Abstract
BACKGROUND: Guideline-directed medical therapy (GDMT) for heart failure with reduced ejection fraction (HFrEF) reduces mortality rates but remains widely underused. A polypill for HFrEF has been proposed as an implementation strategy to improve GDMT delivery, but little is known about physicians' preferences in the design of HFrEF polypills. Discrete choice experiments (DCEs), in which survey respondents choose among hypothetical products, are a powerful tool in health economics research and can lend insight into key tradeoffs in HFrEF polypill design. However, the process of designing DCEs is complex and often poorly reported.
METHODS: We developed a survey instrument, including a DCE, through a 5-stage mixed-methods process including (1) literature review; (2) physician interviews and surveys; (3) attribute generation; (4) expert review; and (5) pilot testing. We applied a D-efficient design approach using a multinomial logic model to determine the number of choice tasks for the DCE.
RESULTS: We designed a DCE with 4 attributes: HFrEF polypill out-of-pocket cost, inclusion of an angiotensin converting enzyme inhibitor, angiotensin receptor blocker, or sacubitril/valsartan, ancillary support for polypill prescribing, and pharmacy availability. The final survey will be distributed to cardiologists in academic and Veterans Administration medical centers across the United States.
CONCLUSIONS: Through a rigorous multistage design process leveraging mixed methods, we developed a DCE that elicits cardiologists' preferences about HFrEF polypills and key tradeoffs in their design. Results from this DCE study will directly inform future HFrEF polypill cluster-randomized clinical trials.