Related Experiment Video
Updated: Aug 14, 2026
Shrinkage of Dental Composite in Simulated Cavity Measured with Digital Image Correlation
Published on: July 21, 2014
Assessing sealant retention using a Poisson frailty model
M V Morgan1, G G Adams, A C Campain
1The School of Dental Science, The University of Melbourne, Australia. m.morgan@unimelb.edu.au
Objective:
This paper utilises a Poisson frailty regression model with clustering to report on the failure rates of sealants placed and repaired during a three-year follow-up school-based comprehensive preventive programme.
Basic Research Design:
The programme included annual application/reapplication of light-cured fissure sealants in 210 12-year-old subjects over three years. Sealants placed at the commencement of each year of the programme were assessed for retention 12 months later and repair/reapplication was undertaken at that time for those surfaces judged to require it. An application of a Poisson regression model with random effects ("frailties") and clustering was used in this paper to allow for heterogeneity between subjects. Whilst it appears that the direct utilization of frailty models in determining fissure sealant failure rates has not been reported, it is clearly an appropriate use of this statistical method. A total of 1,544 sealants were placed during the study period. Of these, 1,038 (67.2%) were first placements at baseline and 506 (32.8%) were repairs/reapplications. Of these repairs/reapplications, 206 (40.7%) were performed at first year follow-up and 300 (59.3%) at second year follow-up.
Results:
The Failure Rate Ratios were statistically significantly greater for second molars cf. first molars (FRR=1.33, p=0.008) and for lower molars cf upper molars (FRR=1.32, p<0.001). Failure rate ratios for sealants placed in the distal occlusal pit and fissures cf. mesial occlusal pit and fissures were significantly higher (FRR=1.33, p<0.001). A significant interaction was found for molar type versus pit and fissure site (p=0.009). There was no significant interaction between molar type and arch.
Conclusions:
Sealants placed on second molars, those placed on lower molars and those placed on the distal fissure sites have a higher failure rate. As individual tooth sites in the mouth are not independent, any statistical analysis should allow for these associations when assessing sealant retention.