Reliability and Validity of Upper Extremity Patient-Reported Outcomes Measures in Assessing Traumatic Finger Amputation Management: What is the Best Instrument?
Alfred P. Yoon, MD, Kevin C. Chung, MD, MS.
University of Michigan, Ann Arbor, MI, USA.
Purpose: This study investigates the psychometric properties of patient-reported outcomes (PRO) instruments in post-surgical traumatic digit amputation patients. We hypothesize the MHQ (Michigan Hand Outcomes Questionnaire) and DASH (Disabilities of the Arm, Shoulder, and Hand Questionnaire) to be the most valid and reliable instruments.
Methods: Patients with traumatic digit amputation were studied as part of the Finger Replantation ANd amputation Challenges in assessing Improvement, Satisfaction, and Effectiveness (FRANCHISE) collaboration. MHQ, DASH, Patient-Reported Outcomes Measurement Information System (PROMIS), and Short Form Health Survey (SF-36) were used to assess PROs at least one year postoperatively. Internal consistency was measured by Cronbach’s alpha(α) and criterion validity with Pearson’s correlation coefficient(r). Construct validity was tested with four pre-defined hypotheses. Receiver operating characteristic(ROC) analysis was conducted to evaluate discriminant validity.
Results: 168 replantation and 74 revision amputation patients met inclusion criteria. All instruments had fair to good internal consistency in both cohorts (0.7<α<0.9). MHQ and DASH scores correlated strongly (r>0.60) in both cohorts. SF-36 had moderate to weak correlation with the remaining instruments, and its mental component had poor ability to discriminate high and low MHQ scores (area under the curve 0.63 - 0.67). MHQ, DASH, and PROMIS demonstrated good construct validity confirming 75% of predefined hypotheses, while SF-36 only confirmed 25%.
Conclusions: We recommend using MHQ and DASH when assessing PROs in finger amputation patients based on good internal consistency and validity. PROMIS has fair validity and reliability but should be an adjunct instrument. SF-36 should not be used to assess digit amputation patients.
Back to 2019 Abstracts