Reliability, Construct Validity, Acceptability and Feasibility of the BruxScreen

The recently developed BruxScreen consists of two parts: BruxScreen-Q (self-report questionnaire) and BruxScreen-C (clinical assessment).

To test the intra- and inter-rater reliability, construct validity, acceptability and feasibility of the BruxScreen-Q and BruxScreen-C and assess their concordance among Dutch dental students.

88 out of 109 potentially eligible dental master students completed a set of questionnaires two times (Q1; Q2) and participated in two clinical workshops (CE1; CE2), using the BruxScreen-Q and BruxScreen-C, respectively. Intra-rater reliability of the BruxScreen-Q and concordance between the BruxScreen-Q and BruxScreen-C were assessed using Cohen's (weighted) Kappa. Intra- and inter-rater reliability of the BruxScreen-C were analysed using intraclass correlation coefficients calculated from generalised linear mixed-effects models. Construct validity of the BruxScreen-Q was tested using Spearman's Rank Correlation or Mann-Whitney U test based on hypothesis testing. Acceptability and feasibility of the BruxScreen were assessed using descriptive statistics.

Intra-rater reliability for BruxScreen-Q was fair to substantial. Intra- and inter-rater reliability for BruxScreen-C varied from poor to excellent. BruxScreen-Q showed moderate construct validity, based on the acceptable consistency between the actual and hypothesised effect size of the questionnaire items. BruxScreen-Q (Q2) and BruxScreen-C (CE2) were found both acceptable and feasible by a majority of the students. There was no agreement between subject-based bruxism according to the BruxScreen-Q and clinically based bruxism according to the BruxScreen-C.

The BruxScreen demonstrates acceptable reliability, construct validity, acceptability and feasibility in assessing both subject-based bruxism and clinically based bruxism. However, there is a discrepancy between self-reported bruxism and the clinicians' diagnosis.