On December 5, NCCOR hosted a Connect & Explore webinar on Assessing Prevalence and Trends in Obesity: Navigating the Evidence. A consensus committee convened by the National Academies of Sciences, Engineering, and Medicine recently explored how reports on obesity prevalence and trends differ and what these differences mean for interpretation and application. During the webinar, speakers presented insights from the resulting National Academies report and provided an overview of the various data collection and analysis approaches that have been used in developing reports on the prevalence and trends in obesity across population groups, but particularly as they relate to children and adolescents. Speakers included Shari L. Barkin, M.D., M.S.H.S. (Chair), William K. Warren Foundation Chair and Professor of Pediatrics, Director of Pediatric Obesity Research in the Diabetes Center, and Chief of General Pediatrics at Vanderbilt University School of Medicine; Lynn Blewett, Ph.D., Professor, Division of Health Policy and Management, School of Public Health, Director, State Health Access Data Assistance Center (SHADAC); Jackson P. Sekhobo, Ph.D., M.P.A., Director of Evaluation, Research, and Surveillance in the Division of Nutrition of the New York State Department of Health; and Cynthia L. Ogden, Ph.D, M.R.P., NHANES Analysis Branch Chief/Epidemiologist, National Center for Health Statistics, Centers for Disease Control and Prevention.
The presentations generated many thoughtful questions, some of which the presenters were unable to answer due to time constraints. As a follow-up to this webinar, the presenters have answered all the questions the audience posed during the webinar.
- Can you describe how you assess for disparities in the obesity prevalence data?
Dr. Lynn Blewett: Many of the major surveys, such as NHANES and BRFSS, often oversample different population groups. It’s important to check with your state on their sampling methodology.
Dr. Shari Barkin: To understand and assess disparities you have to be really clear about how you’re defining disparities and then consider the interaction among those variables in terms of age, sex, gender, socioeconomic status, and race/ethnicity. It seems like it should be a straightforward question, but it isn’t. You can look at those samples that have larger sample sizes, such as NHANES and BRFSS, because they oversample with the hopes of answering more of those questions.