Agreement Between Patient and Surrogate Assessments of Pre-Critical Illness Physical Function.

Abstract

We know little of the impact that pre-critical illness physical function plays in the recovery from critical illness. The agreement between patients and surrogates on physical function questionnaires is unclear. Prospective observational cohort conducted at a tertiary medical center. We enrolled patient-participants who were treated for respiratory failure, shock, and/or sepsis and their well-chosen surrogates. None. We assessed patient-participant physical function anchored on the 4 weeks before admission using the virtual Short Physical Performance Battery (vSPPB); Strength, Assistance in Walking, Rising from a Chair, Climbing Stairs, Falls (SARC-F) questionnaire; and Fatigue, Resistance, Ambulation, Illness, and Loss of weight (FRAIL) questionnaire completed by both patient and surrogate participants. The primary outcome was agreement between patient and surrogate responses on composite scores from each questionnaire. Secondary outcomes included correlations between patient and surrogate responses, and the relationship between surrogate and patient responses. We enrolled 75 patient-surrogate dyads. Patient-participants had a median (interquartile range) age of 61 (52-69), a Charlson Comorbidity Index of 2 (1-2), and a baseline activities of daily living disability score of 3 (1-7). Most surrogates were spouses (68%), who spent an average of 20 hours per day (12-24 hr/d) with patients. We found fair agreement between patient and surrogate scores on the vSPPB (intraclass correlation coefficient [ICC], 0.52; 95% CI, 0.31-0.69), SARC-F (ICC, 0.43; 95% CI, 0.22-0.60), and FRAIL (ICC, 0.45; 95% CI, 0.24-0.63). We found moderate correlation between patient and surrogate scores on the vSPPB, SARC-F, and FRAIL (Spearman r 0.60, 0.46, and 0.53, respectively [ p < 0.001]). We found fair agreement and moderate correlation on previously validated self-report questionnaires of physical function. Patients reported better function than their surrogates particularly at lower levels of function. Our findings highlight the need for further development of tools to assess pre-critical illness physical function.