A methodological study to assess the measurement properties (reliability and validity) of a caries risk assessment tool for young children

Bradley Christian; Hanny Calache; Geoff Adams; Martin Hall; Stuart Dashper; Lisa Gibbs; Mark Gussy

doi:10.1016/j.jdent.2020.103324

A methodological study to assess the measurement properties (reliability and validity) of a caries risk assessment tool for young children

J Dent. 2020 Apr:95:103324. doi: 10.1016/j.jdent.2020.103324. Epub 2020 Mar 19.

Authors

Bradley Christian¹, Hanny Calache², Geoff Adams³, Martin Hall⁴, Stuart Dashper³, Lisa Gibbs⁵, Mark Gussy⁶

Affiliations

¹ Dentistry and Oral Health, La Trobe Rural Health School, La Trobe University, Bendigo, Victoria, Australia; Jack Brockhoff Child Health and Wellbeing Program, Centre for Health Equity, Melbourne School of Population and Global Health, The University of Melbourne, Melbourne, Australia. Electronic address: b.christian@latrobe.edu.au.
² Centre for Population Health Research, Deakin University, Burwood, Victoria, Australia.
³ Melbourne Dental School, The University of Melbourne, Melbourne, Australia.
⁴ Dentistry and Oral Health, La Trobe Rural Health School, La Trobe University, Bendigo, Victoria, Australia; Dental Health Services Victoria, Melbourne, Australia.
⁵ Jack Brockhoff Child Health and Wellbeing Program, Centre for Health Equity, Melbourne School of Population and Global Health, The University of Melbourne, Melbourne, Australia.
⁶ Lincoln International Institute for Rural Health, College of Social Science, University of Lincoln, Lincoln, United Kingdom.

PMID: 32200008
DOI: 10.1016/j.jdent.2020.103324

Abstract

Objectives: To assess the CAMBRA caries risk assessment tool (CRAT) in terms of its measurement properties: reliability, validity and responsiveness to change in caries estimates over time.

Methods: Secondary data from the VicGeneration (VicGen) birth cohort study were used. Caries risk status at child age18 months was compared against caries development at 36 and 48 months. Questionnaire data for the 18-month-old children, were used to generate the CRAT item responses. Five examiners assessed the caries risk profile for each 18-month-old child and assigned a risk status. The analytic method was informed by the COnsensus-based Standards for the selection of health Measurement INstruments Checklist.

Results: Each examiner applied the CAMBRA CRAT to 214 cases from the VicGen study. At 18 months of age, the proportion of low-, moderate- and high-risk children were 17 %, 17 % and 66 %, respectively. Caries prevalence at 36 and 48 months of age was 18 % and 35 %, respectively. Almost perfect reliability (both inter- and intra-reliability) was achieved in caries risk categorisation. For criterion validity and responsiveness, the area under the curve was approximately 0.50. For the dichotomised risk, the sensitivity of the CAMBRA CRAT to predict caries at 36 and 48 months, was 74 % and 70 %, respectively. The corresponding specificity scores were 35 % and 36 %, respectively.

Conclusions: The results from this study show that the CAMBRA CRAT, in its current form, is not a good predictor of future caries among young Victorian children and in fact overestimates the proportion of children at high-risk of future disease. As such, it may not be ideal for use among young Victorian children and it is highly likely it's use could be driving over treatment.

Clinical significance: This study provides evidence for the applicability of the CAMBRA CRAT with young Victorian children to inform decision making on caries risk-based management. It also highlights the important measurement properties and related analytics (applicable to any CRAT) that should be considered when selecting a CRAT for use in clinical practice.

Keywords: Birth cohort; Children; Dental caries; Risk assessment; Risk factors.

MeSH terms

Child
Child, Preschool
Cohort Studies
Dental Caries Susceptibility
Dental Caries*
Humans
Infant
Reproducibility of Results
Risk Assessment