Reliability of clinician judgements of bruxism

Abstract

Bruxism is considered to be a parafunctional disorder requiring treatment and is viewed as a risk factor for the development of temporomandibular disorders (TMDs). The purpose of this investigation is to examine the reliability of clinician judgements of bruxism severity. Twenty dentists who are faculty members in a dental school examined 29 stone casts and gold-plated models of individual teeth for evidence of bruxism. Ordinal ratings of bruxism severity for the 29 augmented models were made on two occasions, approximately 3 months apart. Inter-rater reliability among all clinicians, evaluated using intraclass correlation coefficients (ICCs), was poor at both time one and time two (i.e. ICC = 0.33 and 0.32, respectively), with somewhat better reliability found among those clinicians with above-average time elapsed since completion of dental training (i.e. ICC = 0.48 and 0.50 for time 1 and time 2, respectively). Three-month test-retest reliabilities were fair (ICC = 0.46) for the full group of raters and were unrelated to clinicians' degree of confidence in their ratings. These results indicate a need to standardize methods for clinical assessment of bruxism. Additionally, they have implications for studies using clinical assessments of bruxism to test the association between bruxism and other conditions such as TMDs.