Population-level obesity surveillance: monitoring childhood body mass index z-score in a safety-net system

Abstract

Objective: To determine the utility of repeated patient-level body mass index (BMI) measurements among higher-risk patients seen at safety-net clinics as a community-level monitoring tool for overweight and obesity population trends.

Methods: Data from a network of urban, federally qualified community health centers with computerized tracking of BMI at sequential outpatient visits were analyzed. We performed a longitudinal observational study over 8 years (2005-2012) with children stratified by weight status groups on the basis of BMI. Changes in BMI z-scores were used to estimate population trends among children 2 to 11 years old, with at least 2 visits (at least 1 year apart), for whom weight and height were measured.

Results: Among children (n = 33,542), the rate of overweight was 16% and rate of obesity was 18% at their last visit. Children were followed for an average of 3.24 ± 1.76 years to measure trends and change in weight status from earlier to later childhood. Children who were obese at first visit had increased odds (adjusted odds ratio 27.8, 95% confidence interval 25.6-30.2) of being obese by last visit. Mean change in BMI z-score per person-year of observation was 0.10 ± 0.38, with a differing rate of change based on weight status category at last visit (not overweight = 0.06 ± 0.39; overweight = 0.17 ± 0.34; obese = 0.19 ± 0.36). Change in BMI z-score per person-year decreased for 40% of obese children; however, their weight status group remained unchanged.

Conclusions: Childhood obesity prevalence was high, with substantial progression to overweight and obesity from first to last visit. Clinically derived BMI z-score per person-year measures can effectively show population trends not observed using standard weight status categories.

Keywords: BMI z-score; childhood obesity; electronic health records; population trends.