299270
Next steps in electronic BMI surveillance: Modeling missing data from electronic health record-based surveillance
San Diego received federal funding to expand the use of Electronic Health Records (EHR) for surveillance by leveraging existing infrastructure, the San Diego Immunization Registry (SDIR), as a mechanism to collect Body Mass Index (BMI) data. By the project’s conclusion, BMI data transfer was in place for 12 sites including six community clinic networks, two large medical systems, and four private medical groups. The SDIR contained BMI data for over 20% of the regional population. Initial data review revealed information gaps for geographic areas and population subgroups. Statistical methods were developed to estimate missing data to provide useful public health information during implementation of an EHR-based surveillance system.
Objective
To explore the development and application of a statistical model to estimate missing data for population surveillance using information collected in healthcare settings.
Methods
The model was tested on data from the SDIR with comparison population estimates from the California Health Interview Survey (CHIS) for clinic visitors and non-visitors. The goal was to examine how the estimates of BMI improved with additional predictive factors. The final analysis included gender, public versus private healthcare provider, and age.
Results
Mean BMI from CHIS for visitors and non-visitors respectively was 26.9 and 26.6, both of which were significantly lower than the SDIR mean of 27.9. Additional subgroup comparisons by public and private healthcare provider, gender, and age group yielded mixed results with the best results for the public provider subgroup.
Discussion
Increasing adoption of EHR systems has provided an opportunity to expand mechanisms to monitor population health. However, implementation of these systems is often dependent on outside organizations that influence the schedule and scope of the development. Mechanisms to develop interim estimates are vital to the expansion of disease surveillance to include chronic diseases and provide new tools to assess community health.
Learning Areas:
Chronic disease management and preventionCommunication and informatics
Learning Objectives:
Discuss the importance of developing mechanisms and tools to provide useful public health data from EHR's.
Describe process to calculate estimates for missing data in EHR-based BMI surveillance system.
Keyword(s): Data Collection and Surveillance, Obesity
Qualified on the content I am responsible for because: I am an epidemiologist and the lead evaluator for two large federal grants on obesity prevention through policy, systems, and environmental change. I worked in collaboration with consultants to develop and test the methods described in this abstract.
Any relevant financial relationships? No
I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.