Our objective was to estimate the prevalence of Parkinson's disease (PD) and Parkinsonism (Ps) in Antioquia (Colombia), using the Capture-Recapture method. The two biggest institutions for attending neurological patients in Antioquia were selected as sources for the use of the Capture-Recapture method. Prevalences of PD (PPD) and Ps (PPs) were estimated according to the following expression: PPD (or PPs) = n/Nl105. The number of cases (n) of PD (or Ps), n = a + b + c + d, where a = cases identified from the two sources, b = cases identified only in the first source, c = cases identified only in the second source, and d = nondetected cases from any source = bc/a. The projected Antioquian population for the year 2000 was used as denominator. Information obtained between January 1, 1996, and December 31, 2000, was reviewed in order to identify the clinical records of all patients that fulfilled the Ps or PD criteria. General prevalence of PD in Antioquia was 30.7/100.000 (C195% = 29.2-32.2), and that of Ps was 42.1/100.000 (CI95% = 40.3-43.8). Prevalence of PD in people older than 50 years was 176.4/100.000 (CI95% = 166.6-186.3) and that of Ps was of 339.6/100.000 (C195% = 326.0-353.2). Ps and PD prevalences in Antioquia were lower than the figures reported by the National Neuroepidemiologic Study (470/100.000) and similar to the estimated prevalence of these diseases in Caucasian populations (80 to 270/100.000). These findings evidence the great variability of PD prevalence in different regions; therefore, a nationwide study is necessary to determine the prevalence of PD and Ps in Colombia