Multivariable caries prediction model vs Alternative prediction model(s) within the same study, or no formal comparator (models compared descriptively; no between-model statistical test)
Predictive performance for coronal caries increment (positive likelihood ratio, LR+) · Children and adolescents (aged 2-19) at risk of coronal caries increment (caries onset or progression)
Most externally validated coronal-caries models showed modest LR+ (<5); only Campus 2012 and Gao 2013 NUS-CRA reached the useful threshold; no pooled estimate; ungraded.
2 findings from 1 review
- Multivariable prediction models of caries increment: a systematic review and critical appraisal. 2023 All outcomesFollow-up >=1 year (included studies 1-5 years) · Oral health care; children/adolescents; no geographical restriction · Model validation studiesNo meta-analysis. Per-model LR+ (Table 2) across the 9 validation studies (31 coronal models) ranged roughly 0.78-5.53 (e.g. Beck 1.62-3.00, Christian/CAMBRA 1.09-1.14, Gao 2013 CAT ~1.03 to NUS-CRA 5.31, Campus 5.53,…9 studiesGRADE not reported
- Multivariable prediction models of caries increment: a systematic review and critical appraisal. 2023 All outcomesFollow-up >=1 year (included studies 1-5 years) · Oral health care; children/adolescents; no geographical restriction (11 countries across all studies) · Model development studiesNo meta-analysis (heterogeneity precluded pooling); performance given per model as LR+ recalculated by the review authors (Table 2). Across the 8 development studies (23 coronal models) LR+ spanned roughly 1.31-10.26…8 studiesGRADE not reported