Weight Loss Mediated the Relationship between Tooth Loss and Mortality Risk

Tooth loss is a risk factor for increased mortality; however, the underlying mechanism remains unclear. This study aimed to evaluate the mediating effect of weight change on the relationship between tooth loss and mortality risk. This was a 10-y follow-up prospective cohort study using the data from the Japan Gerontological Evaluation Study (JAGES). The participants were independent older adults aged >= 65 y at baseline and were followed up from 2010 to 2020. The incidence of death in 2013 and 2020, incidence of >5% weight loss/gain in 2010 and 2013, and the number of remaining teeth in 2010 were used as the outcome, mediator, and explanatory variables, respectively. We conducted causal mediation analysis by fitting the Cox proportional hazard model, including possible confounders. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) of the total effect (TE), natural indirect effect (NIE), and proportion mediated (PM) were estimated. Among the 34,510 participants, the mean age was 72.6 (SD = 5.4) y, and 47.6% were men. From 2013 to 2020, 14.0% of the participants (n = 4,825) died, 60.5% (n = 20,871) had 0 to 19 remaining teeth, and 17.2% (n = 5,927) and 8.4% (n = 2,907) experienced >5% weight loss and gain, respectively. The mortality rate was 0.016 per person-year among those with >= 20 remaining teeth and 0.027 per person-year among those with 0 to 19 remaining teeth. Weight loss of >5% significantly mediated the association between tooth loss and higher mortality risk (TE: HR, 1.28 [95% CI, 1.16 to 1.40]; NIE: HR, 1.03 [95% CI, 1.02 to 1.04]; PM, 13.1%); however, we observed a slight mediating effect for >5% weight gain (NIE: HR, 1.003 [95%CI, 1.0001 to 1.01]; PM, 1.3%). The present study suggests that a clinically meaningful level of weight loss mediated the association between tooth loss and increased risk of mortality among independent older adults.