OBJECTIVES: To investigate the prevalence of unacceptable refill-based adherence rates with antiparkinson medicine items. METHODS: A retrospective drug utilisation study was performed on medicine claims data of a pharmacy benefit management company in South Africa during January 1, 2005 until December 31, 2008. Refill-based adherence rates were calculated for 8 768 antiparkinson medicine items that were prescribed more than once during a four-year period (January 1, 2005 to December 2008). The refill-based adherence rate was calculated per trade name by using the following equation: Refill-Adherence rate = (total number of days of antiparkinson medicine items supplied – days supplied at the last refill) / (date last claimed – date first claimed). (RSA Rand(R)/$US = 6.8595 on 31 Dec. 2007). [RSA Rand (R)/$US = 6.38112 (2005); 6.78812 (2006); 7.06926 (2007) and 8.27505 (2008)]  RESULTS: A majority of antiparkinson medicine items (53.50%, n = 4,691) had unacceptably low refill-adherence rates below 90%, that accounted for 41.62% (n = R16,398,512.00) of the total cost (N = R39,402,898.20) of all antiparkinson medicine items included in this study. Only 36.78% (n = 3225) of antiparkinson medicine items had acceptable refill-adherence rates between 90% and 110%. Those with unacceptable high refill-adherence rates accounted for 9.72% (n = 852) of all antiparkinson medicine items and represented 6.5% (n = R2,574,597) of the total cost. No practical significant difference in the average refill-adherence rates was found between male (93.99% ± 186.99) and female (90.83% ± 175.21) patients. Biperidien, carbidopa/levodopa, and levodopa/benserazide containing products had on average unacceptable low refill-adherence rates (<90%).  CONCLUSIONS: Although poor obedience to treatment schedules adds up to aggravation of Parkinson’s disease leading to death and amplified health care cost, it seems that the refill-adherence rate of antiparkinson medicine items is not very favourable