Background: Few studies have evaluated the quality of plant-based diets in relation to chronic diseases, and the potential role of BMI is not clearly explored. Objectives: To study the associations between plant-based diet indices and type 2 diabetes (T2D) and hypertension risks, as well as the extent to which the associations were modified and/or mediated by BMI. Methods: The study included 74,522 women from the Etude Epidemiologique aupres de femmes de la Mutuelle Generale de ('Education Nationale prospective cohort (mean (SD): age, 52.94 (6.7) years; BMI, 22.970 (3.328) kg/m(2)]. Dietary data were collected at baseline (1993) via an FFQ. Overall plant-based diet index (PDI), healthful PDI (hPDI), and unhealthful PDI (uPDI) were developed. Multivariable Cox regression models were used to derive HRs and 95% Cls. Effect modification and mediation by BMI were explored. Results: There were 3292 (4.64%) incident cases of T2D and 12,504 (2714%) incident cases of hypertension over similar to 20 years of follow-up. In the multivariable model further adjusted for BMI, higher adherence to PDI and hPDI was associated with lower T2D and hypertension risks, with an HR per 1-SD increase (95% CI) of 0.88 (0.85, 0.91) and 0.96 (0.94, 0.98) for PDI and 0.88 (0.85, 0.92) and 0.94 (0.92, 0.95) for hPDI, respectively. uPDI was not associated with T2D 10.98 (0.94, 1.01)1, whereas a positive association was observed with hypertension: 1.04 (1.02,1.06). There was interaction between PDI and uPDI, as well as BMI, on T2D (P-interaction < 0.001) but not on hypertension (P-interaction > 0.05). In addition, BMI mediated 26-59% and 0.2-59% of diet-T2D and diet-hypertension associations, respectively. Conclusions: Differential associations between plant-based diets and T2D and hypertension risks were observed among women in this large prospective study. Only healthier plant foods were associated with lower risks, partly through decreasing BMI.