Examining the Utility of Electronic Health Records for the Surveillance of Chronic Illness

Abstract

Introduction: Public health surveillance systems have evolved from the first recorded epidemic in 3180 B.C. in Egypt, to a modern system built upon health surveys and administrative data. Electronic health records (EHR) and electronic laboratory records (ELR) are increasingly seen as the next frontier for surveillance activities, particularly with respect to reportable infectious diseases. However, EHR may be underutilized in the surveillance of chronic illnesses and emerging environmental health challenges. As a pilot study, we examine the utility of EHR and ELR for documenting diabetes prevalence in two California counties, including disparities by geography, income, and race. Methods: Clinical data for diabetes diagnoses and glycohemoglobin test results were obtained for adult patients of a large California healthcare provider over a 5-year period. The validity of using glycohemoglobin laboratory tests to represent diabetes prevalence was evaluated using clinical diagnoses as a gold standard. Disparities were estimated by combining these data with income, race, and ethnicity data from the U.S. Census. Results: A patient’s five-year maximum glycohemoglobin value ≥ 7.0% yielded the best combination of sensitivity, specificity, and positive predictive value using a clinical diagnosis of diabetes as a gold standard. Using this metric, we found it straightforward to document relevant disparities in diabetes by race, income, and census tract. Conclusions: ELR may offer timely public health surveillance data able to document disparities, target interventions, and evaluate changes in population health. These data may be easier to access than a patient’s entire EHR, but outcome metric validation with diabetes diagnoses would need to be ongoing. Future activities could include data validation of laboratory data and EHRs across multiple providers, assessing other environmental health outcomes, and data linkage with environmental exposures.