Educational Research Column: Systematic Review and Meta-analysis as a Means to Publish in Medical Education
Systematic Review and Meta-analysis as a Means to Publish in Medical Education
As 2016 came to a close, JAMA released its list of the top 10 most talked about articles of the year. One of these papers, “Prevalence of Depression, Depressive Symptoms, and Suicidal Ideation Among Medical Students: A Systemic Review and Meta-Analysis” by Rotenstein et al1, falls within the realm of medical education. This paper certainly serves as yet another reminder of the high prevalence of suicide in our medical student population and emphasizes the importance of prevention and wellness during training. However, it is also an example of a research modality that can be used to publish in the area of medical education.
Systematic review and meta-analyses refer to two distinct but related processes that are typically presented together in a single article. A systemic review is the deliberate, pre-determined process of performing a thorough and comprehensive search of the literature on a topic of interest. The findings are then statistically analyzed by meta-analysis, a statistical method, by which data from individual studies are analyzed together to give a single, quantitative description or estimate of the topic of interest. Effect sizes are calculated using one of several statistical models in order to draw conclusions based on an inclusive body of literature. A heterogeneity coefficient, which is used to describe similarity among individual studies, is typically calculated as well. A nonsignificant heterogeneity value signifies that the studies are similar enough for comparison and is more robust.
When publishing a systematic review and meta-analysis, there are several important factors to consider. It is crucial to perform a review that encompasses multiple databases and to have pre-determined inclusion criteria based on subject population and study methodology. A research librarian can be particularly helpful in formulating a search strategy that is most inclusive. Having at least two study investigators independently review each search result and assess for eligibility in the meta-analysis, while noting reasons for exclusion of those that are not eligible, can reduce heterogeneity. It is also important to extract meaningful data, including demographics, in an organized fashion. Invoking the assistance of a statistician for choosing the appropriate effects model is of utmost importance for those without an extensive statistical background. Finally, as with any type of research study, it is important that the conclusions you draw are justified by the analysis performed and not an over generalization.
Successful meta-analyses have been performed on medical education topics including wellness issues, simulation, teaching techniques, assessment modalities.
If there is a medical education topic of interest to you for which there is a growing body of evidence, consider performing a meta-analysis as your next research endeavor and add to our literature!
Nicole Dubosh, MD
Instructor, Harvard Medical School
Associate Clerkship Director
Department of Emergency Medicine
Beth Israel Deaconess Medical Center
- Rotenstein LS, Ramos MA, Torre M et al. Prevalence of depression, depressive symptoms, and suicidal ideation among medical students: a systemic review and meta-analysis. JAMA 2016 Dec 6;316(21):2214-2236