Physiotherapists systematically overestimate the amount of time stroke survivors spend engaged in active therapy rehabilitation: an observational study

Abstract

QUESTIONS: How accurately do physiotherapists estimate how long stroke survivors spend in physiotherapy sessions and the amount of time stroke survivors are engaged in physical activity during physiotherapy sessions? Does the mode of therapy (individual sessions or group circuit classes) affect the accuracy of therapists' estimates? DESIGN: Observational study embedded within a randomised trial. PARTICIPANTS: People who participated in the CIRCIT trial after having a stroke. INTERVENTION: 47 therapy sessions scheduled and supervised by physiotherapists (n = 8) and physiotherapy assistants (n = 4) for trial participants were video-recorded. OUTCOME MEASURES: Therapists' estimations of therapy time were compared to the video-recorded times. RESULTS: The agreement between therapist-estimated and video-recorded data for total therapy time and active time was excellent, with intraclass correlation coefficients (ICC) of 0.90 (95% CI 0.83 to 0.95) and 0.83 (95% CI 0.73 to 0.93) respectively. Agreement between therapist-estimated and video-recorded data for inactive time was good (ICC score 0.62, 95% CI 0.40 to 0.77). The mean (SD) difference between therapist-estimated and video-recorded total therapy time, active time, and inactive time for all sessions was 7.7 (10.5), 14.1 (10.3) and -6.9 (9.5) minutes respectively. Bland-Altman analyses revealed a systematic bias of overestimation of total therapy time and total active time, and underestimation of inactive time by therapists. Compared to individual therapy sessions, therapists estimated total circuit class therapy duration more accurately, but estimated active time within circuit classes less accurately. CONCLUSION: Therapists are inaccurate in their estimation of the amount of time stroke survivors are active during therapy sessions. When accurate therapy data are required, use of objective measures is recommended.