Western and Modern Mexican dietary patterns are directly associated with incident hypertension in Mexican women: a prospective follow-up study

Adriana Monge, Martín Lajous, Eduardo Ortiz-Panozo, Beatriz L. Rodríguez, José Juan Góngora, Ruy López-Ridaura

Research has found that diet and dietary patterns are associated with blood pressure and hypertension. Limited research in this area has been conducted in a Mexican population. We investigated the relation between dietary patterns (principal component analysis) and the incidence of self-reported treated hypertension in 62,913 women from the Mexican Teachers' Cohort, a large population-based cohort of female Mexican teachers, who were free of hypertension at baseline in 2006-2008 when diet was assessed using a food frequency questionnaire. Dietary patterns were categorized into quartiles and logistic regression models were fit. Participants were 42.1 ± 7.3 years old, had a BMI 27.0 ± 4.4 and a cumulative incidence of hypertension of 4.6%. Between baseline and first follow-up (2011-2014) we identified 2916 incident cases of hypertension. We identified three major components. The first was loaded heavily with vegetables, fruits and legumes; the second component was loaded heavily with processed meats, fast foods, and red meat; and finally the third component was loaded heavily with corn tortillas, hot peppers, and sodas. We named the components as Fruits & Vegetables (FV), Western (W), and Modern Mexican (MM). The multivariable-adjusted odds of hypertension in the highest quartile of the W pattern were 24% higher than the odds for individuals in the lowest quartile (95%CI = 1.10, 1.40; P-trend = 0.0004); women in the highest quartile of the MM pattern had 15% higher odds than women in the lowest quartile (95%CI = 1.02, 1.29; P-trend = 0.01). The FV pattern was not significantly associated with hypertension (OR for extreme quartiles = 0.94; 95%CI = 0.84, 1.05; P-trend = 0.19). The Western pattern and the Modern Mexican pattern, which showcases an undergoing nutrition transition, may affect the incidence of hypertension, whereas the FV pattern was not associated with hypertension. These findings are important in the prevention of hypertension and cardiovascular diseases in Mexico and possibly among Mexican people living in the US.