Reliability and validity of artificial intelligence-based innovative digital scale for the assessment of anxiety in children

Background Dental anxiety still persists as a potential problem in managing the child in the dental office. There is a need to develop a gold standard scale to measure anxiety in children incorporating newer technology. An innovative self-reported scale known as RMSDigital Anxiety Scale (RMS-DAS) incorporating artificial intelligence (AI) was developed. Aim To assess the reliability and validity of an AI-based, innovative digital scale for the assessment of dental anxiety in children. Methods Seventy-six children (aged 4-12 years) were included in the reliability group. The RMS-DAS test score was recorded on Day 1 where the child was asked to click on the expression produced by AI that matches his/her anxiety level the most at that moment. RMS-DAS retest score was recorded after 7 days. The validity group included 140 children. The anxiety scores were recorded using three scales; RMS-DAS, RMS-Pictorial Scale (RMS-PS) and Facial Image Scale (FIS) during the same visit where the child was asked to click on the expression that matches his/her anxiety level the most at that moment. Reliability was assessed by the internal consistency using Cronbach’s alpha and the test-retest was assessed using paired t-test, scatterplot, and coefficient correlation. The validity of RMS-DAS was assessed by correlating it with RMS-PS and FIS using Spearman’s correlation coefficient. Results The internal consistency of RMS-DAS was highly reliable (α=0.810). A strong correlation existed between the test-retest scores (r=0.7). There was a strong correlation when RMS-DAS was compared with RMS-PS (r=0.73) and FIS (r=0.76). Conclusions RMS-DAS is a reliable and valid scale that can be used as a new digital tool to assess children’s dental anxiety.