TY - JOUR T1 - Electronic Health Records vs Medicaid Claims: Completeness of Diabetes Preventive Care Data in Community Health Centers JF - The Annals of Family Medicine JO - Ann Fam Med SP - 351 LP - 358 DO - 10.1370/afm.1279 VL - 9 IS - 4 AU - Jennifer E. DeVoe AU - Rachel Gold AU - Patti McIntire AU - Jon Puro AU - Susan Chauvie AU - Charles A. Gallia Y1 - 2011/07/01 UR - http://www.annfammed.org/content/9/4/351.abstract N2 - PURPOSE Electronic Health Record (EHR) databases in community health centers (CHCs) present new opportunities for quality improvement, comparative effectiveness, and health policy research. We aimed (1) to create individual-level linkages between EHR data from a network of CHCs and Medicaid claims from 2005 through 2007; (2) to examine congruence between these data sources; and (3) to identify sociodemographic characteristics associated with documentation of services in one data set vs the other. METHODS We studied receipt of preventive services among established diabetic patients in 50 Oregon CHCs who had ever been enrolled in Medicaid (N = 2,103). We determined which services were documented in EHR data vs in Medicaid claims data, and we described the sociodemographic characteristics associated with these documentation patterns. RESULTS In 2007, the following services were documented in Medicaid claims but not the EHR: 11.6% of total cholesterol screenings received, 7.0% of total influenza vaccinations, 10.5% of nephropathy screenings, and 8.8% of tests for glycated hemoglobin (HbA1c). In contrast, the following services were documented in the EHR but not in Medicaid claims: 49.3% of cholesterol screenings, 50.4% of influenza vaccinations, 50.1% of nephropathy screenings, and 48.4% of HbA1c tests. Patients who were older, male, Spanish-speaking, above the federal poverty level, or who had discontinuous insurance were more likely to have services documented in the EHR but not in the Medicaid claims data. CONCLUSIONS Networked EHRs provide new opportunities for obtaining more comprehensive data regarding health services received, especially among populations who are discontinuously insured. Relying solely on Medicaid claims data is likely to substantially underestimate the quality of care. ER -