Combining data from healthcare databases in Europe: A study of dopamine agonists

Objective: To demonstrate the value of combining data from multiple databases to address urgent safety issues, such as the risk of cardiac valve fibrosis with dopamine agonist use. Background: To explore the association between rare exposures and outcomes, a single automated database may not provide sufficient sample size and heterogeneity of exposure needed for comparative pharmacoepidemiologic drug safety research. Methods: Cohorts of new users of either 1) DAs or 2) levodopa, and 3) database, gender and age-matched controls from general population were identified from THIN-UK, Italian HSD-THALES, Dutch-IPCI GP databases, and Dutch-PHARMO record linkage system between 1996 and 2007. After database-specific extractions using common data model, a unique output dataset was created. Results: During the study period, cohorts identified were: 1) DA users for PD (N510,790) or hyperprolactinemia (N58,386), 2) levodopa (N514,669), and 3) controls (N529,311). Statistically significant differences on DA use were observed among Countries. Overall, cabergoline was the most frequently used DA (43%), irrespective of indication of use, with highest use in Italy (47%) and lowest use in UK (22%). Non-ergot DAs had the highest use in UK (35%) and NL (25%) and lowest use in Italy (18%). Conclusions: Combining data from multiple databases increases study size and heterogeneity of exposure, which can facilitate drug comparisons.