Skip to content

Medicine Matters

Sharing successes, challenges and daily happenings in the Department of Medicine

Medicine Matters Home Article of the Week Methods for Evaluating Natural Experiments in Obesity: A Systematic Review

Methods for Evaluating Natural Experiments in Obesity: A Systematic Review

ARTICLE: Methods for Evaluating Natural Experiments in Obesity: A Systematic Review

AUTHORS: Wendy L. Bennett, Renee F. Wilson, Allen Zhang, Eva Tseng, Emily A. Knapp, Hadi Kharrazi, Elizabeth A. Stuart, Oluwaseun Shogbesan, Eric B. BassLawrence J. Cheskin

JOURNAL: Ann Intern Med. 2018 May 1:1-10. doi: 10.7326/M18-0309. [Epub ahead of print]


BACKGROUND: Given the obesity pandemic, rigorous methodological approaches, including natural experiments, are needed.

PURPOSE: To identify studies that report effects of programs, policies, or built environment changes on obesity prevention and control and to describe their methods.

DATA SOURCES: PubMed, CINAHL, PsycINFO, and EconLit (January 2000 to August 2017).

STUDY SELECTION: Natural experiments and experimental studies evaluating a program, policy, or built environment change in U.S. or non-U.S. populations by using measures of obesity or obesity-related health behaviors.

DATA EXTRACTION: 2 reviewers serially extracted data on study design, population characteristics, data sources and linkages, measures, and analytic methods and independently evaluated risk of bias.

DATA SYNTHESIS: 294 studies (188 U.S., 106 non-U.S.) were identified, including 156 natural experiments (53%), 118 experimental studies (40%), and 20 (7%) with unclear study design. Studies used 106 (71 U.S., 35 non-U.S.) data systems; 37% of the U.S. data systems were linked to another data source. For outcomes, 112 studies reported childhood weight and 32 adult weight; 152 had physical activity and 148 had dietary measures. For analysis, natural experiments most commonly used cross-sectional comparisons of exposed and unexposed groups (n = 55 [35%]). Most natural experiments had a high risk of bias, and 63% had weak handling of withdrawals and dropouts.

LIMITATION: Outcomes restricted to obesity measures and health behaviors; inconsistent or unclear descriptions of natural experiment designs; and imperfect methods for assessing risk of bias in natural experiments.

CONCLUSION: Many methodologically diverse natural experiments and experimental studies were identified that reported effects of U.S. and non-U.S. programs, policies, or built environment changes on obesity prevention and control. The findings reinforce the need for methodological and analytic advances that would strengthen evaluations of obesity prevention and control initiatives.

For a link to the full article, click here:

Link to abstract online:


Kelsey Bennett