Table 2 Conjoint analysis applications in health: a checklist offered by the International Society for Pharmacoeconomics and Outcomes Research [20]
Research question and hypothesis clarity: A sound and testable hypothesis underpinning a research question that is focusing on some aspect of patient health or care. This hypothesis has to be back by existing literature and grounded through some formative qualitative research. Understanding attributes and levels: Conjoint analysis focuses on elicitation of preferences or values over the range of attributes and levels that define key domains in the conjoint-analysis tasks. The attribute levels should encompass the range that may be salient to participants, even if those levels are hypothetical or not feasible in each context. Construction of tasks: “Tasks” describe the choice options presented to patients from which they make their selected preference. Within a choice task, attributes and levels may be offered individually or in “profiles” where multiple attributes and levels are offered together to represent a service or product option. Thoughtful construction of tasks is helpful to understand trade-offs better. Experimental nature of the design: The goal of a conjoint-analysis experimental design is to create a set of tasks that will yield as much statistical information as possible for estimating unbiased, precise preference parameters. In accordance with the experimental nature, the design must be balanced at each level and attribute. Preference elicitation: Offering participants contextual information including motivation and explanation for the tasks helps in eliciting the right choices. It is critical that the overall design is not cognitively or semantically challenging for the participants. Pretests and expert, key stakeholder consultation is critical here. |