A Newly Developed Method for Assessing Exposure and Adherence to Free Dose Combinations: A Proof-of-Concept Using Antihypertensive Drugs in Danish Administrative and Healthcare Registers

Background: Non-adherence to hypertensive treatment poses a significant risk for cardiovascular diseases (CVDs), a leading cause of global mortality. International guidelines advocate for combination antihypertensive therapy as an initial approach to attain treatment objectives. Early identification of patients nonadherent to such therapy can be life-saving.

Objectives: This study proposes a novel method to assess exposure to free dose combinations in secondary data sources, aiming to calculate adherence to antihypertensive treatments. Additionally, it offers insights into the top most commonly prescribed free and fixed-dose combinations of antihypertensive treatments during different stages among treatment initiators in Denmark.

Methods: Utilizing Danish registry databases, a cohort of older individuals (≥ 65 years) commencing antihypertensive medications underwent a 12-month washout period before being recruited for the study and followed for a total of 730 days. The new method involved three steps: 1) estimation of medication event duration using the Sessa Empirical Estimator, 2) assessment of co-exposure to free dose and fixed-dose antihypertensive combinations, and 3) creation of a final treatment episode for each individual by summing up co-exposures of free-dose and fixed-dose combinations. The proportion of days covered (PDC) was estimated.

Results: A total of 1,021,819 individuals aged 65 years or above, who redeemed antihypertensive drugs between 1996 and 2016, were enrolled in the study. The method demonstrated high accuracy in classifying the start and end of the co-exposure period, the observational window, and the PDC, achieving an accuracy rate of 84.88%. In cases where misclassifications occurred, the median number of misclassified exposure days was 22 (with an IQR of 13-60 days), highlighting areas for potential refinement. Free dose combinations of diuretic + CCB, diuretic + ACEis, and CCB + ACEis were the most prevalently observed throughout the entire study period and were consistently prioritized by healthcare providers over the years. We observed a median PDC of 36% (interquartile range 14.-79%).

Conclusions: In conclusion, this study introduces an innovative approach with promising implications for evaluating co-exposure to treatments within administrative and healthcare registers. By applying this method to antihypertensive treatment, we have gained valuable insights into the landscape of hypertension treatment strategies over a span of more than two decades in Denmark.