USING DISCRETE CHOICE EXPERIMENTS (DCE) TO ESTIMATE PREFERENCE-BASED UTILITIES FOR DUPUYTREN'S CONTRACTURE (DC)
Author(s)
Gu NY1, Botteman M1, Gerber R2, Ji X1, Postema R3, Wan W1, Sianos G4, Anthony I4, Cappelleri JC5, Szczypa P6, Van Hout B71Pharmerit International, Bethesda, MD, USA, 2Pfizer, Inc., New London, CT, USA, 3Pharmerit International, Rotterdam, -, Netherlands, 4Glasgow Royal Infirmary, Glasgow, United Kingdom, 5Pfizer, Inc., Groton, CT, USA, 6Pfizer, Inc., Surrey, United Kingdom, 7Pharmerit International, York, United Kingdom
OBJECTIVES: Descrete Choice (DC) is a proliferative connective tissue disorder in which fingers bend towards the palm and cannot be fully extended. DC’s quality of life impact varies according to number/severity of affected fingers. To overcome this heterogeneity, DCE methods were used to estimate DC utilities.METHODS: A sample of the United Kingdom general population was invited, via Internet, to state preferences for varying hypothetical DC hand severity, defined using a combination of contractures (0°, 45° or 90°) in 8 joints [proximal interphalangeal (PIP) and metacarpophalangeal (MCP) joints of the index, middle, ring and little fingers]. Using an optimal DCE design, 56 DC hands were selected from the 6,561 (=38) possible profiles. Right-handed, left-handed, and ambidextrous respondents were asked to compare 10 randomly selected profile pairs and to indicate which hand, within each pair, was preferable. To facilitate consistent comparisons, anatomically precise digital hand drawings were used. To anchor preferences on a traditional 0-1 utility scale, participants indicated how having the most severe hand (all joints bent 90°) would affect their EQ-5D-5L profile. Unaffected hands were assumed to have a utility of 1.0. Conditional logistic models were used to estimate indirect utility weights for the joints, and were subsequently rescaled to EQ-5D-5L. RESULTS: A total of 1745 participants provided valid and complete responses. Anchoring utilities for the most severe dominant, non-dominant, and ambidextrous hands were 0.4909, 0.5688 and 0.6343, respectively. For a dominant hand with 90° contracture in MCP joints of the ring and little fingers, the estimated utility was 0.8848. If only the MCP joint of the ring finger was affected, the utility was 0.9477 and increased to 0.9738 if the contracture was 45° instead.CONCLUSIONS: DC may be associated with relatively large utility decrements. The model provides a convenient and useful tool to estimate any preference-based DC utilities for clinical and economic appraisals.
Conference/Value in Health Info
2012-06, ISPOR 2012, Washington, D.C., USA
Value in Health, Vol. 15, No. 4 (June 2012)
Code
PMS91
Topic
Methodological & Statistical Research
Topic Subcategory
PRO & Related Methods
Disease
Musculoskeletal Disorders