A Cross-Sectional Study of the Clinical Metrics of Functional Status Tools in Pediatric Critical Illness.

OBJECTIVES: To determine the clinical metrics of functional assessments in pediatric critical illness survivors. DESIGN: Cross-sectional observational study. SETTING: PICU follow-up clinic. PATIENTS: Forty-four PICU survivors 6-12 months post PICU stay, and 52 healthy controls 0-18 years old. INTERVENTIONS: Nil. MEASUREMENTS AND MAIN RESULTS: Function was assessed using the Pediatric Quality of Life Inventory 4.0 generic scales and infant scales, the Pediatric Evaluation of Disability Inventory-Computer Adaptive Test, and the Functional Status Scale. Muscle strength was assessed by hand grip strength in children greater than or equal to 6 years. Clinical metrics assessed included floor and ceiling effects, known-group, and convergent validity. Floor and ceiling effects were present if the participants achieving the worst or best scores exceeded 15%, respectively. Known-group validity was assessed by comparing scores between those with and without complex chronic conditions and abnormal versus good baseline function. Convergent validity was assessed using partial correlation between two tools. Functional Status Scale and Pediatric Quality of Life Inventory physical domain scores showed significant ceiling effects in PICU survivors (69.2% and 15.4%, respectively, achieved the highest possible score). Functional scores were not significantly different between children with or without complex chronic conditions or children with good versus abnormal baseline function. In healthy children, Pediatric Quality of Life Inventory physical correlated moderately with hand grip strength (partial r = 0.66; p < 0.001), whereas Pediatric Quality of Life Inventory psychosocial correlated moderately with Pediatric Evaluation of Disability Inventory-Computer Adaptive Test social/cognitive score (partial r = 0.53; p < 0.001). In PICU survivors, only Pediatric Quality of Life Inventory physical and Pediatric Evaluation of Disability Inventory-Computer Adaptive Test mobility scores were correlated (partial r = 0.55; p < 0.001). CONCLUSIONS: PICU functional assessment tools have varying clinical metrics. Considering ceiling effects, Pediatric Evaluation of Disability Inventory-Computer Adaptive Test may be more suitable in survivors than Functional Status Scale. Differences in scores between children with or without complex chronic conditions, and with or without baseline functional impairment, were not observed. Functional assessments likely require a combination of tools to measure the spectrum of pediatric critical illness and recovery.