Outcome measures report different aspects of patient function three months following critical care

Previous investigation of the relationship between physical performance and patient self-report physical function (PF) measures in intensive care unit (ICU) survivors have not been performed. To (1) analyze the extent to which other activity-based measures of physical performance may serve as proxies for the 6-min walk test (6MWT); (2) determine the extent to which the Short Form (SF) 36 domain of PF and physical component summary (PCS) score, reflect components of physical performance and (3) examine the relationship between demographic and ICU variables and the 6MWT. Cross-sectional data from two clinical trials. Two acute care hospitals (Melbourne, Australia and Denver, USA). A total of 177 survivors of ICU. Were evaluated at 3 months. Performance-based measures were: 6MWT, timed up and go test (TUG), the five times sit to stand test (5xSTS), the Berg balance scale (BBS) and two self-report measures: the SF-36 PF domain and the PCS score. 6MWT showed excellent correlation with the TUG (rho = -0.79) and BBS (rho = 0.80); and good correlation with 5xSTS (rho = -0.69) and SF-36 PF scores (rho = 0.69). 6MWT explained 54 and 33 % of variance in SF-36 PF and PCS scores respectively. No variables were clinically important in predicting 6MWT. The 6MWT and TUG may both be acceptable measures of PF performance 3 months after ICU. Performance-based tests measure different constructs than self-report measures and choice of outcome variables should be aligned with study aims to ensure the most appropriate measure is used.