Evaluating the Influence of User-Complexity on Individual Preferences for Upper Limb Prosthetic Selection: Measurement Tool Development for a Discrete Choice Experiment
Author(s)
Ruben Vargas, MPH1, Todd Castleberry, MS, PhD2, Kristin Nalivaika, MOT, OTD2, Elizabeth Gress, MPA1, Jasmine Khan, MPH2, Michael P. Douglas, MS1, Shane R. Wurdeman, MS, PhD2, Leslie Wilson, BS, MS, RN, PhD1.
1University of California, San Francisco, San Francisco, CA, USA, 2Hanger Institute for Clinical Research and Education, Austin, TX, USA.
1University of California, San Francisco, San Francisco, CA, USA, 2Hanger Institute for Clinical Research and Education, Austin, TX, USA.
Presentation Documents
OBJECTIVES: Upper limb loss (ULL) significantly impacts daily living, and prosthetic selection is highly individualized. Limited understanding of risk-benefit trade-offs across user-defined prosthetic complexity levels contributes to low use and misalignment between prosthesis expectations and experiences. This research aims to develop a choice-based conjoint (CBC) tool using discrete choice methodology to capture risk-benefit trade-offs and explore how user-complexity levels influence preferences.
METHODS: We designed and pilot-tested the CBC measurement tool to assess prosthetic preferences across four user-complexity categories: 1) Low: passive device, 2) Some: hook and shoulder harness, 3) Moderate: hybrid harness/myoelectric, and 4) High: myoelectric full. Key attributes were identified through a scoping literature review and refined through one-on-one interviews with 8 ULL individuals and 11 content experts. Using Sawtooth Software, we generated a full-profile, balanced-overlap experimental design that included training, random, and fixed-choice tasks. Pilot analysis assessed attribute-level selection frequency and fixed-choice responses across user-complexity categories.
RESULTS: The CBC included 8 attributes; 3 benefits (functional usefulness, life goals, and noticeability) and 5 risks (weight/comfort, durability/repairs, concentration/energy required, body connection, and training), with 14 random and 6 fixed-choice paired tasks. Preliminary results (N=54) indicated that for functional usefulness, attribute-level selection frequencies were greater for high user-complexity prosthetics (62.1%) when compared to low (31.0%). Individuals preferred hook/shoulder harness 70% over low-complexity (passive devices), 56% over moderate-complexity (hybrid harness/myoelectric), and 63% over high-complexity (myoelectric full). Highest user-complexity prosthetics were least preferred, with only 46%, 37%, and 26% selecting it over low, some, and moderate-complexity devices.
CONCLUSIONS: The CBC tool captured user preferences, facilitating informed decision-making by allowing the weighing of risk-benefit trade-offs when selecting a prosthetic. Preliminary results aligned with expected patterns across attribute levels, highlighting the tool’s potential to guide prosthetic selection based on prosthetic user-complexity preference levels.
METHODS: We designed and pilot-tested the CBC measurement tool to assess prosthetic preferences across four user-complexity categories: 1) Low: passive device, 2) Some: hook and shoulder harness, 3) Moderate: hybrid harness/myoelectric, and 4) High: myoelectric full. Key attributes were identified through a scoping literature review and refined through one-on-one interviews with 8 ULL individuals and 11 content experts. Using Sawtooth Software, we generated a full-profile, balanced-overlap experimental design that included training, random, and fixed-choice tasks. Pilot analysis assessed attribute-level selection frequency and fixed-choice responses across user-complexity categories.
RESULTS: The CBC included 8 attributes; 3 benefits (functional usefulness, life goals, and noticeability) and 5 risks (weight/comfort, durability/repairs, concentration/energy required, body connection, and training), with 14 random and 6 fixed-choice paired tasks. Preliminary results (N=54) indicated that for functional usefulness, attribute-level selection frequencies were greater for high user-complexity prosthetics (62.1%) when compared to low (31.0%). Individuals preferred hook/shoulder harness 70% over low-complexity (passive devices), 56% over moderate-complexity (hybrid harness/myoelectric), and 63% over high-complexity (myoelectric full). Highest user-complexity prosthetics were least preferred, with only 46%, 37%, and 26% selecting it over low, some, and moderate-complexity devices.
CONCLUSIONS: The CBC tool captured user preferences, facilitating informed decision-making by allowing the weighing of risk-benefit trade-offs when selecting a prosthetic. Preliminary results aligned with expected patterns across attribute levels, highlighting the tool’s potential to guide prosthetic selection based on prosthetic user-complexity preference levels.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
PCR198
Topic
Patient-Centered Research
Topic Subcategory
Health State Utilities, Instrument Development, Validation, & Translation
Disease
No Additional Disease & Conditions/Specialized Treatment Areas