Assessing the effectivity of counting the number of teeth with their conditions to predict mortality: the OHSAKA study

Although previous studies have reported associations between the number of teeth and all-cause mortality, the results vary depending on the tooth condition. Few previous studies have focused on the most effective method of counting the number of teeth to predict all-cause mortality. This study aimed to identify an effective method for counting the number of teeth according to their condition in order to predict all-cause mortality. This cohort study used data from the Oral Health Screening to Assess Keys of Aging Well (OHSAKA) study and linked public dental check-ups and healthcare administrative datasets in Japan from 2018 to 2020. A total of 190,282 participants aged 75 years or older who underwent public dental checkups in Japan were evaluated for the study. The exposure in this study was the number of teeth and their condition, excluding the third molars. The outcome measure was all-cause mortality using the National Health Insurance Database of Japan. The median observational period was 3.4 years (interquartile range, 2.8–3.8), and all-cause mortality was observed in 9,713 (12.0%) men and 6,242 (5.7%) women. Adjusted Cox proportional hazard models revealed a dose-dependent association between sound and filled teeth and the all-cause mortality (adjusted hazard ratios [95% confidence intervals] of 0, 1–5, 6–10, 11–15, 16–20, and ≥ 21 teeth for men: 1.74 [1.62, 1.87], 1.51 [1.41, 1.61], 1.42 [1.33, 1.51], 1.38 [1.29, 1.47], 1.15 [1.08, 1.22], and 1.00 [reference], respectively, for women: 1.69 [1.55, 1.85], 1.35 [1.24, 1.48], 1.24 [1.14, 1.34], 1.21 [1.12, 1.31], 1.17 [1.09, 1.26], and 1.00 [reference], respectively). The Net Reclassification Improvement of the sound + filled + decayed teeth count model decreased compared to that of the sound + filled teeth count model in both men and women. This study clarified the deleterious and beneficial associations between decayed and filled teeth and all-cause mortality. The total number of sound and filled teeth predicted all-cause mortality more accurately than the number of sound teeth alone, or the number of sound, filled, and decayed teeth combined.