Abstract Introduction The Barthel Index (BI) is widely used to assess functional independence after critical illness. Although it demonstrates strong reliability and prognostic value, some studies suggest that it may be highly specific but poorly sensitive in detecting subtle or multidimensional impairments in health-related quality of life (HRQoL). Patients may regain autonomy in basic activities of daily living (ADL) yet continue to experience fatigue, pain, or psychological limitations not captured by the BI. We tested the hypothesis that ICU survivors would show a discordance between functional recovery (BI) and perceived health evaluated through the SF-36 and EQ-5D-3L questionnaires. Method A cross-sectional study was conducted including 52 survivors of respiratory critical illness who required mechanical ventilation. Assessments were performed 6-12 months after ICU discharge using the Barthel Index (0-100), the SF-36 (domains: Physical Functioning PF, Role Physical RP, Bodily Pain BP, General Health GH, Vitality VT), and the EQ-5D-3L (utility index UTIL5D, 0-1 and visual analogue scale VAS, 0-100). Clinical discordance was defined as BI ≥ 90 combined with (a) PF 50 or GH 50 (SF-36), (b) UTIL5D 0.80, or (c) VAS 50. t-tests and Pearson correlations were used to compare BI with each PROM domain. Results Mean (±SD) scores were: BI 91.3±14.1; PF 76.4±25.3; RP 46.6±47.2; BP 60.3±24.4; GH 47.3±24.8; VT 44.1±20.7; UTIL5D 0.695±0.213; VAS 0.734±0.186; Age 52.7±15.6; APACHE II 14.8±4.4; SOFA 7.3±2.9; ICU stay 12.6±7.8 days; and Mechanical Ventilation 9.4±5.6 days. Correlations between BI and SF-36 domains were: PF r = 0.80 (p 0.001), RP r = 0.56, BP r = 0.57, GH r = 0.42, and VT r = 0.43. Clinical discordance was observed in 3.8% (PF 50), 32.7% (GH 50), 59.6% (UTIL5D0.80), and 75.0% (VAS50) of patients. Discordance was more frequent among younger survivors (65 years: 41.7% vs ≥ 65 years: 12.5% for GH 50). Conclusion Despite recovering functional independence (BI ≥ 90), a substantial proportion of ICU survivors report poor perceived health and quality of life. These findings suggest that the BI is specific but not sensitive in detecting post-ICU HRQoL impairments. Integrating PROMs such as SF-36 and EQ-5D-3L (including VAS) may improve detection of residual physical, emotional, and social limitations, guiding patient-centered rehabilitation strategies. This abstract is funded by: None
Morales et al. (Fri,) studied this question.