How do patients with primary hypertension assess different endpoints of their treatment? a survey using analytic hierarchy process

Effects of antihypertensive therapy are estimated in clinical trials. There is a need to prioritize the endpoints according to patients' preferences. 26 patients from two regions of Germany rated in 2019 their preferences regarding the importance of various endpoints of hypertension treatment (Mortality, Myocardial infarction, Stroke, Heart failure, and subdivided Adverse events) by a pairwise comparison of individual endpoints. Analytic Hierarchy Process (AHP), a multi-criteria decision analysis method was used to generate relative weights for each endpoint. The robustness of the results was defined by means of consistency. The elicitation yielded the following aggregated group weights: Stroke 0.320, Mortality 0.297, Myocardial infarction 0.202, Heart failure 0.119, and Adverse events 0.062, subdivided in Dyspnea, Pain, Edema, and Cough. The overall consistency reached for efficacy endpoints a consistency ratio below 0.1 (safety endpoints = 0.04) without exceeding established limits. In all sensitivity analyses but one, no rank reversal was observed, and Stroke was rated highest. Individual weights varied extensively. Some participants weighted Mortality (0.021-0.686) higher than Stroke (0.078-0.615) and Heart failure (0,021-0,469) higher than Myocardial infarction (0,047-0.431). Individual inconsistency exceeded the limits in almost half of the cases, with gender, therapy duration, and therapeutic scheme being explaining variables for inconsistency within binary logistic regression models. AHP can be used to obtain preferences of patients with primary hypertension for effectiveness and safety endpoints. Preference elicitation could provide important information for drug assessment (group weights) and shared decision-making (individual weights) following the concept of patient-centeredness at system and patient level.