To Register      SMDM Homepage

Wednesday, 20 October 2004

This presentation is part of: Poster Session - Utility Theory; Health Economics; Patient & Physician Preferences; Simulation; Technology Assessment

IMPACT OF DISCUSSION ON UTILITY VALUES ELICITED IN A GROUP SETTING

Ken Stein, MB, ChB, MSc1, Tania Crabb, BSc1, John Brazier, PhD2, Julie Ratcliffe, PhD2, Ruairidh Milne, BA, MB, ChB, MSc3, and Alison Round, MB, ChB1. (1) Peninsula Medical School, Peninsula Technology Assessment Group, Exeter, United Kingdom, (2) School of Health and Related Research, Sheffield Health Economics Group, Sheffield, United Kingdom, (3) University of Southampton, National Coordinating Centre for Health Technology Assessment, Southampton, United Kingdom

Purpose: To investigate the impact of group discussion following individual preference elicitation

Methods: A group of 15 non-health professionals was established to measure preferences on a range of health states using the standard gamble method. The group met five times over six months. Health state scenarios were derived from disease specific outcome measures reported in clinical trials of health technologies. Preferences were initially elicited using the standard gamble (titration approach) without discussion in the group. Each scenario and the initial preferences expressed by group members were then discussed for five to ten minutes and participants given the opportunity to revise their initial preference following discussion. The number of changes made and their impact on the range and summary utility values for the group were calculated. The importance of group discussion was explored with participants in qualitative interviews.

Results: 40 health states relating to seven specific conditions were valued, giving 445 individual results. Mean utility values ranged -0.27 to 0.98. 14 (3.1%), changes were made in values for 12 scenarios (30%) by seven individuals. One individual changed five values. Changes ranged from -0.075 to 0.45 (mean 0.04). The impact on summary values was limited. Mean utility was affected in only 7 of the 12 scenarios. The average mean change following discussion was 0.01 (range -0.01 to 0.07). Only three median values were affected by changes (range -0.05 to 0.03). Qualitative data revealed a range of perceived benefits from discussion.

Conclusions: Preferences elicited using the standard gamble are unaffected by brief discussion in a group. Changes have no significant impact on the summary measures but participants value discussion highly.


See more of Poster Session - Utility Theory; Health Economics; Patient & Physician Preferences; Simulation; Technology Assessment
See more of The 26th Annual Meeting of the Society for Medical Decision Making (October 17-20, 2004)