OHRQoL for 8-10 year olds: assessing a new measure

Objectives: To assess the reliability and construct validity of the Child Oral Health-Related Quality of Life for 8-10 year-olds (COHRQoL [8-10]) using confirmatory factor analysis (CFA) and to test the measurement properties of latent variables believed to define the multi-dimensional construct of OHRQoL. Methods: A convenience sample of 270, Year 4 children from 6 schools was obtained. The administered questionnaire included the 25-item COHRQoL [8-10] and the Coopersmith Self-Esteem Inventory-School Form. The analytical method was based upon CFA using maximum likelihood estimation. A second-order factoring approach was applied to determine the extent that the latent variables tapped a single over-arching domain of quality of life. Results: 7 items were withdrawn for low endorsement and poor association with resultant factors. The COHRQoL [8-10] was confirmed to measure a single construct of three latent variables invariant to gender. Internal consistency of the 3 scales derived comprising a total of 18 retained items was acceptable. Associations with self-esteem and with a single question on the extent that the mouth was a problem were confirmed and strengthened the construct validity of the COHRQoL [8-10] measure. Conclusions: Reliability and construct validity were demonstrated for COHRQoL [8-10]. CFA showed that the 3 constructs or latent variables underlying the overall COHRQoL ratings were discrete measures that can be reliably assessed in children. Further model testing with additional data will increase generalisation of these findings.