RISKY BUSINESS: HOW HEALTH PREFERENCE RESEARCH CAPTURES RISK TOLERANCE
Author(s)
Tina Cheng1, January Cornelius, MPH2, Jui-Chen Yang, MEM2, Juan M. Gonzalez, PhD3;
1Duke University, Doctoral Student, Durham, NC, USA, 2Duke University, Durham, NC, USA, 3Duke Clinical Research Institute, Durham, NC, USA
1Duke University, Doctoral Student, Durham, NC, USA, 2Duke University, Durham, NC, USA, 3Duke Clinical Research Institute, Durham, NC, USA
OBJECTIVES: Risk-tolerance measures (RTMs) from health preference (HP) studies quantify the acceptability of uncertain health outcomes associated with medical interventions. While much of the discussion has focused on the diversity of preference-elicitation methods for RTM derivation, less attention has been given to potential influences of study design and analytical choices on RTMs. This scoping review aimed to identify and synthesize key study features that would affect the estimation, interpretation, and reporting of RTMs in HP research.
METHODS: A systematic search on PubMed, Scopus, and Web of Science identified quantitative HP studies examining benefit-risk tradeoffs involving probabilistic outcomes. Studies were evaluated, and data were extracted by experts in stated-preference methods, including ISPOR task force members. Elicitation method, basis for attribute selection, framing of benefits and harms, risk presentation, and assumptions regarding preference and risk-tolerance heterogeneity were among the data extracted. Reporting practices also were assessed, including the availability of survey instruments, parameter estimates, and stated limitations.
RESULTS: Of the 60 studies included, the most common elicitation methods were discrete-choice experiments (n=43) and threshold-technique methods (n=11). While all studies framed risks as decision-induced utility losses, benefits were framed as decision-induced utility gains in 43 studies, avoidance of utility losses in 9 studies, and both in 7 studies. Although most studies modeled continuous preference heterogeneity (n=36), many ultimately reported homogenous RTMs (n=21).
CONCLUSIONS: Current practices reflect broad agreement on core methods for eliciting risk tolerance in HP studies. However, we observed inconsistencies in the presentation of benefits, reporting of modeling assumptions, and estimation of RTMs. Results often reported homogenous RTMs despite accounting for heterogeneous preferences, suggesting a loss of potentially important information. Greater transparency and standardization in the presentation of uncertain outcomes and reporting are needed to ensure RTMs accurately reflect underlying preference heterogeneity and support valid interpretation and comparability of RTMs across studies.
METHODS: A systematic search on PubMed, Scopus, and Web of Science identified quantitative HP studies examining benefit-risk tradeoffs involving probabilistic outcomes. Studies were evaluated, and data were extracted by experts in stated-preference methods, including ISPOR task force members. Elicitation method, basis for attribute selection, framing of benefits and harms, risk presentation, and assumptions regarding preference and risk-tolerance heterogeneity were among the data extracted. Reporting practices also were assessed, including the availability of survey instruments, parameter estimates, and stated limitations.
RESULTS: Of the 60 studies included, the most common elicitation methods were discrete-choice experiments (n=43) and threshold-technique methods (n=11). While all studies framed risks as decision-induced utility losses, benefits were framed as decision-induced utility gains in 43 studies, avoidance of utility losses in 9 studies, and both in 7 studies. Although most studies modeled continuous preference heterogeneity (n=36), many ultimately reported homogenous RTMs (n=21).
CONCLUSIONS: Current practices reflect broad agreement on core methods for eliciting risk tolerance in HP studies. However, we observed inconsistencies in the presentation of benefits, reporting of modeling assumptions, and estimation of RTMs. Results often reported homogenous RTMs despite accounting for heterogeneous preferences, suggesting a loss of potentially important information. Greater transparency and standardization in the presentation of uncertain outcomes and reporting are needed to ensure RTMs accurately reflect underlying preference heterogeneity and support valid interpretation and comparability of RTMs across studies.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
PCR119
Topic
Patient-Centered Research
Disease
No Additional Disease & Conditions/Specialized Treatment Areas