. [ 25 ] and the controlled clinical trial conducted by Chen et al. [ 14 ], suggesting that TADs in the maxillary alveolar can provide some vertical force even during sagittal retraction, necessitating attention to the direction of traction and the vertical position of the anterior teeth to avoid deepening of the overbite. Following active maxillary dental intrusion, the MMA + VC group exhibited greater upper molar intrusion (U6: -2.25 ± 1.03 mm) and mandibular counterclockwise rotation (MP-SN: -1.82 ± 1.38°), which is slightly lower than that reported by Ding et al. [ 15 ] and Deguchi et al. [ 26 ] This difference may be attributed to differences in inclusion criteria. In Ding’s study, the inclusion criteria were shallow overbite, while in Deguchi’s study were open bite. In contrast, our study included many patients with normal or even deep overbite. To achieve a favorable overbite after treatment, we conducted intrusion of not only molars but also anterior teeth (U1: -1.30 ± 1.61 mm; U3: -1.81 ± 1.28 mm) with the help of TADs in the anterior segment, which represented a more challenging improvement compared to the aforementioned studies.

In terms of soft tissue evaluation, many previous studies have discussed the main factors contributing to changes in various soft tissue landmarks. For instance, Maetevorakul et al. found that the improvement in incisor angle was most crucial for enhancing lower lip prominence, and the mandibular plane angle as well as different treatment modalities had significant effects on changes of soft tissue chin prominence [ 27 ]. Regarding the overall assessment of soft tissue profiles, Zhao et al. demonstrated that the Z angle had the best discriminative ability for female adults with Angle Class II Division 1 malocclusion [ 28 ]. Therefore, in this study, we stressed on the Z angle and found that the MMA + VC group showed a more significant improvement compared to the MMA group (15.25 ± 5.30° in the MMA + VC group; 10.54 ± 5.11° in the MMA group, P  = 0.011), which correlates with the poorer profiles before treatment in the MMA + VC group. To better identify which patients require active dental intrusion, we conducted a multiple linear regression analysis and found that this improvement was most closely associated with the retraction of the upper anterior teeth, prominence of the pogonion, and counterclockwise rotation of the mandibular plane. Therefore, we can conclude that vertical control is more necessary for patients with limited space for retraction or poor chin morphology.

Limitations and prospects of TAD-assisted vertical control

Although the occlusal plane’s counterclockwise rotation is considered an effective method to reduce the angle of the mandibular plane [ 29 ], in the current study, we observed a trend of clockwise rotation. However, this unexpected result is consistent with the findings of many similar studies in this field [ 12 , 13 ]. We speculate that this rotation results from the pendulum effect of the upper anterior teeth. Compared with the molars, the upper incisors have less intrusion, suggesting that we must pay particular attention to controlling the occlusal plane.

Additionally, despite TADs’ advantages of simplicity, flexibility, and independence from patient cooperation, they remain an invasive treatment [ 30 , 31 ]. In the current study, however, six miniscrews were needed to achieve effective vertical control. This approach does not apply to patients with improper bone conditions, and it also increases the difficulty of operation. Therefore, we hope to develop further methods that are more convenient and minimally invasive. The use of midpalatal miniscrews and personalized palatal bars may be an alternative option [ 12 ]; however, such an approach would still pose challenges in terms of operation and hygiene maintenance. Accordingly, we hope to further reduce orthodontic devices’ complexity in order to meet the requirements of comfortable treatment.

Methodologically, the current study’s evaluation of muscle response and profile changes was limited to a cephalometric analysis. Since soft tissue yields inaccurate measurements during lateral cephalograms, 3D facial scanning and electromyography could allow a more precise examination of patients’ aesthetic and functional changes. We plan to enhance the refinement of assessment modalities for both soft and hard tissues, endeavoring to substantiate vertical control’s efficacy and constraints through various methodologies, including randomized controlled trials.

The conclusions of this retrospective study are as follows.

TAD-reinforced maxillary anchorage with vertical control achieves intrusion of approximately 2 mm for the upper first molars.

TAD-reinforced maxillary anchorage with vertical control induces mandibular counterclockwise rotation of approximately 1.8° and improves patients’ hyperdivergent skeletal pattern.

When the upper incisors are not sufficiently retracted or the chin shape is not satisfying, active vertical control should be applied to help patients achieve better profiles.

Taken together, these conclusions demonstrate that TAD-assisted vertical control is essential for patients with skeletal class II and a hyperdivergent facial type. This approach constitutes a good alternative to improving occlusion and profiles via orthodontic camouflage treatment.

Data availability

The data sets used and analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

Maxillary maximum anchorage

Temporary anchorage devices

• Vertical control

Tucker MR. Orthognathic surgery versus orthodontic camouflage in the treatment of mandibular deficiency. J Oral Maxillofac Surg. 1995;53(5):572–8.

Article   CAS   PubMed   Google Scholar  

Baherimoghaddam T, Oshagh M, Naseri N, Nasrbadi NI, Torkan S. Changes in cephalometric variables after orthognathic surgery and their relationship to patients’ quality of life and satisfaction. J Oral Maxillofac Res. 2014;5(4):e6.

Article   PubMed   PubMed Central   Google Scholar  

Thomas PM. Orthodontic camouflage versus orthognathic surgery in the treatment of mandibular deficiency. J Oral Maxillofac Surg. 1995;53(5):579–87.

Raposo R, Peleteiro B, Paco M, Pinho T. Orthodontic camouflage versus orthodontic-orthognathic surgical treatment in class II malocclusion: a systematic review and meta-analysis. Int J Oral Maxillofac Surg. 2018;47(4):445–55.

Freitas MR, Lima DV, Freitas KM, Janson G, Henriques JF. Cephalometric evaluation of class II malocclusion treatment with cervical headgear and mandibular fixed appliances. Eur J Orthod. 2008;30(5):477–82.

Jung MH. Vertical control of a class II deep bite malocclusion with the use of orthodontic mini-implants. Am J Orthod Dentofac Orthop. 2019;155(2):264–75.

Article   Google Scholar  

Peng J, Lei Y, Liu Y, Zhang B, Chen J. Effectiveness of micro-implant in vertical control during orthodontic extraction treatment in class II adults and adolescents after pubertal growth peak: a systematic review and meta-analysis. Clin Oral Investig. 2023;27(5):2149–62.

Article   PubMed   Google Scholar  

Jamilian A, Showkatbakhsh R, Rad AT. A novel approach for treatment of mandibular deficiency with vertical growth pattern. Int J Orthod Milwaukee. 2012;23(2):23–7.

PubMed   Google Scholar  

Lione R, Franchi L, Laganà G, Cozza P. Effects of cervical headgear and pendulum appliance on vertical dimension in growing subjects: a retrospective controlled clinical trial. Eur J Orthod. 2015;37(3):338–44.

Sambataro S, Rossi O, Bocchieri S, Fastuca R, Oppermann N, Levrini L, et al. Comparison of cephalometric changes in Class II growing patients with increased vertical dimension after high-pull and cervical headgear treatment. Eur J Paediatr Dent. 2023;24(1):36–41.

CAS   PubMed   Google Scholar  

Ulger G, Arun T, Sayinsu K, Isik F. The role of cervical headgear and lower utility arch in the control of the vertical dimension. Am J Orthod Dentofac Orthop. 2006;130(4):492–501.

Lee J, Miyazawa K, Tabuchi M, Kawaguchi M, Shibata M, Goto S. Midpalatal miniscrews and high-pull headgear for anteroposterior and vertical anchorage control: cephalometric comparisons of treatment changes. Am J Orthod Dentofac Orthop. 2013;144(2):238–50.

Yao CC, Lai EH, Chang JZ, Chen I, Chen YJ. Comparison of treatment outcomes between skeletal anchorage and extraoral anchorage in adults with maxillary dentoalveolar protrusion. Am J Orthod Dentofac Orthop. 2008;134(5):615–24.

Chen M, Li ZM, Liu X, Cai B, Wang DW, Feng ZC. Differences of treatment outcomes between self-ligating brackets with microimplant and headgear anchorages in adults with bimaxillary protrusion. Am J Orthod Dentofac Orthop. 2015;147(4):465–71.

Ding S, Liu M, Zou T. Comparative study on vertical effect between miniscrew and face-bow in orthodontic treatment of hyperdivergent class II protrusion. J Oral Sci Res. 2019;35(4):351–4.

Google Scholar  

Khlef HN, Hajeer MY, Ajaj MA, Heshmeh O. Evaluation of treatment outcomes of En masse Retraction with Temporary skeletal Anchorage devices in comparison with two-step retraction with Conventional Anchorage in patients with Dentoalveolar Protrusion: a systematic review and Meta-analysis. Contemp Clin Dent. 2018;9(4):513–23.

Kokodynski RA, Marshall SD, Ayer W, Weintraub NH, Hoffman DL. Profile changes associated with maxillary incisor retraction in the postadolescent orthodontic patient. Int J Adult Orthodon Orthognath Surg. 1997;12(2):129–34.

Kuroda S, Yamada K, Deguchi T, Kyung HM, Takano-Yamamoto T. Class II malocclusion treated with miniscrew anchorage: comparison with traditional orthodontic mechanics outcomes. Am J Orthod Dentofac Orthop. 2009;135(3):302–9.

Wang XD, Zhang JN, Liu DW, Lei FF, Liu WT, Song Y, et al. Nonsurgical correction using miniscrew-assisted vertical control of a severe high angle with mandibular retrusion and gummy smile in an adult. Am J Orthod Dentofac Orthop. 2017;151(5):978–88.

Deng JR, Li YA, Wang XD, Li J, Ding Y, Zhou YH. Evaluation of Long-Term Stability of Vertical Control in Hyperdivergent patients treated with Temporary Anchorage devices. Curr Med Sci. 2018;38(5):914–9.

Wang XD, Zhang JN, Liu DW, Lei FF, Zhou YH. Nonsurgical correction of a severe anterior deep overbite accompanied by a gummy smile and posterior scissor bite using a miniscrew-assisted straight-wire technique in an adult high-angle case. Korean J Orthod. 2016;46(4):253–65.

Wang XD, Lei FF, Liu DW, Zhang JN, Liu WT, Song Y, et al. Miniscrew-assisted customized lingual appliances for predictable treatment of skeletal class II malocclusion with severe deep overbite and overjet. Am J Orthod Dentofac Orthop. 2017;152(1):104–15.

Wang Y, Zhou Y, Zhang J, Wang X. Long-term stability of counterclockwise mandibular rotation by miniscrew-assisted maxillary intrusion in adult patients with skeletal class II high-angle malocclusion: a 10-year follow-up of 2 patients. AJO-DO Clin Companion. 2022;2(6):601–17.

Haralabakis NB, Sifakakis IB. The effect of cervical headgear on patients with high or low mandibular plane angles and the myth of posterior mandibular rotation. Am J Orthod Dentofac Orthop. 2004;126(3):310–7.

Al-Sibaie S, Hajeer MY. Assessment of changes following en-masse retraction with mini-implants anchorage compared to two-step retraction with conventional anchorage in patients with class II division 1 malocclusion: a randomized controlled trial. Eur J Orthod. 2014;36(3):275–83.

Deguchi T, Kurosaka H, Oikawa H, Kuroda S, Takahashi I, Yamashiro T, et al. Comparison of orthodontic treatment outcomes in adults with skeletal open bite between conventional edgewise treatment and implant-anchored orthodontics. Am J Orthod Dentofac Orthop. 2011;139(4 Suppl):S60–8.

Maetevorakul S, Viteporn S. Factors influencing soft tissue profile changes following orthodontic treatment in patients with Class II Division 1 malocclusion. Prog Orthod. 2016;17:13.

Zhao Z. The study on sensitivity of Asethetic Index from six kinds of Profile Soft tissue analysis methods on female adult in Liaoning Province. Dalian, China: Dalian Medical University; 2016.

Li X, Zhao Q, Zhao R, Gao M, Gao X, Lai W. Effect of occlusal plane control procedure on hyoid bone position and pharyngeal airway of hyperdivergent skeletal class II patients. Angle Orthod. 2017;87(2):293–9.

Papadopoulos MA, Papageorgiou SN, Zogakis IP. Clinical effectiveness of orthodontic miniscrew implants: a meta-analysis. J Dent Res. 2011;90(8):969–76.

Yamaguchi M, Inami T, Ito K, Kasai K, Tanimoto Y. Mini-implants in the anchorage armamentarium: new paradigms in the orthodontics. Int J Biomater. 2012;2012:394121.

Download references

Acknowledgements

We thank LetPub ( www.letpub.com ) and Scribendi ( www.scribendi.com ) for its linguistic assistance during the preparation of this manuscript.

This project is supported by the National Program for Multidisciplinary Cooperative Treatment on Major Diseases No. PKUSSNMP-202013; China Oral Health Foundation No. A2021-021; Beijing Municipal Science & Technology Commission No. Z171100001017128; National Natural Science Foundation of China No. 81901053, No.81900984 No. 82101043 and No. 8237030822; National High Level Hospital Clinical Research Funding No. 2023-NHLHCRF-YXHZ-TJMS-05; Elite Medical Professionals Project of China-Japan Friendship Hospital No.ZRJY2023-QM05; Beijing Municipal Natural Science Foundation No. 7242282.

Author information

Yan-Ning Guo and Sheng-Jie Cui contributed equally to this work.

Authors and Affiliations

Department of Orthodontics, National Center of Stomatology & National Clinical Research Center for Oral Diseases & National Engineering Laboratory for Digital and Material Technology of Stomatology & Beijing Key Laboratory of Digital Stomatology & Research Center of Engineering and Technology for Computerized Dentistry Ministry of Health & NMPA Key Laboratory for Dental Materials, Peking University School and Hospital of Stomatology, No.22, Zhongguancun South Avenue, Haidian District, Beijing, 100081, China

Yan-Ning Guo, Sheng-Jie Cui, Jie-Ni Zhang, Yan-Heng Zhou & Xue-Dong Wang

Dental Medical Center, China-Japan Friendship Hospital, Beijing, 100029, China

Yan-Ning Guo

Department of Orthodontics, the School of Stomatology, The Key Laboratory of Stomatology, Hebei Medical University, Shijiazhuang, 050017, China

Fourth Division Department, National Center of Stomatology & National Clinical Research Center for Oral Diseases & National Engineering Laboratory for Digital and Material Technology of Stomatology & Beijing Key Laboratory of Digital Stomatology & Research Center of Engineering and Technology for Computerized Dentistry Ministry of Health & NMPA Key Laboratory for Dental Materials, Peking University School and Hospital of Stomatology, Beijing, 100081, China

You can also search for this author in PubMed   Google Scholar

Contributions

Y-N G and S-J C: Investigation, Visualization, Writing - Original draft preparation, contributed equally to this work and joint first authors. Y L: Methodology, Investigation, Data Curation, and critically revised the manuscript. Y F: Investigation, Data Curation, and critically revised the manuscript. J-N Z, Y-H Z: Supervision and critically revised the manuscript. X-D W: Conceptualization, Validation, Writing- Reviewing and Editing, Corresponding author. All authors reviewed the manuscript.

Corresponding author

Correspondence to Xue-Dong Wang .

Ethics declarations

Ethics approval and consent to participate.

This study was approved by the Biomedical Ethics Committee of Peking University School and Hospital of Stomatology. (PKUSSIRB-201630096, retrospectively registered) All methods were carried out in accordance with relevant guidelines and regulations. Informed consent was obtained from all subjects and/or their legal guardian(s).

Consent for publication

Not Applicable.

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.

Yan-Ning Guo, Sheng-Jie Cui joint first authors.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

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.

Guo, YN., Cui, SJ., Liu, Y. et al. Quantitative evaluation of vertical control in orthodontic camouflage treatment for skeletal class II with hyperdivergent facial type. Head Face Med 20 , 31 (2024). https://doi.org/10.1186/s13005-024-00432-2

Download citation

Received : 07 November 2023

Accepted : 29 April 2024

Published : 14 May 2024

DOI : https://doi.org/10.1186/s13005-024-00432-2

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

• Skeletal class II
• Profile improvement

Head & Face Medicine

ISSN: 1746-160X

• General enquiries: [email protected]

• Open access
• Published: 17 May 2024

Identification of patients’ smoking status using an explainable AI approach: a Danish electronic health records case study

• Ali Ebrahimi 1 ,
• Margrethe Bang Høstgaard Henriksen 2   na1 ,
• Claus Lohman Brasen 3 , 4 ,
• Ole Hilberg 5 , 4 ,
• Torben Frøstrup Hansen 2 , 4 ,
• Lars Henrik Jensen 2 ,
• Abdolrahman Peimankar 1 &
• Uffe Kock Wiil 1  

BMC Medical Research Methodology volume  24 , Article number:  114 ( 2024 ) Cite this article

Metrics details

Smoking is a critical risk factor responsible for over eight million annual deaths worldwide. It is essential to obtain information on smoking habits to advance research and implement preventive measures such as screening of high-risk individuals. In most countries, including Denmark, smoking habits are not systematically recorded and at best documented within unstructured free-text segments of electronic health records (EHRs). This would require researchers and clinicians to manually navigate through extensive amounts of unstructured data, which is one of the main reasons that smoking habits are rarely integrated into larger studies. Our aim is to develop machine learning models to classify patients’ smoking status from their EHRs.

This study proposes an efficient natural language processing (NLP) pipeline capable of classifying patients’ smoking status and providing explanations for the decisions. The proposed NLP pipeline comprises four distinct components, which are; (1) considering preprocessing techniques to address abbreviations, punctuation, and other textual irregularities, (2) four cutting-edge feature extraction techniques, i.e. Embedding, BERT, Word2Vec, and Count Vectorizer, employed to extract the optimal features, (3) utilization of a Stacking-based Ensemble (SE) model and a Convolutional Long Short-Term Memory Neural Network (CNN-LSTM) for the identification of smoking status, and (4) application of a local interpretable model-agnostic explanation to explain the decisions rendered by the detection models. The EHRs of 23,132 patients with suspected lung cancer were collected from the Region of Southern Denmark during the period 1/1/2009-31/12/2018. A medical professional annotated the data into ‘Smoker’ and ‘Non-Smoker’ with further classifications as ‘Active-Smoker’, ‘Former-Smoker’, and ‘Never-Smoker’. Subsequently, the annotated dataset was used for the development of binary and multiclass classification models. An extensive comparison was conducted of the detection performance across various model architectures.

The results of experimental validation confirm the consistency among the models. However, for binary classification, BERT method with CNN-LSTM architecture outperformed other models by achieving precision, recall, and F1-scores between 97% and 99% for both Never-Smokers and Active-Smokers. In multiclass classification, the Embedding technique with CNN-LSTM architecture yielded the most favorable results in class-specific evaluations, with equal performance measures of 97% for Never-Smoker and measures in the range of 86 to 89% for Active-Smoker and 91–92% for Never-Smoker.

Our proposed NLP pipeline achieved a high level of classification performance. In addition, we presented the explanation of the decision made by the best performing detection model. Future work will expand the model’s capabilities to analyze longer notes and a broader range of categories to maximize its utility in further research and screening applications.

Peer Review reports

Introduction

Information on smoking status is crucial especially in cardiovascular, pulmonary, diabetes, and cancer research, since in addition to being a common risk factor it is also a confounder for various diseases [ 1 ]. Smoking accounts for more than eight million deaths annually [ 2 ]. In the specific area of lung cancer, the implementation of screening and detective models is becoming more relevant. The models, however, lack the ability to identify high-risk individuals who are dependent on tobacco [ 3 ]. In Denmark, smoking habits are not formally registered unless patients are diagnosed with cancer or a chronic disease that includes them in the National Clinical Registries. For example, information on the smoking habits of a lung cancer patient will be registered in the Danish Lung Cancer Registry, while a patient with chronic obstructive pulmonary disease, followed at a hospital level, will appear in the Danish Register of Chronic Obstructive Lung Disease [ 4 ]. Patients with milder conditions often do not appear in national registries, and information on smoking habits is only available as unstructured free-text in electronic health records (EHRs) [ 5 ]. The records often have an unrestricted format leading to differences between clinicians in terms of spelling errors, abbreviations, and a field-specific jargon that may be difficult for outsiders to interpret [ 6 ]. Clinicians have to manually search for smoking habits, which is feasible when dealing with a small number of patients, but it becomes impractical with larger cohorts, such as a high-risk population for lung cancer screening or large-scale research with smoking as an essential risk factor [ 7 ].

Natural Language Processing (NLP), a sub-field of artificial intelligence, focuses on analyzing linguistic data, particularly unstructured textual data using machine learning. The main goal of NLP is to transform free text into structured data that can be easily identified by machines [ 8 ]. NLP has been used in healthcare for various tasks such as detecting heart failure criteria [ 9 ], identifying adverse drug effects [ 10 ] detecting symptoms of specific disease, and improving quality of life [ 11 ]. In 2006, the “Informatics for Integrating Biology and the Bedside” research center announced the “Smoking challenge” funded by the National Institute of Health in the USA. The challenge aimed to address the problem of classifying smoking status based on EHRs and compare the performance with classifications made by pulmonologists. By means of supervised and unsupervised classifiers, several models demonstrated the ability to classify smoking status using a limited number of key textual features [ 12 ]. More recently, applying deep neural networks to EHRs has been in focus due to their better performance and lower preprocessing requirements [ 13 , 14 ]. In 2018, Google developed a new technique called Bidirectional Encoder Representations from Transformers (BERT). Unlike traditional word embedding methods such as word2vec, BERT is context-sensitive and generates a representation of each word based on the other words in the sentence [ 15 , 16 ]. BERT is considered state-of-the-art, as it allows for transfer learning and adaptations to other domains [ 17 ]. In 2020, a Danish edition of BERT was introduced, trained on 1.6 billion words from various online repositories (Common Crawl, Wikipedia, OpenSubtitles, etc.) [ 18 , 19 ]. Additionally, in 2021, Derczynski et al. presented the first Danish Gigaword Corpus, a billion-word corpus encompassing a wide range of the Danish language from different domains, settings, time periods, registers, and dialects [ 20 ].

Despite the advancements, a high-performing model capable of detection smoking status in the Danish language is yet to be developed. This limitation can be attributed to both the limited availability of text data due to access restrictions and the lack of advanced model development. The complex structure of EHRs further limits the possibility of transfer learning from other languages [ 21 ]. Consequently, this paper aims to address these challenges by presenting a high-performing NLP-based model capable of detecting smoking status in Danish EHRs using both binary and multiclass labels. The model is expected to be valuable in future screening scenarios and various research fields, including other types of cancer and cardiovascular diseases.

As machine learning and deep neural networks continue to advance, they often remain mysterious for both developers and end-users, resembling black boxes. The lack of transparency obstructs the broad adoption of such models, especially in domains where decision making holds critical importance such as the medical field. To effectively implement a model in a medical context, explainability becomes imperative in allowing clinicians and researchers to trust and comprehend the precise detection made [ 22 ]. An explainable model enhances the chance of identifying systematic errors and hence improves the model’s performance. Understanding the rationale behind a detection and the potential for model enhancement is of utmost importance for clinicians or researchers who will ultimately be responsible for the outcomes. While research and applications in explainable artificial intelligence have grown in the context of image and structured data models, those based on free-text datasets have received comparatively less attention [ 23 ]. Consequently, in addition to developing highly accurate detection models, this study seeks to provide transparent post-hoc explanations for the models.

The contributions of our study can be summarized as follows:

Formulate several NLP-based architectures to identify smoking status : To the best of our knowledge, this is the first study to detect smoking status based on Danish text from EHR. Several NLP-based architectures formulated resulting from the integration of advanced feature extraction techniques with ensemble-based machine learning and deep learning models.

Analyzing the detection performance of developed architectures and comparing them with state-of-the-art detection models : Comprehensive analysis and comparison of the performance of the developed models against existing state-of-the-art predictive models, with the superior models identified through rigorous statistical evaluation. This not only highlighted the detection performance of our model in comparison to others but also explored into a non-parametric statistical assessment based on the Friedman test.

Post-hoc explanations for the detection models : The study is the first study to provide model explanations for smoking status detection based on EHR. Explanation of the models’ decision-making processes using the state-of-the-art XAI approach, LIME, highlighting the significance of individual features and the underlying rationale for model decisions.

The subsequent sections outline the process of data collection followed by preprocessing, feature extraction, model development, evaluation, and explanation. Figure  1 provides a comprehensive overview of the study’s methodology, encompassing all stages of the pipeline.

Data collection

Data for this project were obtained from EHRs within a cohort of 38,944 patients who underwent assessments for a potential risk of lung cancer between January 1, 2009 and December 31, 2018 in the Region of Southern Denmark. This cohort has been comprehensively described in a related work [ 24 ]. We collected all types of documents from the EHRs containing the subheaders “smoking” or “risk factors” without imposing any time constraints. The subheaders were chosen, as they most commonly contain documentation of smoking history. Moreover, the data annotation process would have been impractical which we used on complete patient notes from the EHRs. We carefully eliminated duplicate entries, instances with missing gender information, and we pseudonymized the data to ensure privacy and confidentiality.

Pre-processing

Clinical notes underwent manual annotation by a medical doctor and the results were subsequently reviewed by the same doctor. The dataset underwent further refinement with a decision to include only one note per patient. As the annotated data have been employed in previous studies to predict lung cancer status, our selection focused on the note that provided the most comprehensive details regarding smoking status. Patients were primarily categorized as Active Smoker if they had detailed information on current pack-years (a widely recognized measure of smoking intensity calculated by multiplying the number of packs of cigarettes smoked per day by the number of years of smoking) [ 25 ]. The remaining patients, lacking information on pack-year, were categorized as Active-Smoker or Former-Smoker, or status unknown. To resolve the “unknown” category, additional notes for these patients were evaluated and a smoking label was assigned based on the note containing the most comprehensive information. Any duplicate entries were removed, retaining only the note responsible for the patient’s label.

To validate the smoking status annotation from the EHRs, the distributions were compared with registrations of pack-years obtained from the Danish Lung Cancer Registry. They had been recorded independently of the EHRs and manually completed by clinicians upon a patient’s lung cancer diagnosis. While the registration of smoking status from the Danish Lung Cancer Registry is not integrated in the EHRs, it is expected to align with the EHRs annotations overall. Following the annotation process the text underwent cleaning in the following sequence: Handling of abbreviations, conversion of all text to lowercase, removal of stop words, numbers, and punctuation marks. Consecutive spaces where there were two or more spaces in a row were either removed or converted to a single space. Finally, a word tokenizer was applied to convert sentences into word tokens.

The cleaning steps employed in this study are carefully tailored to enhance the analysis of Danish EHRs, recognizing the unique linguistic and structural characteristics of the language. Handling abbreviations initially is crucial in Danish language, where abbreviations can carry significant meanings or denote specific terminology, ensuring that such condensed forms are correctly interpreted or expanded for analysis. Converting all text to lowercase addresses the case sensitivity of Danish language, promoting uniformity and reducing the risk of duplicate representations for the same words.

The removal of stop words, numbers, and punctuation marks, beside the consolidation of consecutive spaces, streamlines the text, focusing the analysis on the most meaningful content without the noise of non-informative elements. This step is particularly effective in Danish, where functional words and punctuation can obscure key linguistic patterns if not properly managed. Applying a word tokenizer as the final step effectively breaks down sentences into individual tokens, a process that is essential for capturing the morphological richness of Danish words and phrases. Each of these steps, collectively, prepares the Danish EHRs for a more accurate and efficient computational analysis, ensuring that subsequent NLP tasks, such as feature extraction and model training, are performed on clean, consistent data that accurately reflects the intricacies of the Danish language.

Given the challenge of imbalanced class distribution, a stratified split approach was chosen, which entailed dividing the data into a training set (70%) and a test set (30%). By using a stratified split, the proportion of records in all classes remained consistent between the training and test sets. Preprocessing techniques, including data cleaning and feature extraction, were exclusively learned from the training set, and subsequently applied to the test set with necessary adaptations. This prevented a possible information leakage from the test set to the model training process, which could have led to an overly optimistic evaluation of model performance. It is important to note that the test set was exclusively used for evaluating the performance of the final models and did not contribute to the model learning process.

Feature extraction

Before choosing a classification algorithm for the task, it is essential to transform the unstructured data into a numerically vectorized representation. Feature extraction can be done with word embedding methods referring to the representation of words and whole sentences in a numerical manner. Words are converted into numeric vectors, and vectors of words closely related would be closer to each other [ 26 ]. In this study, we consider three methods to encode the tokens of a given technical text into a vectorized representation: The well-known Word embedding, BERT, Count Vectorizer and Word2Vector. General descriptions of all methods are described in detail in Table  1 . We applied a hyperparameter tuning step for the Count Vectorizer method using a randomized search cross validation to identify the threshold for the removal of frequent tokens and the number of n-grams.

Selecting Word Embedding, BERT, Count Vectorizer, and Word2Vec as methods for encoding tokens of Danish EHR into vector representations aligns with our objective to capture the linguistic nuances inherent to the Danish language effectively. Word Embeddings and Word2Vec, both deeply rooted in learning contextual relationships and semantic similarities, are particularly adept at navigating the intricate morphological characteristics of Danish language, such as its compound words and diverse verb forms. These methods excel in creating nuanced vector representations that reflect the semantic richness of words within their specific context, a crucial feature for the Danish language with its nuanced meanings and expressions.

BERT, with its deep contextualized training, excels in understanding the syntax and semantics of Danish text, leveraging its transformer architecture to capture subtle language cues and idiomatic expressions unique to Danish language. This is particularly beneficial given the contextual richness and syntactic flexibility of Danish. Lastly, Count Vectorizer provides a straightforward yet powerful approach to text representation, capturing the frequency of terms in a manner that supports the identification of domain-specific terminology prevalent in technical texts. Additionally, these methods provide a comprehensive toolkit for Danish text analysis, balancing deep semantic understanding with robust statistical approaches to ensure accurate and meaningful representation of Danish EHR.

Model development

Stacking-based ensemble (se).

The SE method was created by Wolpert et al. and is different from previous ensemble learning techniques in that it employs meta-learning to combine multiple types of machine learning algorithms [ 30 ]. SE is used in a two-level structure where the level-1 meta learner combines the outputs of the level-0 base learners. Figure  1 , Sect. 4 illustrates the stacking structure used in this study, which comprises three stages. The first stage involves training the base classifiers, which are K-Nearest Neighbor, Decision Trees, Random Forest, and XGBoost algorithms. The second stage involves gathering the output detection (feature vectors) of the base classifiers to generate a new reorganized training set. Finally, in the third stage, the Logistic Regression algorithm is utilized to train the meta-classifier using the new training set, resulting in the development of SE. Detailed descriptions of the developed machine learning algorithms are provided in Table  2 .

For the detection of smoking status, we also used the architecture CNN-LSTM. It consists of five layers, i.e., an input layer for word embedding, a one-dimensional convolutional network layer for local feature extraction, an LSTM network layer for capturing long-term dependencies, a dropout layer, and a classification layer for label detection. The structure of our model is shown in Fig.  1 (Sect. 4). In the input layer, input texts are treated as a matrix. Each row of the matrix represents a word, derived from the feature extraction method. In this study, the dimension of 300 is considered for the input layer. We used a one-dimensional convolution layer (Conv1D) to capture the sequence information and reduce the dimensions of the input data. A convolution operation involves a convolutional kernel applied to a fixed window of words to compute a new feature. The kernel, also called a filter, completes the feature extraction. Each filter is applied to a window of m words to obtain a single feature. To ensure the integrity of the word as the smallest granularity, the width of the filter is equal to the width of the original matrix. In this study, we employed the Conv1D layer with 256 filters and a kernel size of 3 in the output of the embedding layer to learn the lower-level features from words. A nonlinear activation function ReLU is used to reduce the number of iterations needed for convergence in deep networks.

Following the above steps, the result of the convolution was pooled using the maximum pooling operation to capture essential features in the text. To improve the quality of our text classification task, the different calculated features were concatenated to constitute the input of the LSTM layer. LSTM solves the vanishing gradient problem because it learns to regulate the flow of information. Due to high memory power, LSTMs can efficiently capture contextual information from the input text and produce high-level features that are used for further classification. We added a dropout layer to reduce the chance of overfitting. Finally, the last component is the fully connected layer, which takes as input the characteristics generated from a sentence by the LSTM layer and consequently detects the most appropriate label according to semantic and syntactic content. The probability that a sentence belongs to the smoking categories is calculated by the Softmax activation function.

Model architectures

Combining the different feature extraction methods with CNN-LSTM and the SE resulted in seven architectures: (1) Embedding with CNN-LSTM, (2) Embedding with SE, (3) Bert with CNN-LSTM, (4) Bert with SE, (5) Word2Vector with CNN-LSTM, (6) Word2Vector with SE, and (7) Count Vectorizer with SE. The details of these architectures are presented in Fig.  1 , Sect. 4.

In this study, we chose not to employ the combination of Count Vectorizer with a CNN-LSTM architecture. The rationale behind the decision lies in the intrinsic design of the Count Vectorizer, which produces a bag-of-words representation, consequently discarding word order. CNN-LSTM architectures are specifically tailored to capture sequential patterns in data; therefore, using a bag-of-words representation compromises their primary advantage. Furthermore, the integration of CNN-LSTM introduces substantial complexity to the model. In the absence of sequential data to leverage the unique strengths of CNN-LSTM, alternative simpler models may potentially offer comparable or superior performance without the computational overhead of such intricate architectures.

Model evaluation

To assess the detection performance of the created classifiers, several metrics were employed, including the receiver operating characteristics curve (ROC), area under the receiver operating characteristics curve (AU-ROC), Precision, Recall, F1-Score, and detection accuracy. These performance metrics are determined by searching for the values of true positive (TP), false positive (FP), false negative (FN), and true negative (TN). Detailed descriptions of the evaluation metrics used are presented in Table  3 .

Model explanation

Explainable Artificial Intelligence (XAI) techniques helps to explain the decisions made by machine learning models so that humans can understand. Ensuring that clinical staff and end users trust a machine learning model’s decisions requires making it’s reasoning process clear and comprehensible [ 37 ]. The local interpretable model-agnostic explanations (LIME) framework is one of the most extensively used XAI packages that enables classifiers to explain individual detection [ 38 ]. It explains a decision by locally approximating the classifier’s decision boundary in the given instance’s neighborhood. LIME builds locally linear models to explain the detection of a machine learning model. It corresponds to the rule-based regional explanations through the simplification category. Explanations through simplification build an entirely new model based on the trained machine learning model to be explained. The newly simplified model then attempts to optimize its similarity to its previous model functions while lowering complexity and maintaining comparable performance. As a result, after the machine learning decision is achieved, the LIME is used to assess the features’ importance and probabilities in the decision. As a result, we can determine the importance of the features in the decision input, which assists in interpreting the model outputs. We applied this technique to the models, which has the highest detection. Since the data are private and contain sensitive information, only the non-sensitive portions of the sentences are displayed in the examples.

Dataset description

From the total cohort of patients examined on suspicion of lung cancer ( N  = 38,944), notes containing the two subheaders were available on 23,542 patients (59%). After removing duplicates and patients missing data on gender, the final cohort was reduced to 23,132 patients, each with multiple registrations (92,113 notes). Each note contained an average of 60 tokens, but the range of the token length varied between 1 and 1051. The annotation of the 23,132 patients with available notes resulted in the following distribution of smoking habits: 6121 (26%) Never-Smoker, 10,617 (46%) Former-Smoker and 6394 (28%) Active-Smoker. They were further pooled into binary labels of Non-Smoker (26%) and Smoker (74%), which is former and active smokers.

To validate the data annotation, the results were matched against the registrations in the Danish Lung Cancer Registry. From the 23,132 patients with EHR-annotated smoking status, 4719 had lung cancer. Among these, data on smoking status registered in the Danish Lung Cancer Registry was available on 4168 patients. In the registry 217 patients were listed as Non-Smoker, of which the EHR annotation was equivalent in 83% of the cases. The registration as Smoker was made on 3787 patients of which the EHR annotation was equivalent in 97% of cases. This was overall considered to be a high correlation between the results and acceptable validity of the manual annotation from free text.

Binary classification

It is important to note that in terms of precision, recall, and F1-score, the SE-based architecture was low on average and class-specific performance. As presented in Fig.  2 , BERT with SE and Embedding with SE achieved the worst results compared with the other feature extraction methods, in which the accuracy reached 97%. This might be due to high dimensionality, causing the SE to be less effective when compared to alternative methods. On the other hand, BERT with CNN-LSTM could achieve almost the highest overall accuracy and precision of 99% among all developed architecture. However, as shown in Table  4 , BERT using CNN-LSTM shared the best precision of 99% with Embedding using the CNN-LSMT architecture for the Smoker class.

In terms of recall, Embedding with CNN-LSTM and Count Vectorizer with SE achieved the highest precision of 98% as shown in Fig.  2 . For the single class of Smoker, however, Bert with CNN-LSTM achieved the highest recall of 100% (Table  4 ). In terms of F1-Score, Word2Vector achieved the highest overall performance of 98%. As to F1-Score of a single class of Smoker, three architectures achieved the highest score of 99%, i.e., BERT with CNN-LSTM, Word2Vector with CNN-LSTM and Count Vectorizer with SE.

Results based on confusion matrix (Fig.  3 ) indicates that Word2Vector with CNN-LSTM architecture had the best performance in terms of detecting Smoker class with a true detection rate of about 99%. BERT with CNN-LSTM architecture performed best in detecting Non-Smoker patients at a true detection rate of about 98%. The results of other machine learning classifiers including KNN, DT, RF, and XGBoost are presented in Supplementary Fig.  1 and Supplementary Table 1 .

Multiclass classification

As presented in Fig.  4 , BERT with SE had the lowest performance compared to the other feature extraction methods, in which the accuracy reached 89%. Contrarily, BERT with CNN-LSTM achieved the highest accuracy, precision, recall, F1-score, and AUC of 95%. This architecture also performed the best in most of the class specific outcomes. As presented in Table  5 , BERT with CNN-LSTM had the highest performance for precision and F1-score of the Never-Smoker and Active-Smoker classes. In terms of precision, this architecture achieved 98% and 95% in the Never-Smoker and Active-Smoker classes, respectively. In terms of F1-score, it achieved 97% and 93% in the Never-Smoker and Active-Smoker classes, respectively.

Other architectures also achieved reasonable detection performances close to the performance of BERT with CNN-LSTM architecture. Embedding with CCN-LSTM and Count Vectorizer with SE achieved an overall accuracy of 94% (Fig.  4 ), which is only 1% lower than BERT with CNN-LSTM. Considering the results in Table  5 , Embedding with CCN-LSTM and BERT with CNN-LSTM architecture achieved the highest precision and F1-scores of 94% and 95%, respectively, for the Former-Smoker class. In terms of recall, the results for each class varied. For the Never-Smoker class, Count Vectorizer with SE achieved the highest recall of 98%. For the Active-Smoker class, Embedding with CNN-LSTM, BERT with CNN-LSTM, and Count Vectorizer with SE achieved the highest recall of 91%. In the Former-Smoker class, BERT with CNN-LSTM achieved the highest recall of 97%.

Results derived from the confusion matrix reveal that the Embedding with CNN-LSTM and Count Vectorizer with SE architectures exhibited superior performance in detecting the Active-Smokers and Never-Smoker classes, yielding true detection rates of approximately 91% and 97%, respectively (Fig.  5 ). BERT with CNN-LSTM excelled in identifying samples from the Former-Smoker class, with a true detection rate of 98%. When accounting for the smallest discrepancy in detection rates across all classes, both the Embedding with CNN-LSTM and Count Vectorizer with SE architectures were the most consistent. This suggests a marginal difference of about 4% between the Former-Smoker and Active-Smoker classes, which is the narrowest gap observed across all architectures. The marginal difference between the Never-Smoker class and other classes in the Embedding with CNN-LSTM architecture presents the narrowest gap compared to all other architectures developed. The results of other machine learning classifier including KNN, DT, RF, and XGBoost are presented in Supplementary Fig.  2 and Supplementary Table 2 .

Post-hoc comparison of model architectures

Since the results derived from detection performances and confusion matrices did not provide sufficient insight to determine the optimal model, we conducted a Friedman test on the mean of average results from the seven developed architectures. As shown in Fig.  6 there was no significant difference in average performance between the classifiers, neither concerning the binary (A) nor the multiclass architectures (B).

XAI to explain detection model decisions

The results indicate that classifying between ‘Former-Smoker’ and ‘Active-Smoker’ status is challenging, as the models occasionally underperformed in these categories. Nonetheless, the architecture of Embedding with CNN-LSTM reached a nearly optimal performance. In this section we explain the framework of the architecture utilizing the LIME technique as depicted in Fig.  6 . All examples come with the original text and plots illustrating the importance of features for the detected class compared to the remaining two classes. Figure  7 A displays the data on a Former-Smoker accurately detected with a probability of 94% of being categorized as a Former-Smoker. The key feature, “rygeophør” (smoking cessation), played a central role in assigning the case to the Former-Smoker category. Figure  7 B presents the data of an Active-Smoker that was correctly detected with a probability of 100% as an Active-Smoker. This outcome was primarily influenced by the words “fortsat” (continued) and “dgl” (daily), which classified the patient into the Active-Smoker category. Figure  7 C, however, portrays an Active-Smoker that was misclassified as a Former-Smoker, with a detected high probability of 99% of being a Former-Smoker and merely 1% of being an Active-Smoker. The words “rygestop” (smoking cessation) and “2017” contributed significantly to the detection, while the words “dagligt” (daily) and “ryger” (smoker) skewed the classification toward the Active-Smoker label. Figure  7 D exhibits a Smoker incorrectly labeled as a Non-Smoker, due to the misinterpretation of the word “nihil” (nothing) within an alcohol assessment context.

Summary of findings.

We proposed effective detection NLP-based architectures for detection of smoking status using Danish EHRs. The data were collected from 23,132 patients who underwent examinations on suspicion of lung cancer. They were conducted at pulmonary departments in the Region of Southern Denmark from 2009 to 2018. Our proposed method encompassed the utilization of seven diverse model architectures developed through a combination of feature extraction techniques (embedding, BERT, Word2Vector, and count vectorizer), machine learning (SE) and deep learning (CNN-LSTM) models. We evaluated the performance of the architectures by examining various metrics for binary (Non-Smoker and Smoker) and multiclass (Never-Smoker, Active-Smoker, and Former-Smoker) classification tasks. Each metric focuses on a special aspect of the performance. Except for the AU-ROC, all metrics were constructed based on a confusion matrix (TP, FP, TN, and FN).

Given the complex nature of Danish language, particularly its compound word formation and unique syntactic structures, our proposed methodology was accurately designed to ensure the relevance and effectiveness of selected NLP pipeline in processing Danish language. The developed models and feature extractions were chosen for their robust linguistic capture capabilities, essential for the syntactic and morphological complexities of Danish. Adaptations included specialized preprocessing for Danish abbreviations and punctuation, and the fine-tuning of the BERT model with Danish EHR, enhancing its syntactic and semantic understanding of the language. The superior performance of the developed scenarios within our experimental validation highlights the success of these adaptations. Such outcomes not only validate our methodological choices but also underline the potential of our approach in advancing Danish language processing.

Performance metrics exhibited general similarity across the models, and post hoc tests revealed no significant differences when considering the mean of all outcomes. In terms of binary classification, however, the evaluations specific to each class indicated that BERT with CNN-LSTM outperformed the other models in all performance metrics.

In terms of multiclass classification, we observed that BERT with SE achieved the worst results compared with the other feature extraction methods in which the accuracy reached 89%. This was somehow expected due to the low amount of labeled data. BERT embeddings are high-dimensional vectors, which can lead to a large number of features when applied to the classical machine learning models. It resulted in high dimensionality causing the SE to become less efficient compared to other techniques.

On the other hand, the architecture of BERT with CNN-LSTM demonstrated overall superiority in terms of weighted average performance as well as class-specific performance metrics. It involves using BERT to generate contextual embeddings for the input text, passing them through a CNN layer to capture local features, and feeding the resulting features into an LSTM layer for sequential modeling and final classification. The superior performance of the BERT with CNN-LSTM architecture can be attributed to several key factors. Firstly, BERT, which is a state-of-the-art pre-trained language model, excels in capturing contextual information and semantic understanding from textual data. This enables it to extract intricate patterns and nuances in the EHRs related to smoking status, which can be highly context dependent. Furthermore, the combination of CNN and LSTM layers in this architecture allows for the effective extraction of both local and sequential features from the EHR text. CNNs are adept at capturing local patterns and features, while LSTMs excel at modeling sequential dependencies. The synergistic integration of these two components enables the model to capture a wide range of relevant information, from short-term textual features to long-term contextual dependencies, making it particularly well-suited for the nuanced task of smoking status identification. The combined approach helps the model effectively capture both global contextual information and local sequential patterns, resulting in improved performance in text classification tasks compared to using BERT with classic machine learning algorithms.

However, we believe that the Embedding with CNN-LSTM demonstrated the optimal results since the discrepancy in detection rates across all classes based on confusion matrix was the narrowest gap observed across the developed architectures. The Embedding with CNN-LSTM architecture exhibited more consistent detection rates across all classes compared to BERT with CNN-LSTM. This approach, with its straightforward embeddings, ensures efficient capture of semantic meanings, leading to faster training and reduced computational demands. Moreover, when tailored to specific datasets, the embeddings can potentially offer more aligned representations for the task at hand.

The consistent detection rates exhibited by the Embedding with CNN-LSTM architecture compared to BERT with CNN-LSTM can be attributed to its more structured feature representation, simpler model complexity, and potential alignment with the dataset’s characteristics. The use of word embeddings facilitates a focused representation of text data, aiding in the consistent identification of smoking-related terms across various classes. Additionally, the Embedding with CNN-LSTM’s relative simplicity may contribute to improved generalization across classes, particularly in the presence of class imbalances. This suggests that the architecture’s suitability for the dataset, combined with effective hyperparameter tuning, plays a crucial role in achieving stable and reliable detection rates across all classes. Hence, for the collected dataset in this study and the classification goals to detect smokers (Never-Smoker, Former-Smoker, Active-Smoker), the Embedding with CNN-LSTM architecture might be the more adaptable and optimal choice.

To provide additional insight into the interpretability of our results, we explored LIME-plots from the Embedding with CNN-LSTM architecture. Notably, these plots unveiled clinically relevant top features associated with each specific class. The utilization of explainable AI methods, notably the LIME, in the developed NLP pipeline, plays a pivotal role in enhancing the interpretability and trustworthiness of our smoking status identification process within the complex landscape of EHRs. With the natural complexity of EHR data, it is essential that our AI model’s decision-making is transparent and understandable to healthcare professionals. LIME enables us to provide detailed, human-readable explanations for each prediction, highlighting the most influential features and factors that led to a specific outcome. This not only empowers clinicians to gain deeper insights into the model’s reasoning but also allows them to validate the models’ decisions against their domain expertise. By bridging the gap between AI-driven predictions and clinical understanding, the explainable AI methods contribute significantly to the credibility and reliability of our smoking status identification system in the EHR environment, ultimately adding greater confidence in its utility and accuracy. The results were discussed with domain experts, who were in favor of a balanced performance across all classes in the dataset.

Comparison to previous study results

Different studies have evaluated the application of NLP based on machine learning and deep learning techniques for the detection of smoking status through EHRs with different languages [ 12 , 14 , 39 , 40 ]. Rajendran et al. developed a binary and multiclass classification model using English EHRs from the United States [ 14 ]. The model incorporated a CNN that utilized both a word-embedding layer pre-trained from the Google news corpus and a word2vec model, resulting in superior performance compared to conventional machine learning methods. The binary classification achieved an F1-measure of 85%, while the multiclass classification reached 68% for smoking status identification. Bae et al. developed a multiclass classification model using Korean and English EHR data extracted from 4711 clinical notes [ 39 ]. The most effective model employed an unsupervised keyword extraction technique in combination with a linear support vector machine, achieving an impressive F1-score of 91% for multiclass classification. Of note, both studies encountered challenges due to limited data availability and the extensive length of patient notes. Additionally, the Korean study faced limitations in terms of the relevant corpus available for pre-training, which necessitated the use of seed keywords pre-defined by clinicians for the keyword extraction method.

To the best of our knowledge, the most comparable study to ours is one based on Swedish EHR notes [ 40 ]. It developed classic machine learning models to classify smoking status into Current-Smoker, Ex-Smoker, Non-Smoker, and Unknown. Among the 32 developed detection models, support vector machine achieved the highest F1-score of 98%. The authors did not present the performance of developed models for each of the classes, which makes it difficult to understand the ability of models in different classes. Also, they did not consider any feature extraction method to transform the text into features and capture the essential information from the text. Consequently, the reasons for models’ decisions were not presented.

Limitation and Future Work

To the best of our knowledge this study represents the first exploration of a Danish NLP-model derived from a sizable dataset of manually annotated EHR-notes, but it has some limitations. It is important to acknowledge that the models are based on constrained input data. We exclusively considered text from the relatively short and simplistic subfield associated with smoking and risk factors in the EHR systems. Applying the established models on the complete EHR note is unquestionably bound to result in a performance decrease. Nevertheless, it is worth noting that the current Danish hospital systems store information on smoking status and other risk factors in a sub-header format similar to the structure observed in this dataset.

Another limitation pertains to the absence of an “unknown” category. Following the initial data annotation process, patients with unknown smoking status were further evaluated using additional notes. Ultimately, we selected the note containing the most detailed information on smoking status, resulting in the complete exclusion of the unknown category. This, however, represents a potential drawback since the model was not trained to classify “unknown” smoking status. Finally, it would be ideal to expand the model to include more detailed information on smoking status such as smoking duration and intensity. Incorporating these factors into a model would be relevant when determining eligibility for lung cancer screening. This would require a higher standard of quality and standardization in documenting smoking status compared to the current practices.

Based on the findings of this study, we plan to further explore the potential of this algorithm on longer EHR-notes without limitations to the subfield relevant to smoking. It would be valuable to incorporate free text from general practice to identify patients at risk of lung cancer or other chronic diseases where smoking status is a significant risk factor. However, data annotation remains a time-consuming task, and the size of the dataset may be limited by this factor when dealing with larger patient notes. Additionally, there is potential to annotate other risk factors, such as alcohol consumption, to expand the current model to different outcomes beyond smoking.

Clinical perspectives

To the best of our knowledge, this is the first model based on Danish EHR data. Despite its limitations, the current model holds potential for application to Danish EHR data acquired at a hospital level. The ability to extract smoking status directly from free-text material would be highly advantageous, given that smoking status is a crucial risk factor for various acute and chronic illnesses. Having such information readily available for large patient populations allows for further investigation, as this variable is typically only accessible for specific populations such as patients with lung cancer or coronary heart disease. The incorporation of explainable AI, specifically LIME plots, opens possibilities for enhancing future models by identifying potential systematic errors. Additionally, it offers valuable insights into predictions, a crucial aspect for responsible clinicians. In addition to its potential in advancing research, this model could also find utility in screening scenarios, providing valuable information for risk assessment tools.

We present the outcomes of a novel model capable of categorizing the smoking status of patients using Danish EHRs. By combining a transformer with a convolutional neural network, specifically BERT with CNN-LSTM, we achieved a remarkable performance, with low discrepancy in detection rates across all classes. This outcome accentuates the promising possibility of classifying smoking status based on unstructured free text data. The availability of comprehensive and precise information on smoking habits could potentially prove advantageous in future research endeavors. Moreover, it can aid in identifying high-risk individuals who are eligible for screening programs such as those aimed at detecting lung cancer.

Flowchart depicting the study design in each step of the NLP pipeline. Bidirectional Encoder Representations from Transformers (BERT). Convolutional neural network with a long short-term memory layer (CNN-LSTM). K-Nearest Neighbors (KNN). Decision Tree. Created with Biorender.com

Average performance measures based on binary classification of the seven model architectures. CNN-LSTM: Convolutional neural network with a long short-term memory layer. SE: Stacking-Based Ensemble. BERT: Bidirectional Encoder Representations from Transformers. AU-ROC: Area under Receiver Operating Characteristic Curve

Confusion matrixes based on binary classification of all seven model architectures

Average performance measures based on multiclass classification of the seven model architectures. CNN-LSTM: Convolutional neural network with a long short-term memory layer. SE: Stacking-Based Ensemble. BERT: Bidirectional Encoder Representations from Transformers. AU-ROC: Area under Receiver Operating Characteristic Curve

Confusion matrixes based on multiclass classification of the architectures of all seven models

Results of the Friedman test and Nemenyi post-hoc test, α = 0.05

LIME plots representing the outcomes of multiclass classification of four distinct samples derived from Embedding with CNN-LSTM. A : Former-Smoker accurately detection with a 94% probability of being a Former-Smoker. B : Active-Smoker correctly detection with a 100% probability of being an Active-Smoker. C : Active-Smoker misclassified as a Former-Smoker. D : Smoker wrongly classified as a Non-Smoker.

Data availability

The dataset used for this study is not publicly available due to the possibility of compromising individual privacy but is available from the corresponding author on reasonable request.

Abbreviations

Natural Language Processing

Electronic Health Records

Convolutional Neural Network with a long short-term memory layer

K-Nearest Neighbors

Decision Tree

Extreme Gradient Boosting

Random Forest

Stacking-based Ensemble

Area Under the Receiver Operating Characteristics Curve

Local interpretable model-agnostic explanations

Bidirectional Encoder Representations from Transformers

True Positive

True Negative

False Positive

False Negative

de Boer AR et al. Data mining to retrieve smoking status from electronic health records in general practice. Eur Hear Journal-Digital Heal, 2022.

Roth G, Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2017 (GBD 2017) Results. Seattle, United States: Institute for Health Metrics and Evaluation (IHME), 2018, Lancet , vol. 392, pp. 1736–1788, 2018.

Malhotra J, Malvezzi M, Negri E, Vecchia CL, Boffetta P. Risk factors for lung cancer worldwide. Eur Respir J. 2016;48(3):889–902.

Article   PubMed   Google Scholar  

Lange P et al. Danish Register of chronic obstructive pulmonary disease. Clin Epidemiol, pp. 673–8, 2016.

Schmidt M, et al. The Danish health care system and epidemiological research: from health care contacts to database records. Clin Epidemiol. 2019;11:563.

Article   PubMed   PubMed Central   Google Scholar  

Afzal Z. Text mining to support knowledge discovery from electronic health records, Erasmus University Rotterdam , 2018. http://hdl.handle.net/1765/105993 (accessed Oct. 03, 2023).

Groenhof TKJ, et al. Data mining information from electronic health records produced high yield and accuracy for current smoking status. J Clin Epidemiol. 2020;118:100–6.

Liddy ED. Natural language processing, in Encyclopedia of Library and Information Science, 2nd Ed. NY. Marcel Decker, Inc , 2001.

Byrd RJ, Steinhubl SR, Sun J, Ebadollahi S, Stewart WF. Automatic identification of heart failure diagnostic criteria, using text analysis of clinical notes from electronic health records. Int J Med Inf. 2014;83(12):983–92.

Article   Google Scholar  

Sarker A, Gonzalez G. Portable automatic text classification for adverse drug reaction detection via multi-corpus training. J Biomed Inf. 2015;53:196–207.

Koleck TA, Dreisbach C, Bourne PE, Bakken S. Natural language processing of symptoms documented in free-text narratives of electronic health records: a systematic review. J Am Med. 2019;26(4):364–79. Informatics Assoc.

Google Scholar  

Uzuner Ö, Goldstein I, Luo Y, Kohane I. Identifying patient smoking status from medical discharge records. J Am Med Inf Assoc. 2008;15(1):14–24.

Li I, et al. Neural natural language processing for unstructured data in electronic health records: a review. Comput Sci Rev. 2022;46:100511.

Rajendran S, Topaloglu U. Extracting smoking status from electronic health records using NLP and deep learning, AMIA Summits Transl. Sci. Proc , vol. 2020, p. 507, 2020.

Devlin J, Chang M-W, Lee K, Toutanova K. Bert: Pre-training of deep bidirectional transformers for language understanding, arXiv Prepr. arXiv1810.04805 , 2018.

Devlin M-W, Jacob and, Chang O, Sourcing BERT. State-of-the-Art Pre-training for Natural Language Processing, 2018. https://blog.research.google/2018/11/open-sourcing-bert-state-of-art-pre.html (accessed Oct. 03, 2023).

Wonsild F, Møller MG. Danish Clinical Event Extraction Developing a clinical event extraction system for electronic health records using deep learning and active learning, 2020, [Online]. Available: https://api.semanticscholar.org/CorpusID:220267240 .

Hvingelby R, Pauli AB, Barrett M, Rosted C, Lidegaard LM, Søgaard A. DaNE: A named entity resource for danish, in Proceedings of the 12th Language Resources and Evaluation Conference , 2020, pp. 4597–4604.

Certainly. Certainly has trained the most advanced Danish BERT model to date, 2020. https://certainly.io/blog/danish-bert-model/ (accessed Oct. 03, 2023).

Derczynski L et al. The Danish Gigaword Corpus, in Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa) , 2021, pp. 413–421.

Kirkedal A, Plank B, Derczynski L, Schluter N. The Lacunae of Danish Natural Language Processing, in Proceedings of the 22nd Nordic Conference on Computational Linguistics , 2019, pp. 356–362. [Online]. Available: https://aclanthology.org/W19-6141.

Tjoa E, Guan C. A survey on explainable artificial intelligence (xai): toward medical xai. IEEE Trans Neural Networks Learn Syst. 2020;32(11):4793–813.

Zini JE, Awad M. On the explainability of natural language processing deep models. ACM Comput Surv. 2022;55(5):1–31.

Henriksen MB, et al. A collection of multiregistry data on patients at high risk of lung cancer—a Danish retrospective cohort study of nearly 40,000 patients. Transl Lung Cancer Res. 2023;12(12):2392.

Article   CAS   PubMed   PubMed Central   Google Scholar  

U.S. Department of Health and Human Services. National Cancer Institute Dictionary of Cancer Terms, 2024. https://www.cancer.gov/publications/dictionaries/cancer-terms/def/pack-year (accessed Apr. 04, 2024).

Jurafsky D, Martin JH. Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition.

Selva Birunda S, Kanniga Devi R. A review on word embedding techniques for text classification, Innov. Data Commun. Technol. Appl. Proc. ICIDCA 2020 , pp. 267–281, 2021.

Manning C, Schutze H. Foundations of statistical natural language processing . MIT press, 1999. [Online]. Available: https://doc.lagout.org/science/0_ComputerScience/2_Algorithms/Statistical Natural Language Processing.pdf.

Mikolov T, Chen K, Corrado G, Dean J. Efficient estimation of word representations in vector space, arXiv Prepr. arXiv1301.3781 , 2013.

Wolpert DH. Stacked generalization. Neural Netw. 1992;5:241–59. https://doi.org/10.1016/S0893-6080(05)80023-1 .

Mucherino A, Papajorgji PJ, Pardalos PM, Mucherino A, Papajorgji PJ, Pardalos PM. K-nearest neighbor classification. Data Min Agric, pp. 83–106, 2009.

Quinlan JR. Induction of decision trees. Mach Learn. 1986;1:81–106.

Rigatti SJ. Random forest. J Insur Med. 2017;47(1):31–9.

Bentéjac C, Csörg\Ho A, Mart\’\inez-Muñoz G. A comparative analysis of gradient boosting algorithms. Artif Intell Rev. 2021;54:1937–67.

Tolles J, Meurer WJ. Logistic regression: relating patient characteristics to outcomes. JAMA. 2016;316(5):533–4.

Fang Z, Wang Y, Peng L, Hong H. A comparative study of heterogeneous ensemble-learning techniques for landslide susceptibility mapping. Int J Geogr Inf Sci. 2021;35(2):321–47.

Jain R, Ciravegna G, Barbiero P, Giannini F, Buffelli D, Lio P. Extending Logic Explained Networks to Text Classification, arXiv Prepr. arXiv2211.09732 , 2022.

Ribeiro MT, Singh S, Guestrin C. ‘ Why should i trust you?’ Explaining the predictions of any classifier, in Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining , 2016, pp. 1135–1144.

Bae YS, et al. Keyword extraction algorithm for classifying smoking status from Unstructured Bilingual Electronic Health Records Based on Natural Language Processing. Appl Sci. 2021;11(19):8812.

Article   CAS   Google Scholar  

Caccamisi A, Jørgensen L, Dalianis H, Rosenlund M. Natural language processing and machine learning to enable automatic extraction and classification of patients’ smoking status from electronic medical records., Ups. J. Med. Sci , vol. 125, no. 4, pp. 316–324, Nov. 2020, https://doi.org/10.1080/03009734.2020.1792010 .

Download references

Acknowledgements

The authors would like to thank Karin Larsen, Research secretary, The Department of Oncology, Lillebaelt Hospital, University Hospital of Southern Denmark, for helping in proofreading the manuscript.

The study was funded by The Region of Southern Denmark, The University of Southern Denmark, The Danish Comprehensive Cancer Center, The Dagmar Marshall Foundation, The Beckett Foundation, The Lilly and Herbert Hansen Foundation and The Hede Nielsen Family Foundation. The funding bodies played no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Open access funding provided by University of Southern Denmark

Author information

Shared first Author.

Authors and Affiliations

SDU Health Informatics and Technology, The Maersk Mc-Kinney Moller Institute, University of Southern Denmark, Odense, 5230, Denmark

Ali Ebrahimi, Abdolrahman Peimankar & Uffe Kock Wiil

Department of Oncology, Lillebaelt Hospital, University Hospital of Southern Denmark, Vejle, 7100, Denmark

Margrethe Bang Høstgaard Henriksen, Torben Frøstrup Hansen & Lars Henrik Jensen

Department of Biochemistry and Immunology, Lillebaelt Hospital, University Hospital of Southern Denmark, Vejle, 7100, Denmark

Claus Lohman Brasen

Institute of Regional Health Research, University of Southern Denmark, Odense, Denmark

Claus Lohman Brasen, Ole Hilberg & Torben Frøstrup Hansen

Department of Internal Medicine, Lillebaelt Hospital, University Hospital of Southern Denmark, Vejle, 7100, Denmark

Ole Hilberg

You can also search for this author in PubMed   Google Scholar

Contributions

A.E designed, developed, and analyzed the methodology, performed the modeling, analyzed the results, and conducted the computations. M.B.H designed the methodology, analyzed the results, collected the data, and annotated the texts. A.E and M.B.H wrote the manuscript. C.L.B, O.H, and T.F.H contributed to analyzing the results from a clinical perspective and reviewed the manuscript. U.K.W. and A.P. contributed to the computations, result analysis and manuscript review. All authors discussed the results and contributed to the final manuscript.

Corresponding author

Correspondence to Ali Ebrahimi .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Ethical approval and consent to participate

The authors are accountable for all aspects of the work and will ensure that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Danish Data Protection Agency (19/30673, 06-12-2020) and the Danish Patient Safety Authority (3-3013-3132/1, 03-30-2020), and individual consent for this retrospective analysis was waived.

Consent for publication

Not Applicable.

Additional information

Publisher’s note.

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

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

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.

Ebrahimi, A., Henriksen, M.B.H., Brasen, C.L. et al. Identification of patients’ smoking status using an explainable AI approach: a Danish electronic health records case study. BMC Med Res Methodol 24 , 114 (2024). https://doi.org/10.1186/s12874-024-02231-4

Download citation

Received : 25 November 2023

Accepted : 23 April 2024

Published : 17 May 2024

DOI : https://doi.org/10.1186/s12874-024-02231-4

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

• Natural language processing
• Text classification
• Stacking-based ensemble
• Deep learning
• Explainable Artificial Intelligence (XAI)
• Electronic health record
• Smoking status

BMC Medical Research Methodology

ISSN: 1471-2288

• General enquiries: [email protected]