OBJECTIVES: Poor diet due to poor oral health was proposed as a potential mediator of the association between poor oral health and frailty. This study explores the cross-sectional associations between poor oral health, poor diet, and frailty in Canadian adults, aged 45 to 85 years, and then tests whether the expected oral health-frailty association is changed by taking into account the pathway through poor diet. DESIGN: Cross-sectional study. SETTING: The baseline wave of the Canadian Longitudinal Study on Aging, a nationally representative population study. PARTICIPANTS: Data from 28 738 community-dwelling participants, aged 45 to 85 years. MEASUREMENTS: We characterized poor oral health (cumulative count of 24 items of oral health problems), poor diet (scale of poor food consumption from 7 healthy foods groups), and frailty (cumulative frailty index of 76 items). Regression-based path analyses were used to investigate associations between poor oral health, poor diet, and frailty, adjusted for age group, sex, income, smoking, living alone, education, physical activity, social support, and dental visit. RESULTS: Poorer oral health was associated with low income, smoking, low physical activity, low social support, and no dental visit. There were associations between poorer oral health and poorer diet (effect size β adjusted = .40; 95% confidence interval [CI] = .20-.61) and between poorer oral health and increased frailty (β adjusted = .85; 95% CI = .68-1.02). The indirect effect through the path of poor diet was approximately 0.01 (95% CI = 0.01-0.02) (ie, explaining 1% of the effect of poor oral health on frailty). CONCLUSIONS: Poor oral health was associated with poor diet and frailty. Each additional oral health problem was associated with an increase of approximately 1 frailty index point, even after full adjustment for poor diet. The indirect effect of poor oral health through poor diet was modest. Though poor oral health was associated with poorer diet quality, a more direct effect of poor oral health on increasing frailty may be indicated. J Am Geriatr Soc 68:959-966, 2020.