Measuring health‑ and oral health-related quality of life in secondary school pupils: a head‑to‑head psychometric comparison of CHU9D and CARIES-QC-U

Objectives

Dental caries impacts children’s health- and oral health-related quality of life. Preference-based measures (PBM) can quantify these impacts as utilities, facilitating economic evaluation of interventions. Two paediatric PBMs (one generic (CHU9D) and one condition-specific (CARIES-QC-U)) were used in the BRIGHT randomised control trial investigating the impact of a behaviour change intervention on schoolchildren’s oral health. No comparison has been made of these two instruments previously. This study aimed to compare the psychometric properties of CHU9D and CARIES-QC-U using trial data.

Methods

Baseline trial data were assessed. Mean utility scores, missing values and floor and ceiling effects were determined for each instrument. Cronbach’s alpha was assessed to indicate internal consistency for each instrument. Correlations were explored between CARIES-QC-U and CHU9D, the dimensions within the two instruments, and between each instrument and DMFT. Effect sizes (Cohen’s d) were explored for each component of DMFT in relation to overall utility values from each instrument.

Results

Baseline data from 4542 schoolchildren aged 11–13 years were analysed. Over a third of participants had obvious caries experience. Mean utility scores for CARIES-QC-U and CHU9D were 0.76 and 0.91 respectively. Missing data was low for both instruments. Floor and ceiling effects were greater for CARIES-QC-U. Internal consistency was acceptable for both instruments. Correlation between utilities of CARIES-QC-U and CHU9D was weak at 0.35. Correlation between clinical caries experience and utilities from CARIES-QC-U was negative (r=-0.09) and stronger than with CHU9D (r=-0.02). Correlations between dimensions within the instruments were weaker than anticipated. Small, statistically significant effects were seen for both instruments and the decayed (D) component of DMFT, though this was stronger with CARIES-QC-U.

Conclusions

The burden of caries was reflected in participant utility scores. Whilst both PBMs performed well psychometrically, CARIES-QC-U demonstrated greater ability to capture impacts related to dental caries, indicating better suitability for caries research than CHU9D.