The prevalence of Parkinson's disease (PD) in sub-Saharan Africa is still a matter of debate. The few studies that have been conducted have shown prevalences lower than those in the western world. Whether this represents a genuine finding is unclear to date. In northern Tanzania, we have conducted a hospital-based evaluation and a community-based door-to-door study to assess the prevalence of Parkinsonism, including that of PD. Over a period of 8 months, all patients admitted to a mid-size rural hospital were screened for PD. In parallel, 1,569 people aged >or=50 years were recruited from the communities and assessed for PD with standard questions. Sampling was performed according to the method of "multistage cluster sampling." The questions had previously been tested in a pilot study prior to the survey. People who screened positive were examined by a specialist neurologist. In the hospital, eight of 740 people with neurological diagnoses had Parkinsonism, of whom three patients had a diagnosis of PD. In the community-based study, 18 people answered positively to least one of the 12 screening questions. However, the diagnosis of PD could not be confirmed by further examination in any of them. The prevalence of PD in northern Tanzania was found to be very low. This result would need confirmation in studies with larger populations, ideally of different African ethnicities.