Abstract P235: A Virtual Baseline Examination: Effect of Baseline Timeframe on Risk Factor Assessment
Introduction: Prospective cohort studies in classical cardiovascular disease (CVD) epidemiology begin with a baseline exam at which risk factors are measured, after which participants are followed over time. Longitudinal patient data from electronic health records (EHR) hold potential for CVD epidemiology but may necessitate flexibility in defining the timeframe of baseline risk factor assessment.
Objective: In creating the first CVD cohort study using patient data from a large health system, we tested the hypotheses that widening the timeframe used to measure baseline CVD risk factors increases the yield of eligible patient data without introducing selection bias.
Methods: The New England network (VISN 1) of the Veterans Health Administration cares for >1 million patients. To define a veteran cohort, we used EHR data to identify 240,479 eligible VISN 1 patients, defined as having valid demographic data and ≥1 set of blood lipid results between 2000-2007. We anchored the index date to the date of the first eligible lipid results. We then expanded the definition of the baseline timeframe by 1-week intervals before or after this date, assessing the proportion of eligible patients with blood pressure (BP) measurements with each successive widening of the baseline timeframe. We compared 3 groups of patients: 1) those with BP from the exact index date, 2) those with BP up to the 85th percentile to either side of the index date, and 3) those with no BP +/-180 days of the index date. We identified baseline CVD and diabetes from ICD-9 codes.
Results: Group 1 contained 146,636 (61.0%) of the eligible patients. This proportion reached 85% by +21 or -56 days from the index date. The 3 groups did not differ in age, BP, or LDL levels, but Group 3 (1.2% of total) had less available race data, lower prevalence of baseline comorbidities, and fewer outpatient visits.
Conclusions: Creating a prospective CVD cohort from EHR data is feasible but may require special handling of patients with infrequent visits to minimize selection bias.
Author Disclosures: J.L. Vassy: None. Y. Ho: None. K. Cho: None. J.M. Gaziano: None. J. Honerlaw: None. S. Raju: None. P.W.F. Wilson: None. D.R. Gagnon: None.
- © 2015 by American Heart Association, Inc.