The growing availability of medical-administrative databases presents an opportunity to identify patients with certain diseases. Before using them for this purpose, it is necessary to assess their validity. We studied the validity of drug reimbursement databases for identifying patients treated for Parkinson’s disease in five branches of the Mutualité sociale agricole in 2007. We compared the reimbursement profiles for antiparkinsonian medications among patients whose diagnosis of Parkinson’s disease was confirmed by a neurologist and among patients treated for another condition. Using a structured approach, we developed a model to estimate the probability that individuals with a reimbursement for an antiparkinsonian medication are being treated for Parkinson’s disease. Among 1,114 participants with at least one reimbursement for an antiparkinsonian medication in 2007, a diagnosis of Parkinson’s disease was confirmed for 320 (29%) of them, while 794 (71%) were treated for another reason. The logistic model, which includes cumulative doses of different types of antiparkinsonian medications and treatment regularity, performs well (R² = 0.953; sensitivity = 92.5%; specificity = 86.4%). Our results show that it is possible to identify individuals with certain conditions using medical-administrative databases without collecting additional information. In addition to being used for epidemiological surveillance, since these data are collected systematically, they are available for both respondents and non-respondents to a study and can be used to address non-response.