Effect of disease management on prescription drug treatment: what is the right quality measure?

Measures of medication adherence have become common parameters with which disease management (DM) programs are being evaluated, leading to the question of how this concept should be measured in the particular context of a DM intervention. We hypothesize that DM improves adherence to prescriptions more than the rate with which prescriptions are being filled. We used health plan claims data to construct 13 common measures of medication adherence for five chronic conditions. The measures were operationalized in three different ways: the Prescription Fill Rate (PFR), which requires only one prescription; the Medication Possession Ratio (MPR), which requires a supply that covers at least 80% of the year; and the Length of Gap (LOG), which requires no gap greater than 30 days between prescriptions. We compared results from a baseline year to results during the first year of a DM program. Changes in adherence were quite small in the first year of the intervention, with no changes greater than six percentage points. In the intervention year, three measures showed a significant increase based on all three operational definitions, but two measures paradoxically decreased based on the PFR. For both, the MPR and the LOG suggested either no change or significant improvement. None of the MPR and LOG measures pointed toward significantly lower compliance in the intervention year. Different ways to operationalize the concept of medication adherence can lead to fundamentally different conclusions. While more complex, MPR- and LOG-based measures could be more appropriate for DM evaluation. Our initial results, however, need to be confirmed by data covering longer term follow-up.