Prevalence of parkinsonism estimated using the drug prescription archive: a possible method to estimate the prevalence of a chronic neurological disease?

Many surveys estimated prevalence of parkinsonism, with results varying largely. We used prescription records of medications for parkinsonism to estimate the prevalence of this condition. Retrospective survey based on Lazio (Italy) regional drugs' prescriptions registry. Cases of parkinsonism were defined as those who received a medication for parkinsonism (Dopa and dopa derivatives or Monoamine oxidase B-inhibitors) for at least 6 months in a 5-year period (2005-2009). Crude and standardized prevalence rates at June 2009 were calculated. Crude and standardized prevalence rates of parkinsonism in Lazio were, respectively, 283 per 100,000 (95 % CI 278-287), and 294 per 100,000 (95 % CI 289-298), higher in men than in women (292 per 100,000 vs. 274 per 100,000). The highest overall prevalence rate was observed among people aged 85-89 years (246 per 100,000), while the lowest in subjects aged <65 (38 per 100,000). Prevalence rates in people older than 65 and older 75 were, respectively, 1275 per 100,000 and 1912 per 100,000. Using a regional drug registry, based on currently available health information systems, may be a suitable method to estimate prevalence of parkinsonism, which is essential for public health programming, the more in presence of a demographic shift as the current one.