Performance of Dental Undergraduate Students’ Using CAST and ICDAS II

Objectives: To compare caries detection, treatment decisions, level of difficulty and confidence between two groups of dental undergraduate students using CAST and ICDAS II
Methods: Twenty-nine students from the 7th and 8th semesters were randomly allocated to two groups: CAST (n=15) and ICDAS II (n=14). Each group was trained on the caries detection method and immediately after, assessed 37 clinical pictures. Scores were compared to a gold-standard established by 3 experts. Data were analysed using t-Test, Pearson correlation, Wilcoxon and MANOVA tests
Results: The correlation of CAST scores and the gold-standard were r=0.847 and r=0.800 (p<0.01) for the 7th semester and 8th semester, respectively. Regarding ICDAS II, the correlation coefficient between the first and the second digit with the gold-standard were r=0.568 (p<0.01) and r=0.721(p<0.01), respectively and no difference between students from the 7th and 8th semesters was observed. The inter-examiner agreement was κ=0.70 for CAST and κ=0.55 for ICDAS II. The level of perceived difficulty between students form the 7th and 8th semesters was similar (p=0.778) and between CAST and ICDAS II (p=0.975). A difference for treatment decisions was observed when ICDAS II was compared to the gold-standard (p<0.0001). The treatment decisions were less invasive when CAST was used in comparison to ICDAS II (p=0.004)
Conclusions: CAST had a higher correlation with the gold-standard than ICDAS II for both caries detection and treatment decisions and achieved higher inter-examiner agreement. The treatment decisions associated with ICDAS II were more invasive. Although the level of difficulty and confidence for both systems were similar, the ICDAS II two-digit system created difficulties in analyzing the data and required more complex statistical analysis