This presentation is part of: Oral Concurrent Session A - Patient and Physician Behavior/Preferences 1
LUNG TRANSPLANT CANDIDATES' ESTIMATION OF POST-TRANSPLANT UTILITIES
Lianne G Singer, MD, University of Toronto, Medicine, Toronto, ON, Canada and Christine LB Gordon, MSc, University Health Network, Clinical Decision Making and Health Care, Toronto, ON, Canada.
Purpose: To determine how accurately lung transplant candidates envisage transplant recipients’ visual analog scale (VAS) and standard gamble (SG) utilities, and how these perceptions influence readiness for transplant listing. Methods: VAS and SG utilities for current health were elicited from 71 transplant candidates (median age 54, 57% male) and 99 transplant recipients (median age 51, 49% male). Additionally, candidates completed VAS and SG utilities imagining both that they had received a transplant with a good outcome (no BOS) and that they had developed chronic rejection, called bronchiolitis obliterans syndrome (BOS). Candidates who predicted higher scores for BOS than no BOS were excluded from the analysis, leaving 64 VAS and 59 SG scores. Candidates also indicated readiness for transplant listing on a Likert scale. Recipients were separated into those with BOS (n=23) and those without BOS (n=76) for derivation of actual utility values for these health states. Two-sample Wilcoxon rank-sum tests were used to compare candidates’ predicted no BOS and BOS scores with recipients’ actual no BOS and BOS scores respectively for both VAS and SG utilities. Fisher’s exact test was used to compare readiness for transplantation between patient groups. Results: Candidates overestimated recipients’ actual VAS scores for no BOS (p=0.03), although the difference was not clinically significant (median for both = 85). Candidates accurately estimated SG utilities for no BOS (median 0.91 for candidates and 0.94 for recipients, p=0.83). Candidates significantly underestimated recipient’s actual VAS (median 40 for candidates and 75 for recipients, p<0.00005) and SG (median 0.37 for candidates and 0.93 for recipients, p<0.00005) scores for BOS. Candidates with current health SG scores exceeding predicted BOS scores were less likely to indicate readiness for transplant listing than those whose SG scores for current health were lower than predicted BOS scores (52% vs. 88% ready, p=0.015). VAS scores followed a similar pattern. Conclusions: Lung transplant candidates accurately estimated VAS and SG utilities for no BOS, while significantly underestimating VAS and SG utilities for BOS. Candidates who predicted BOS to be worse than their current health were less likely to be ready for transplant listing. Some medically acceptable candidates decline transplant listing. These results suggest that inaccurate perceptions of post-transplant outcomes for BOS may be a contributing factor for those not ready to be listed.
