Identifying geographic differences in children's sugar‐sweetened beverage and 100% fruit juice intake using health system data. Issue 11 (17th June 2020)

• Record Type: Journal Article
• Authors: Palakshappa, Deepak; Lenoir, Kristin; Brown, Callie L.; Skelton, Joseph A.; Block, Jason P.; Taveras, Elsie M.; Lewis, Kristina H.
• Is Part Of: Pediatric obesity. Volume 15:Issue 11(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-06-17
• Subjects: electronic health records -- population health -- public health -- sugar‐sweetened beverages; Obesity in children -- Periodicals; Obesity in adolescence -- Periodicals; Obesity -- Periodicals; Overweight children -- Periodicals; 618.92398
• DOI: 10.1111/ijpo.12663
• Languages: English
• ISSNs: 1747-7174

Abstract

Summary: Background: Identifying areas with higher sugar‐sweetened beverage intake could help tailor policy and public health efforts intended to reduce sugary beverage consumption. Objective: Demonstrate the feasibility of using health system data to examine the geographic distribution of sugar‐sweetened beverage intake and evaluate neighbourhood characteristics associated with intake. Methods: We extracted electronic health record data from a sugar‐sweetened beverage and 100% fruit juice screener used for children ages 1 to 17 years in eight pediatric practices in North Carolina (March 2017‐2018) and dichotomized intake to high (≥3 sugar‐sweetened beverages/day) vs not. We geocoded address and mapped the proportion of consumers in each census tract. We combined electronic health record data with US census data and evaluated associations of census tract income and race/ethnicity with intake. We used multivariable models to evaluate the association between geographic concentrations of income and race/ethnicity and sugar‐sweetened beverage intake, controlling for demographics extracted from the electronic health record and clustering by tract. Results: Of 19 451 patients, 4579 (23.5%) reported consuming ≥3 sugar‐sweetened beverages/day. In multivariable models, children living in tracts with high concentrations of low‐income (OR: 1.45, 95% CI: 1.26, 1.68) and non‐white residents (OR: 1.44, 95% CI: 1.21, 1.71) were more likely to consume ≥3 sugar‐sweetened beverages/day than children in tracts with a high concentration of high‐income and white residents. Conclusion: We demonstrate how health system data could be used to characterize geographic variation in sugar‐sweetened beverage and 100% fruit juice consumption. This approach could help target public health efforts and monitor the effects of community‐level interventions.