Abstract
The longevity of resin-based composite (RBC) restorations is influenced by many factors, and it is difficult to assess all of them in small data sets. This study uses a large data set from a university and its dental clinics to assess many potentially predictive factors of restoration survival.
This retrospective cohort study used de-identified data from electronic dental records of patients attending records the University of Iowa and its dental clinics. Multivariable Cox proportional hazards models were generated to determine which variables influenced the longevity of the restorations, and model selection via best subset selection on Bayesian Information Criterion (BIC) was used to select the best model.
The number of surfaces covered by the restoration showed a very strong influence, with more surfaces leading to reduced longevity (HR 2.49 (CI 1.97–3.15), p < 0.001 for 4 surfaces), but the inclusion of a proximal surface did not have a significant influence. The presence of an eating disorder (HR 1.72 (CI 1.15–2.56), p = 0.008), dry mouth (HR 1.20 (1.02–1.43), p = 0.033), or dental fear (HR 1.39 (CI 1.14–1.69), p = 0.001) also had a negative effect on restoration longevity, but bruxism/clenching did not. The type of filler used in the RBC, conventional or supra-nano spherical, also affected restoration longevity, with greater longevity for supra-nano spherical filler (HR 0.61 (CI 0.45–0.82), p = 0.001).
Type of provider showed no impact on restoration longevity in this study, and only a very limited number of patient factors were significant. Differences in the type of filler and the number of surfaces covered by the restoration did have a significant effect on restoration longevity.
•This retrospective cohort analysis investigated the longevity of direct resin-based composite.•This study used de-identified data from electronic dental records.•Multivariable Cox proportional hazards models were generated to determine which variables.•Type of provider showed no impact on the RBC.•Differences in the material type and the number of surfaces covered longevity.