Abstract
This retrospective analysis of longitudinal data was developed to determine which types, combinations, and intensities of topical fluorides more effectively prevent new caries-related restorations and extractions in high caries risk adults. We included data from October 1, 2008, through June 30, 2018, from electronic dental and medical records and pharmacy database from the US Department of Veterans Affairs. Veterans who were eligible for continuing and comprehensive care, met the criteria of high caries risk (received 2 or more caries-related restorations within a 365-d period), and had 3 y of follow-up were included. Multivariable logistic regression models estimated the odds of caries-related treatment during the 1-y observation period, controlling for age, gender, race and ethnicity, illness burden (Selim comorbidity index), use of prescription medications, attendance at dental prophylaxis appointments, number of caries-related restorations during the index year, and time between first and last caries-related restoration during the index year. The study sample included 68,757 veterans, who were primarily male (91.5%), were White (73.6%), had a mean age of 59.2 ± 13.5 y, and had significant medical comorbidity as measured by the Selim index (3.7 ± 2.4 physical and 1.3 ± 1.2 mental diagnoses). They had 10.8 ± 6.3 prescription VA drug classes, took 0.6 ± 0.8 strong anticholinergic medications, and had 3.9 ± 2.6 teeth restored due to caries during the index year. Adjusted multivariable logistic regression models showed veterans who received a varnish or gel/rinse fluoride intervention versus no fluoride had an approximately 29% decreased odds of receiving caries-related treatment during the observation period (gel/rinse adjusted odds ratio [AOR] = 0.72; 95% confidence interval [CI], 0.67-0.76; varnish AOR = 0.71; 95% CI, 0.67-0.75). The receipt of a varnish and gel/rinse did not demonstrate statistically better odds than each intervention alone (AOR = 0.69; 95% CI, 0.64-0.75). A dose-response effect was observed. Two-plus applications of varnish versus none (AOR = 0.73; 95% CI, 0.69-0.77) and 2-plus applications of gel/rinse versus none (AOR = 0.71; 95% CI, 0.67-0.75) were more effective than 1 application of either modality versus none.