Using factor analysis to confirm the validity of children's self-reported health-related quality of life across different modes of administration

Abstract

Background The measurement of pediatric health-related quality of life (HRQOL) in pediatric medicine and health services research has grown significantly over the past decade. While HRQOL instruments are often administered in clinic settings, alternative modes of administration are also utilized such as mail and telephone surveys. In order for comparisons across in-person, mail, and telephone administration to be valid, items on a HRQOL measure must have equivalent meaning across mode of administration. That is, they must demonstrate factorial invariance (stable factor structure) across mode of administration.

Purpose The objective of the present study was to implement the multigroup confirmatory factor analysis (CFA) method for invariance testing across mode of administration for children's self-reported HRQOL.

Methods Multigroup CFA was performed specifying a five-factor model across three modes of administration groups (in-person, mail, and telephone survey). The 23-item PedsQL^TM 4.0 Generic Core Scales was utilized as the measure of children's self-reported HRQOL in a sample of 3741 children ages 5—18.

Results Strong factorial invariance across the mode of administration groups was demonstrated based on stability of the Comparative Fit Index (CFI) between the models, and several additional indices of practical fit including the Root Mean Squared Error of Approximation (RMSEA), the Non-Normed Fit Index (NNFI), and the Parsimony Normed Fit Index (PNFI).

Limitations The mode of administration subpopulations in our study differed with regard to health status.

Conclusion The multigroup CFA statistical methods utilized in the present study have important implications for clinical trials in which mixed modes of administration are used. The present findings support an equivalent five-factor structure of the PedsQL ^TM 4.0 Generic Core Scales across the three modes of administration studied. Based on these data, it can be concluded that children across the three modes of administration groups interpreted items on the PedsQL ^TM 4.0 Generic Core Scales in a similar manner. Clinical Trials 2009; 6: 185—195. http://ctj.sagepub.com