I SURVIVE: inter-rater reliability of three physical functional outcome measures in intensive care unit survivors

We prospectively assessed inter-rater reliability of three physical function measures in Canadian intensive care unit (ICU) survivors in the inpatient setting.

We enrolled patients who had an ICU length of stay of ≥ three days, were mechanically ventilated for ≥ 24 hr, and were ambulating independently before hospital admission. Weekly from ICU discharge to hospital discharge, two trained frontline clinicians, blinded to each other's findings, independently performed the Physical Function ICU Test-scored (PFIT-s; score out of 10), 30-sec sit-to-stand (30STS; # of stands), and two-minute walk test (2MWT; distance in m), all within 24 hr. We calculated the intraclass correlation coefficient (ICC), standard error of measurement (SEM), and minimal detectable change (MDC₉₀).

We enrolled 42 patients. PFIT-s: in 36 patients with 66 paired scores, the ICC was 0.78 (95% confidence interval [CI], 0.66 to 0.86), the SEM was 1.04, and the MDC₉₀ was 2.42. 30STS: in 35 patients with 67 paired scores, the ICC was 0.85 (95% CI, 0.76 to 0.90), the SEM was 1.91, and the MDC₉₀ was 4.45. 2MWT: in 35 patients with 58 paired scores, the ICC was 0.78 (95% CI, 0.66 to 0.87), the SEM was 20.87, and the MDC₉₀ was 48.69.

These three measures show good inter-rater reliability when used by trained frontline clinicians to assess physical function in ICU survivors in the inpatient setting.