An Interrater Reliability Study of the Nottingham Stroke Dressing Assessment

Abstract

Introduction:

Dressing problems after stroke are common. There is therefore a need to establish the psychometric properties of appropriate dressing outcome measures for use in clinical practice and research. The Nottingham Stroke Dressing Assessment (NSDA) is a validated assessment of post-stroke dressing ability. This study investigated the interrater reliability of the NSDA and the accompanying dressing error analysis form.

Method:

Twenty patients with persistent dressing difficulties at 2 weeks post-stroke were recruited to the study. Two research occupational therapists and one clinical occupational therapist acted as raters. They observed all 20 patients during dressing and independently completed the NSDA and error analysis form without discussion.

A kappa (κ) test was performed to assess the level of agreement between the three raters on each dressing item assessed.

Results:

Of the 44 items tested, there was excellent agreement (κ > 0.75) on 29 items, good agreement (κ>0.6) on 8 items, fair agreement (κ>0.4) on 5 items and poor agreement (κ <0.4) on 2 items. The intraclass correlation coefficient between the three raters' final percentage score was 0.988, representing excellent agreement. The error analysis form also proved to have interrater agreement.

Conclusion:

The NSDA and error analysis form can be considered to have psychometrically proven interrater reliability.