JMIR Publications

ABSTRACT

Background:

Hypertension is the most prevalent risk factor for mortality globally. Uncontrolled hypertension is associated with excess morbidity and mortality, and nearly one half of individuals with hypertension do not have the condition under control. Data from electronic health record (EHR) systems may be useful for community hypertension surveillance, filling a gap in local public health departments’ community health assessments.

Objective:

This study sought to assess the comparability of six different, EHR-based hypertension prevalence estimates with estimates from a national survey.

Methods:

Using six different EHR-based computable phenotypes, we calculated hypertension prevalence estimates for Marion County, Indiana for the period of 2014-2015. We employed the two one-sided t-test (TOST) to test equivalence between BRFSS- and EHR-based prevalence estimates. The TOST was performed at the overall level as well as stratified by age, gender, and race.

Results:

Utilizing both 80% and 90% confidence intervals, the TOST analysis resulted in two computable phenotypes demonstrating rough equivalence to BRFSS. Variation in performance varies across phenotype as well as demographics. TOST with 80% confidence intervals demonstrated less variance compared to BRFSS within subpopulations, particarly related to racial categories. Overall, less variance occurred on phenotypes which included vitals measurements.

Conclusions:

This study demonstrates that some EHR-derived prevalence estimates may serve as rough substitutes for population-based survey estimates. It demonstrates the importance of assessing which data elements to include in EHR-based estimates. Utilizing comprehensive data sources containing complete clinical data as well as data representative of the population are crucial to producing robust estimates of chronic disease. Clinical Trial: not applicable