Remote Assessment of Wrist Range of Motion: Inter- and Intra-Observer Agreement of Provider Estimation and Direct Measurement With Photographs and Tracings.

PURPOSE

Information is limited regarding the validity and reliability of measurements made during remote assessment of wrist range of motion (ROM) motion. We sought to determine intra- and inter-observer agreement among visual estimation, direct goniometric measurement, and patients' self-taken digital photographs and line tracings by comparing the degree differences among measurements. We hypothesized that inter- and intra-observer differences would be less than 10° at least 90% of the time for all measurement modalities.

METHODS

Thirty-seven patients were enrolled in this prospective cohort study. Visual estimation immediately followed by direct goniometry of maximal active wrist ROM (extension [E], flexion [F], radial deviation, and ulnar deviation) were independently and blindly assessed by 3 different providers: a hand surgeon, a hand therapist, and an orthopedic resident. Self-taken photographs and line tracings were blindly evaluated 3 weeks later. Intra- and inter-observer agreement was described using the Bland-Altman method.

RESULTS

The surgeon and hand therapist observed intra-observer agreement within 10° for visual estimation of all 4 directions of motions greater than 90% of the time, but inter-observer agreement for E/F was lower (76% to 86%). Intra-observer agreement by the resident was within 10° 78% of the time for E/F. Intra-observer agreement for photographs and tracings were lower than visual estimation for all observers. Inter-observer agreement for photographs and tracings was higher than intra-observer agreement. The surgeon and hand therapist agreed within 10° at least 76% of the time, the surgeon and resident agreed within 10° at least 62% of the time, and the hand therapist and resident agreed within 10° at least 54% of the time.

CONCLUSIONS

Visual estimation may be a valid method of remote assessment, but compared with goniometry, measurements may be susceptible to observer bias. Self-taken photographs and line tracings are unreliable, perhaps falsely lower owing to submaximal effort from task distraction, and we question their current use for remote assessment of wrist ROM.

CLINICAL RELEVANCE

These results represent an initial step in evaluating potential methods of remote assessment of wrist ROM.