Using Electronic Medical Records to Estimate the Prevalence of Pediatric Obesity in Mississippi

Pediatric obesity is a major public health concern in Mississippi. However, existing surveillance systems, including the Youth Risk Behavioral Surveillance System, National Survey of Children's Health, and Child and Youth Prevalence of Obesity Survey, are limited by self-reported data, restricted age ranges, and incomplete subgroup coverage.

This study evaluates the plausibility of using electronic medical records to improve population-level surveillance of pediatric obesity in Mississippi while addressing issues of data completeness, bias, and representativeness.

Electronic medical records from 40,215 patients aged 2-19 years who received care at the University of Mississippi Medical Center (2013-2022) were analyzed via multilevel regression to generate pediatric obesity prevalence estimates overall and by age, sex, and race/ethnicity for the medical center and the broader Mississippi population. Missing values were managed using multiple imputation by chained equations, and inverse probability weighting addressed ascertainment bias. Results were compared with existing surveillance systems.

The estimated Mississippi obesity prevalence ranged from 20.0% to 21.3% across years, with higher estimates among non-Hispanic Black youth and older age groups. Estimates aligned closely with existing surveillance systems.

Electronic medical record-based surveillance produced plausible, detailed obesity prevalence estimates for Mississippi, complementing existing systems and providing timely data across the pediatric age range.