Causal effect of tooth loss on depression: evidence from a population-wide natural experiment in the USA

Abstract

Abstract Aims Depression severely affects people's health and well-being. Oral diseases have been suggested to be associated with depression, but so far, there is no causal evidence. This study aimed to identify the causal effect of tooth loss on depression among US adults in a natural experiment study. Methods Instrumental variable analysis was conducted using data from 169 061 respondents born in 1940–1978 who participated in the 2006, 2008 or 2010 waves of the Behavioral Risk Factor Surveillance System (BRFSS). Random variation in tooth loss due to differential childhood exposure to drinking water fluoride was exploited as an instrument. Results US adults who were exposed to drinking water fluoride in childhood had more remaining teeth, therefore providing a robust instrument (F = 73.4). For each additional tooth loss, depressive symptoms according to the eight-item Patient Health Questionnaire depression (PHQ-8) score increased by 0.146 (95% CI 0.008–0.284), and the probability of having clinical depression (PHQ ⩾10) increased by 0.81 percentage points (95% CI −0.12 to 1.73). Conclusions Tooth loss causally increased depression among US adults. Losing ten or more teeth had an impact comparable to adults with major depressive disorder not receiving antidepressant drugs.