OBJECTIVES: Medication Possession Ratio (MPR) is the most commonly used method for calculating adherence to single-medication from pharmacy dispensing records. In general, MPR is calculated by dividing the total days’ supply of a medication by the number of days in the observation period; however, there are many variations of MPR used in the literature. This research aims to perform a comparative analysis of different MPR measures for adherence to single-medication and to demonstrate the impact of the length of the observation period and the number of minimum dispensing events on MPR. METHODS:  A hypothetical database containing 1-year dispensing records of 150 subjects for one chronic medication was constructed. Four alternative measures of MPR were compared applying 365-day, and 180-day observation periods. Each analysis was repeated including only subjects with at least one, or two dispensing events during the observation period. Analysis was carried out in R 3.2.5. Statistical comparisons were performed using Mann-Whitney U test. RESULTS: Significant differences were found in the results of the four evaluated MPR measures (all P<0.05). Differences were greater when applying shorter observation period or measurement strategies that required less dispensing events for inclusion. Change of the length of the observation period or the number of minimum dispensing events resulted in notable difference in the results of a given MPR measure (for most P<0.05). CONCLUSIONS:  Differences in the calculation methods of MPR have a significant effect on adherence estimates. Results of studies that use different MPR measures or the same MPR formula but different measurement strategies (i.e., difference in the length of the observation period or the number of minimum dispensing events) are not comparable. These findings highlight an important methodological issue which is inter alia relevant for the data synthesis of MPR results of individual studies for systematic literature reviews and meta-analysis.