On November 17, 2014, Steve Gortmaker, PhD, moderated a panel discussion titled “Mapping Local Obesity Prevalence in the United States: Challenges and Opportunities for Surveillance and Prediction in the 21st Century.”
Ongoing state- and locally-specific BMI surveillance is a critical component of the public health response to the obesity epidemic. However, fewer than one half of states operate statewide childhood BMI surveillance systems, many with limited sample sizes lacking representativeness at the city or school-district level. Data collected on adult BMI distributions are based on self-reported measures of height and weight that may not be accurate, affecting overweight/obese classifications and potentially shifting some state rankings. Objectively-measured data are needed to project future state obesity-related medical expenditures and plan public health interventions.
The session explored the strengths and limitations of existing BMI surveillance structure at the state and national level and best practices for surveillance systems. Presenters, which included several CHOICES researchers, discussed how statistical matching and bootstrap sampling methods can be used to combine data from a range of publicly-available sources to provide bias-corrected prevalence estimates at state and local levels in the absence of a robust surveillance infrastructure.