142nd APHA Annual Meeting and Exposition

Annual Meeting Recordings are now available for purchase

301112
Filling in the Gaps: State-specific childhood obesity prevalence corrected for self-report bias

142nd APHA Annual Meeting and Exposition (November 15 - November 19, 2014): http://www.apha.org/events-and-meetings/annual
Monday, November 17, 2014 : 2:50 PM - 3:10 PM

Michael W. Long, ScD , Department of Social and Behavioral Sciences, Harvard School of Public Health, Boston, MA
Zachary J. Ward, MPH , Center for Health Decision Science, Harvard School of Public Health, Boston, MA
Kelly Blondin, SM , Department of Social and Behavioral Sciences, Harvard School of Public Health, Boston, MA
Stephen Resch, Ph.D. , Center for Health Decision Science, Harvard School of Public Health, Boston, MA
Angie L. Cradock, ScD , Department of Social and Behavioral Sciences, Harvard School of Public Health, Boston, MA
Y. Claire Wang, MD, MSc, ScD , Department of Health Policy and Management, Mailman School of Public Health, Columbia University, New York, NY
Amber Hsiao, MPH , Department of Health Policy and Management, Mailman School of Public Health, Columbia University, New York, NY
Steven L. Gortmaker, PhD , Department of Social and Behavioral Sciences, Harvard School of Public Health, Prevention Research Center, Boston, MA
Background: Accurate state-specific obesity prevalence data are critical to public health efforts to address the childhood obesity epidemic.  However, few states administer statewide objectively-measured childhood BMI surveillance programs.  For the first time, this study aims to estimate state-specific childhood obesity prevalence correcting for reporting bias and validate these estimates against published state BMI surveillance data.

Methods: We developed a non-parametric statistical matching approach to combine data from the 2010 U.S. Census, 2010 American Community Survey (ACS) (5-year), 2003-2004 and 2007-2008 National Survey of Children’s Health (NSCH) (n=133,214), and 2005-2010 National Health and Nutrition Examination Surveys (NHANES) (n=9,422 ages 2-17).  Measured height and weight data from statistically matched NHANES individuals were assigned to a random sample of the Census population by state, age, sex, race/ethnicity and household income, accounting for state-level variation in parent-reported height and weight from NSCH.   The resulting obesity prevalence estimates were validated against surveillance data from five states (AR, FL, MA, PA, TN) conducting a census of all children across a range of grades.

Results: Estimated state-specific childhood obesity prevalence ranged from 11.7% to 21.5%. For the five validation states, our model-derived obesity prevalence estimates aligned closely with measured data, with a mean difference of 0.68 percentage points (range: 0.31-1.38) and a high correlation coefficient (r=0.90, p=0.037).

Conclusion: Non-parametric statistical matching provides a novel method for estimating state-specific childhood obesity prevalence corrected for reporting bias.  In the absence of statewide surveillance systems, this approach provides data needed to target limited obesity prevention resources.

Learning Areas:

Biostatistics, economics
Epidemiology
Public health or related research

Learning Objectives:
Demonstrate importance of non-parametric statistical matching methods for synthesizing new datasets from existing publicly available administrative and public health sources. Evaluate validity of estimated childhood obesity prevalence using non-parametric statistical matching against data from published statewide BMI surveillance systems.

Keyword(s): Surveillance, Obesity

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I have earned a masters and doctoral degree in social and behavioral sciences and obesity epidemiology and prevention. I have presented at APHA and published on a range of obesity epidemiology and quantitative obesity policy evaluation topics. For the past four years, I have been working with a transdisciplinary team to build a simulation model of the U.S. population and ongoing obesity epidemic to evaluate costs and benefits of obesity prevention policies.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.