INTRODUCTION: The objective of the study was to estimate the prevalence of parkinsonism in the Province of Rome using antiparkinsonian prescription histories from 1986 to 1991. METHODS: A subject was defined as a case of parkinsonism if he/she had received "specific" and "consistent" antiparkinsonian therapy in the study period. RESULTS: In November 1990, 6,572 patients were defined as prevalent cases of parkinsonism. The crude prevalence ratio, for the total population of the Province of Rome, is 173.5 per 100,000 inhabitants (165.9 per 100,000 in men and 180.5 per 100,000 in women). The method was validated by record-linkage with clinical records of all patients visited during 1990 at the Department of Neurological Sciences of the University of Rome "La Sapienza". The sensitivity of the prevalence study was 83.6%. CONCLUSIONS: The use of a computerized data base of all prescription data, routinely collected for administrative purposes, enabled us to obtain a prevalence estimate based on a very large population, with low costs and in a relatively short time.
INTRODUCTION: The objective of the study was to estimate the prevalence of parkinsonism in the Province of Rome using antiparkinsonian prescription histories from 1986 to 1991. METHODS: A subject was defined as a case of parkinsonism if he/she had received "specific" and "consistent" antiparkinsonian therapy in the study period. RESULTS: In November 1990, 6,572 patients were defined as prevalent cases of parkinsonism. The crude prevalence ratio, for the total population of the Province of Rome, is 173.5 per 100,000 inhabitants (165.9 per 100,000 in men and 180.5 per 100,000 in women). The method was validated by record-linkage with clinical records of all patients visited during 1990 at the Department of Neurological Sciences of the University of Rome "La Sapienza". The sensitivity of the prevalence study was 83.6%. CONCLUSIONS: The use of a computerized data base of all prescription data, routinely collected for administrative purposes, enabled us to obtain a prevalence estimate based on a very large population, with low costs and in a relatively short time.