Our objective was to assess the psychometric performance of the Health Utilities Index 2 and 3 in a pediatric population following admission to a pediatric intensive care unit.
As part of a larger study of pediatric intensive care outcomes, children were followed up at 6 and 12 months post admission from pediatric intensive care using the Health Utilities Index 2 and 3. We evaluated and compared the psychometric performance of the Health Utilities Index 2 and 3 in this population by assessing their practicality, reliability, and limited information regarding validity.
Twenty-two pediatric intensive care units in the United Kingdom.
A total of 685 children aged 5 yrs and over.
The Health Utilities Index 2 and 3, which are both generic preference-based measures of health-related quality of life, were completed by proxy and children over 11 yrs of age were invited to self-complete.
Both Health Utilities Index 2 and 3 demonstrated good practicality, with excellent completion rates (>97%) and a mean time to complete of around 8 mins. Both Health Utilities Index 2 and 3 demonstrated very good inter-rater reliability and evidence of sensitivity to change. At 6 months after admission, mean scores of the Health Utilities Index 2 and 3 were different in some groups of children with different degrees of in-hospital severity of illness, but those differences were not found at 12 months of follow-up.
The Health Utilities Index 2 and 3 both perform well in a pediatric intensive care setting whether by self-complete or proxy complete. Evidence of good inter-rater reliability gives confidence that the measures can be reliably used with a proxy completer, such as parent or caregiver. Additional research is important to investigate their construct validity further in this population, ideally using baseline data collected at the time of hospital stay in pediatric intensive care and other measures of health status at the times of follow-up.
From the Health Economics and Decision Science (KJS) and Health Services Research (JCF), ScHARR, The University of Sheffield, UK.
*See also p. 482.
Supported, in part, by the U.K. Medical Research Council, Development and assessment of risk adjustment methods for outcomes in paediatric intensive care in the United Kingdom, Grant Number: G9900013.
The authors have not disclosed any potential conflicts of interest.
For information regarding this article, K.Stevens@Sheffield.ac.uk