PROPREDICT Decision Support Tool: Using Evidence to Guide Precision Prosthesis Prescription and Rehabilitation Planning.

Objective: To translate the previously validated AMPREDICT PROsthetics prediction model into a web-based, clinical decision support tool (DST) for use at the time of initial prosthetic evaluation to predict 4 levels of prosthetic mobility 12 months after prescription.

Design: The previously validated AMPREDICT PROsthetics prediction model provided the predictors with corresponding coefficients to develop an online DST. The content and aesthetics of the DST was created by the authors with iterative input from an expert panel gathered via a user-friendly input page and access to a DST simulation with graphic display of predicted mobility levels. This beta DST underwent formal usability testing using a think aloud approach with a quantitative assessment of usability and qualitative interviews to address user-friendliness, readability, functionality, and potential implementation challenges.

Setting: The Veterans Health Administration.

Participants: Twelve clinicians (N=12), who regularly participate in prosthetic prescription and represent the intended target users of the PROPREDICT DST, were included.

Interventions: Not applicable.

Main outcome measures: Qualitative themes and the Post-Study System Usability Questionnaire (PSSUQ).

Results: The PSSUQ overall (and subscale scores) were highly favorable, with a mean overall score of 1.58 from a potential range 1.0-7.0 (lower score greater usability). Participant feedback identified the following potential clinical utility of the DST: (1) assisting in counseling patients on prosthesis selection; (2) setting expectations for future mobility; and (3) sharing DST input with the clinical team to aid in rehabilitation planning.

Conclusions: The PROPREDICT DST (www.prodecide.org) was successfully developed and tested for usability and clinical relevance. Rehabilitation providers are optimistic about its potential.