Is the Modified Dental Anxiety Scale (MDAS) a Single or Two Construct Measure? A Theoretical and Pragmatic Perspective.

Background: The MDAS questionnaire is one of a number of scales available to assess dental anxiety. It is widely used and translated into many world languages; however, it lacks an explicit theoretical backdrop to the content and structure of the measure. This paper draws upon original expositions of dental anxiety: how it develops, is maintained, and how this draws attention to a re-evaluation of the measure. To assist this inspection it was proposed to investigate a two latent construct formulation through a stepwise analysis using data from a representative survey of English respondents on their oral health (the Adult Dental Health Survey). Aim: To present a brief theoretical framework to underpin the measure and, as part of this study's objectives, to provide some evidence to support the measure's potential two-construct structure. Method: Narrative review, structural equation modelling, and testing of specific associations to indicate a two latent construct formulation. Data included the MDAS items (where items 1 and 2 comprise the anticipatory subscale, and items 3 to 5 describe the treatment-related subscale). These items were completed by the representative sample of respondents from the most recent Adult Dental Health Survey conducted in 2009. Results: The two latent construct solution for describing dental anxiety was supported. The anticipated and treatment-related subscales could be discriminated, although they were strongly correlated, demonstrating overlap. Comparison of how each construct varied across the three age groups suggests an interesting heterogeneity. In addition, the two constructs behaved differently when acknowledging previous experience of respondents' last dental visit. Anticipatory dental anxiety was more strongly related to oral health-related quality of life (OHRQoL) than treatment-related dental anxiety as predicted. This partial evidence from empirical data and previous reports in other studies suggests that the separation of the MDAS measure into the two subscales may be warranted. Discussion: Researchers are recommended to report not only the total score of the MDAS in their studies but also consider presenting the two subscale scores, namely, anticipatory and treatment-related dental anxiety. Further work is indicated to determine if clinicians may find the subscales of use when assessing their patients.