Response Shifts in the Canadian Occupational Performance Measure: A Convergent Mixed-Methods Study.

Abstract

Importance: A response shift (RS) is a phenomenon in which there is an individual perceptual gap between pre and post assessments. RS effects were not considered in the Canadian Occupational Performance Measure (COPM) development process.

Objective: To detect the effects of RS on the COPM.

Design: Convergent mixed-methods research.

Setting: Subacute rehabilitation hospital in Japan.

Participants: Nineteen adult patients with a range of neurological and musculoskeletal conditions recruited from a subacute rehabilitation hospital.

Outcomes and measures: In the qualitative analysis, patients' perceptions regarding occupation identified by the COPM were compared between the initial assessment (Time 1 [T1]) and a reassessment (Time 2 [T2]). In the quantitative study, patients were asked to re-rate the occupations in which the RS had occurred, giving feedback on their perceptions at T1 (T2'). The difference between T2 and T2' was calculated to clarify the magnitude of the RS.

Results: Of the 19 patients, 18 had an RS in at least one occupation. The RS effects were classified into five categories: Replacing, Adding, Reducing, Unspecified, and Embodiment. Ninety occupations were extracted from all the patients, and 46 (51.1%) were affected by RS. The percentages of occupations for which the change in score due to RS exceeded the minimal clinically important difference (±2 points) was 26.1% (12 of 46) for COPM-Performance scores and 30.4% (14 of 46) for COPM-Satisfaction scores.

Conclusions and relevance: Diverse RS effects have been identified in the COPM, which also affect score interpretation. Plain-Language Summary: The Canadian Occupational Performance Measure has a potential measurement bias that is due to a response shift in which there is an individual perceptual gap between pre and post assessments. The results of this study reveal a need to establish more accurate measurement methods to reduce the impact of response shifts on COPM scores.