Mapping EHR Data to Examine Neighborhood-Level Variation in Obesity Rates in San Diego

BACKGROUND:  To monitor obesity trends, the County of San Diego Health and Human Services Agency collects electronic health record (EHR) data on height and weight via the San Diego Immunization Registry. BMI data is available for over 20% of the regional population, with representation greater than 50% in some neighborhoods. To understand the health needs of patients in the context of their communities, this project will use geographic information systems (GIS) for mapping and spatial analytics to examine neighborhood-level variation in obesity rates in San Diego.

METHODS:  Healthy weight surveillance is conducted through collection of individual-level data on height, weight, home address, date of birth, and additional demographic characteristics at six community clinic networks, two large medical systems, and four private medical groups. Using ZIP Code Tabulation Areas (ZCTA) from the 2010 Census, we calculated the percentage of residents with measured BMI, separately for adults and children. Patient-level BMI data were aggregated to the ZCTA to protect patient confidentiality. Hot spot analysis will be used to identify communities with higher than average obesity rates for adults and children. BMI data will be linked to data from the Census and the American Community Survey to examine characteristics of communities with higher than average obesity rates.

RESULTS:  Complete data were available for 1,102,722 person-years for individuals aged 2 to 99 years living within the County of San Diego during 2009 to 2014. Most individuals were female (57%) and aged 18 years or older (64%). On average, 10% of the child population of each ZCTA (range=0 to 29%) and 4.5% of the adult population of each ZCTA (range=0 to 18%) were included in our 2014 sample. Across 108 ZCTAs, the mean child BMI z score varied from -0.27 to 1.15 and mean BMI varied from 25.48 to 34.27 for adults. Results of the hot spot analysis will also be presented.

CONCLUSIONS:  The expansion of EHR systems provides new opportunities to monitor population health. Using GIS methods to analyze EHR data can help local health departments to better understand how obesity and other health outcomes vary across neighborhoods, identify health disparities, and target limited resources accordingly.