Validity and concordance of electronic health records in primary care (AP-Madrid) for surveillance of diabetes mellitus. PREDIMERC study

Objective: To assess the validity and concordance of diabetes data in the electronic health records of primary care (Madrid-PC) by comparing with those from the PREDIMERC study. Methods: The sensitivity, specificity, positive predictive value, negative predictive value and kappa index of diabetes cases recorded in the health records of Madrid-PC were calculated by using data from PREDIMERC as the gold standard. The prevalence of diabetes was also determined according to each data source. Results: The sensitivity of diabetes recorded in Madrid-PC was 74%, the specificity was 98.8%, the positive predictive value was 87.9%, the negative predictive value was 97.3%, and the kappa index was 0.78. The prevalence of diabetes recorded in Madrid-PC was 6.7% versus 8.1% by PREDIMERC, where known diabetes was 6.3%. Conclusions: The electronic health records of primary care are a valid source for epidemiological surveillance of diabetes in Madrid. (C) 2013 SESPAS. Published by Elsevier Espana, S.L.U. All rights reserved.