Identifying outcome measures for atrial fibrillation value-based contracting using the Delphi method.

Abstract

Value-based contracts that tie payments for pharmaceuticals to predefined outcomes aim to promote value through shared risk and aligned incentives between manufacturers and payers. We conducted a Delphi study among diverse stakeholders (patients, providers, payers, pharmacy benefits managers, pharmaceutical company representatives) to identify top meaningful outcomes for inclusion in value-based contracts for atrial fibrillation medications. The final panel (n=55) rated the importance of each outcome on a 5-point Likert scale and selected their top 3 most meaningful outcomes. Non-patient participants rated the feasibility of collecting each outcome on a 5-point Likert scale. Consensus was defined as ≥75% agreement (Likert scores ≥4/5 or selection of an outcome as most meaningful). Differences between stakeholder groups were examined using Fisher's Exact Test. Consensus was achieved for importance of 10 outcomes (Likert scale), where "preventing stroke or mini-stroke" reached 100% agreement (55/55). Eighty-one percent (44/54) of participants selected "preventing stroke or mini-stroke" as the most meaningful outcome (rank order question). The measures rated as most feasibly collected were "reducing hospitalizations" (97%, 36/37) followed by "preventing stroke or mini-stroke" and "reducing emergency department visits" (both 92%, 34/37). There were statistically significant differences between patients and non-patients [0% (0/17) vs 22% (8/37), P=0.047] and patients and providers [0% (0/17) vs 39% (7/18), P=0.008] in selection of "improving health-related quality of life" as a most meaningful outcome. These findings will inform the design of atrial fibrillation value-based pharmaceutical contracts and provide additional insight into preferences for outcomes which could be used to improve the quality of atrial fibrillation care.