Low education and incident complete tooth loss: cross-national associations and cohort-specific mediation pathways in HRS and CLHLS

Low educational attainment reflects accumulated socioeconomic disadvantage and may contribute to severe oral health inequalities across the life course. Whether its association with incident complete tooth loss is consistent across populations with different social and behavioral contexts remains unclear. This study examined the association between low education and incident complete tooth loss in US and Chinese cohorts and explored potential cohort-specific mediating pathways. We analyzed 4,728 participants from the Health and Retirement Study (HRS) and 2,091 from the Chinese Longitudinal Healthy Longevity Survey (CLHLS). In HRS, low education was defined as educational attainment below upper secondary school; participants without complete tooth loss at wave 8 (2006) were followed to wave 9 (2008). In CLHLS, low education was defined as no formal schooling, and participants with natural teeth in 2014 were followed through 2018. Logistic regression models estimated odds ratios (ORs) and 95% confidence intervals (CIs), with adjustment for demographic, behavioral, and clinical covariates. Mediation analyses were performed using the SimpleMediation package in R, focusing on drinking in HRS and absence of social activity in CLHLS. Incident complete tooth loss occurred in 360 h participants (7.61%) and 319 CLHLS participants (15.26%). After multivariable adjustment, low education remained associated with higher odds of complete tooth loss in HRS (OR = 2.26, 95% CI 1.73–2.94; P < 0.001) and CLHLS (OR = 1.35, 95% CI 1.02–1.80; P = 0.037). The mediating pathways differed across cohorts. In HRS, drinking showed an indirect effect of 0.1355 (95% CI 0.0731–0.1979), accounting for 14.1% of the association. In CLHLS, absence of social activity showed an indirect effect of 0.0471 (95% CI 0.0003–0.0939), accounting for 14.7%. Low education was consistently associated with incident complete tooth loss in two socially and culturally distinct cohorts, although the strength of association differed between populations. Approximately one-seventh of the observed association was mediated through cohort-specific behavioral or social pathways, highlighting the need to consider both socioeconomic disadvantage and its context-dependent mechanisms when addressing inequalities in severe tooth loss.