Accuracy of administrative databases in identifying patients with hypertension.

摘要: Background : Traditionally, the determination of occurrence hypertension in patients has relied on costly and time-consuming survey methods that do not allow to be followed over time. Objectives To determine accuracy using administrative claims data identify rates a large population living single-payer health care system. Methods: Various definitions for databases were compared with 2 other reference standards: (1) obtained from random sample primary physician offices throughout province, (2) self-reported national census. Results A case-definition algorithm employing outpatient billing 3-year period had sensitivity 73% (95% confidence interval [CI] 69%–77%), specificity 95% (CI 93%–96%), positive predictive value 87% 84%–90%), negative 88% 86%–90%) detecting hypertensive adults physician-assigned diagnoses. Compared data, 64% 63%–66%), 94%(CI 93%–94%), 77% (76%–78%), 89% 88%–89%). When this was applied entire province Ontario, age- sex-standardized prevalence older than 35 years increased 20% 1994 29% 2002. Conclusions It is possible use accurately those who have been diagnosed hypertension. Given are already routinely collected, their likely substantially less expensive serial cross-sectional or cohort studies surveillance outcomes time population.