ISPOR Conjoint Analysis Experimental Design Good Research Practices
This guideline helps researchers design surveys that ask patients or healthcare providers to choose between different treatment options. It ensures the choices are presented fairly and the results reliably show what people value in treatments.
At a glance
Use when
Designing discrete-choice experiments to elicit patient or provider preferences in health care settings
Avoid when
When simpler preference elicitation methods (e.g., rating scales) are sufficient or when resources for complex survey design are limited
Inputs
Health care attributes and levels of interest, research objectives, target population characteristics
Outputs
Validated experimental design for a discrete-choice experiment, including choice tasks and design properties
How it works
Provides methodological guidance for constructing experimental designs in discrete-choice experiments (DCEs) used in health care research. Covers attribute selection, level definition, experimental design types (e.g., full factorial, fractional factorial, D-optimal), and strategies to reduce respondent burden while maintaining statistical efficiency. Intended to improve the validity and reliability of preference data collected via conjoint analysis.
- HTA domains
- Clinical Effectiveness, Costs & Economic Evaluation, Patient and Social Aspects
- Assumptions
- Respondents can meaningfully trade off attributes in hypothetical scenarios; attribute levels are well-defined and relevant to the population
- Strengths
- Promotes methodological rigor in preference studies; supports generation of reliable, interpretable data for decision-making; widely accepted in health economics
- Limitations
- Requires expertise in experimental design and statistics; may not capture real-world behavior perfectly due to hypothetical nature of choices
- Also known as
- ISPOR Conjoint Analysis Design Guidelines, ISPOR Good Research Practices for Conjoint Analysis
Questions this answers
- › How should attributes and levels be selected for a discrete-choice experiment in health care?
- › What experimental design types are appropriate for health care conjoint studies?
- › How can survey length be minimized without compromising data quality?
- › How can dominance and overlap in choice tasks be avoided?
- › What are best practices for ensuring statistical efficiency in DCEs?
- › How can researchers handle interactions between attributes in the design phase?
References & sources
Related methods
Similar by meaning
- ISPOR Statistical Methods for the Analysis of Discrete-Choice Experiments
- Discrete Choice Experiments in Health Economics: Past, Present and Future
- ISPOR Prospective Observational Studies to Assess Comparative Effectiveness Good Research Practices Task Force Report
- ISPOR Roadmap for Patient-Preference Studies in Decision Making
- ISPOR Retrospective Database Analysis Good Research Practices — Part III (Analytic Methods)
Beta record. Based on the original catalogue summary; primary-source enrichment pending.