Comparative evaluation and agreement between ICDAS and UniViSS caries detection systems among toddlers and preschoolers.

OBJECTIVES: To compare the agreement and concordance between ICDAS and UniViSS caries detection systems among toddlers and preschoolers. METHODS: This cross-sectional study included 281 children aged 2-6 years at a tertiary care hospital in South India. All primary molars were examined using both systems and converted to binary format (caries present vs. sound) for comparison, as full ordinal-scale comparison was not feasible given their non-equivalent scoring dimensions. Agreement was assessed using Cohen's Kappa coefficient; sensitivity, specificity, PPV, NPV, and likelihood ratios were calculated with ICDAS as the reference comparator. As ICDAS is a visual index rather than a histological gold standard, all parameters reflect inter-method concordance, not true diagnostic accuracy. RESULTS: The mean age was 5.01 years (SD = 1.07), with 52.7% females. Cohen's Kappa values ranged from 0.886 to 0.983, indicating almost perfect agreement. Against ICDAS as comparator, UniViSS showed inter-method sensitivity of 0.88-1.00, specificity of 0.96-0.99, PPV of 0.88-0.97, and NPV of 0.94-1.00; these values represent agreement with ICDAS and not true diagnostic accuracy. Overall concordance ranged from 95% to 99%. CONCLUSION: UniViSS demonstrated almost perfect inter-method concordance with ICDAS for binary caries detection in primary molars of toddlers and preschoolers. UniViSS may serve as a practical clinical alternative to ICDAS for binary caries screening; however, findings should be interpreted in the context of the high-prevalence study population, and histological validation remains a priority.