Want to learn more about our research in medication adherence and pharmacoepidemiology?
Jump to a question:
A: No, PDC and MPR are related but distinct measures of medication adherence. MPR is typically calculated as the total days’ supply of medication dispensed divided by the number of days in the observation period, which can result in values over 100%. In contrast, PDC calculates the proportion of days within a defined period during which a patient has access to at least one dose of their medication, and it is capped at 100%. PDC is generally preferred in adherence research, particularly for chronic medications, because it better accounts for overlapping refills and is less likely to overestimate adherence.
A: The use of an 80% threshold to define adherence is common but not strongly supported by clinical or scientific evidence (Baumgartner et al. 2018; Dalli et al. 2021). This arbitrary cut-off can reduce statistical power and may not reflect meaningful differences in adherence. We recommend analysing PDC as a continuous variable whenever possible. If dichotomisation is necessary, sensitivity analyses using alternative thresholds should be performed to assess the robustness of findings and minimise the risk of type I error.
A: While some studies suggest that specific adjustments to the PDC can improve the accuracy of medication adherence estimates, we recognise that certain adjustments may not be feasible due to data limitations or study design constraints. Rather than promoting a single "default" approach, the TEN-SPIDERS tool emphasises transparency. It is essential that all adjustments made—or not made—are clearly reported using the TEN-SPIDERS tool, along with justifications for any adjustments that could not be implemented.
A: Yes. The PDC framework is suitable for measuring and reporting medication adherence for any chronic medication intended for long-term use, not just cardiovascular therapies. Visit the Exemplars page to see how the TEN-SPIDERS tool has been applied in other disease areas and clinical contexts.