AimEmerging evidence points to a two‐way relationship between periodontitis and dietary choices and, thus, nutrition. This study aimed to assess the potential cause–effect relationship between the periodontitis stage, loss of functional tooth units (FTUs), masticatory function, and intake of different food groups using path analysis.Materials and MethodsA single calibrated examiner determined the periodontitis stage of a consecutive sample of 241 Chinese subjects reporting for tooth replacement. Their masticatory function was quantified by the mixing ability of a two‐colour chewing gum. Validated food frequency questionnaires were used to calculate the intake of 33 food group items by an experienced calibrated rater. After verification of assumptions, visual structural equation modeling was performed with Amos 23. The consistency of results and the potential modifying effect of age were assessed in 9043 subjects from the NHANES database.ResultsHighly significant models were constructed using periodontitis stage and age as exogenous factors. Periodontitis stage diagnosis significantly affected the number of posterior FTUs and oral health‐related quality of life (OHRQoL, path coefficient [PC] = –0.55 and −0.20, p < .05, respectively). In the model, FTUs also had an independent effect on OHRQoL (PC = 0.23, p < .05). FTUs determined the level of masticatory function (PC = –0.38, p < .05), which in turn affected vegetable intake but not fruit or meat intake (PC = –0.18, p < .0.5, PC = 0.06, NS and PC = 0.11, NS, respectively). The effect of age was significant for vegetable and meat intake and was also correlated with periodontitis stage diagnosis. Analysis of the NHANES database confirmed the negative impact of periodontitis on the number of occluding pairs and vegetable consumption for the 18–44, 45–60 and >60 age groups.ConclusionsPeriodontitis showed a potential cause–effect pathway affecting vegetable intake across cultures and age groups. The size of the effect is potentially of clinical and public health significance. Additional studies, including intervention trials, are required to test this potential mechanism linking oral health to nutrition.