Initial validation of the global assessment of severity of illness

This study assessed the validity of the Global Assessment of Severity of Illness (GASI), a single-item scale designed for quick and simple assessment of illness severity in children with various chronic physical conditions. Study objectives were to examine validity, reliability, and responsiveness of the GASI. Clinicians assessed the severity of asthma, food allergy, epilepsy, diabetes, and juvenile arthritis in 55 children, and parents reported on children ’s health-related quality of life. Area under the curve (AUC) computed by logistic regression and Kendall’s Tau-c (τc) assessed the strength of association between the GASI and other study measures. The  kappa coefficient (κ), weighted kappa (κw), and McNemar ’s test assessed stability in GASI ratings over time. The standardized response mean and Guyatt’s responsiveness index examined internal and external responsiveness of the GASI, respectively. The GASI correlated strongly with established severity scales (AUC = 0.83–0.96; τc = 0.57–0.78) and did not correlate with health-related quality of life (τc <  0.1). Moderate (κw = 0.57) to substantial (κ = 0.79) test–retest reliability was supported and the magnitude of responsiveness was large (d = 0.83–3.83). This study provided initial evidence of excellent construct validity and responsiveness in the GASI and also supported acceptable test–retest rel iability in a clinical sample of children with physical conditions. Futu...
Source: Health Services and Outcomes Research Methodology - Category: Statistics Source Type: research