Abstract
Background: Vegetarian diets have been associated with reduced mortality. Because a pure vegetarian diet might not easily be embraced by many individuals, consuming preferentially plant-derived foods would be a more easily understood message. A provegetarian food pattern (FP) emphasizing preference for plant-derived foods might reduce all-cause mortality. Objective: The objective was to identify the association between an a priori-defined provegetarian FP and all-cause mortality. Design: We followed 7216 participants (57% women; mean age: 67 y) at high cardiovascular risk for a median of 4.8 y. A validated 137-item semiquantitative food-frequency questionnaire was administered at baseline and yearly thereafter. Fruit, vegetables, nuts, cereals, legumes, olive oil, and potatoes were positively weighted. Added animal fats, eggs, fish, dairy products, and meats or meat products were negatively weighted. Energy-adjusted quintiles were used to assign points to build the provegetarian FP (range: 12-60 points). Deaths were confirmed by review of medical records and the National Death Index. Results: There were 323 deaths during the follow-up period (76 from cardiovascular causes, 130 from cancer, 117 for noncancer, noncardiovascular causes). Higher baseline conformity with the provegetarian FP was associated with lower mortality (multivariable-adjusted HR for ≥40 compared with <30 points: 0.59; 95% CI: 0.40, 0.88). Similar results were found with the use of updated information on diet (RR: 0.59; 95% CI: 0.39, 0.89). Conclusions: Among omnivorous subjects at high cardiovascular risk, better conformity with an FP that emphasized plant-derived foods was associated with a reduced risk of all-cause mortality. This trial was registered at www.controlled-trials.com as ISRCTN35739639.
Generated Summary
The Prevención con Dieta Mediterránea (PREDIMED) study, a multicenter, randomized clinical trial, investigated the effects of a provegetarian food pattern (FP) on all-cause mortality among 7,216 participants at high cardiovascular risk. The study defined a provegetarian FP emphasizing plant-derived foods (fruit, vegetables, nuts, cereals, legumes, olive oil, and potatoes) and de-emphasizing animal-derived foods (animal fats, eggs, fish, dairy products, and meats). Participants, aged 55-80 years, were followed for a median of 4.8 years, with dietary information collected through a validated 137-item food-frequency questionnaire. The research aimed to determine if better adherence to this provegetarian FP, as defined a priori, was associated with reduced all-cause mortality. The study employed a Cox proportional hazards model to assess the association between the provegetarian FP and total mortality, adjusting for multiple covariates and center stratification, and utilized yearly updated dietary information.
Key Findings & Statistics
- During the follow-up period, there were 323 deaths (76 from cardiovascular causes, 130 from cancer, and 117 from other causes).
- Higher baseline conformity with the provegetarian FP was associated with lower mortality. The multivariable-adjusted hazard ratio (HR) for those with a provegetarian FP score ≥40 compared with <30 points was 0.59 (95% CI: 0.40, 0.88).
- Similar results were found with the use of updated information on diet (RR: 0.59; 95% CI: 0.39, 0.89).
- The crude absolute rates of death decreased monotonically across successive quintiles of the provegetarian FP.
- The death rates decreased from 14.9/1000 person-years among participants with low adherence (score 40).
- The significance of the inverse linear trend in the crude Cox model stratified by center (P = 0.014) strengthened after sex and age were adjusted for (P = 0.003).
- The fully adjusted HRs (95% CIs) for categories of low, moderate, or high/very high adherence compared with the very low (reference) adherence category were 0.71 (0.50, 1.02), 0.68 (0.48, 0.96), and 0.59 (0.40, 0.88), respectively, with a significant inverse linear trend (P = 0.027).
- When the high and very high adherence categories were separated, only 9 deaths were observed in the latter category (>44 points). For this group, the HRs (95% CIs) were as follows: crude, 0.46 (0.22, 0.95); age- and sex-adjusted, 0.38 (0.18, 0.79); and multivariable-adjusted, 0.44 (0.21, 0.92).
- The correlation coefficient between the two scores assessing the provegetarian FP was 0.66 (95% CI: 0.64, 0.67).
- The cross-classification of both scores in 3 categories classified 57.5% of subjects in the same category, and only 4.8% were misclassified in extremely opposed categories.
- The cross-classification in approximate quintiles of both scores placed 79.7% of subjects in the same or adjacent quintile, and only 4.2% were misclassified >2 quintiles apart.
- When we positively weighed eggs and dairy products and added them to the positive part of the score together with plant-based foods, the age- and sex-adjusted HRs of death from any cause (95% CIs) were 1 (reference), 0.83 (0.56, 1.23), 0.75 (0.51, 1.09), and 0.52 (0.33, 0.80) for <30, 30–34, 35–39, and ≥40 points, respectively.
- When we restricted the analysis to cardiovascular deaths, the multivariable-adjusted HRs (95% CIs) for 30-34, 35-39 and ≥40 points of adherence to the provegetarian FP were 0.48 (0.24, 0.99), 0.44 (0.22, 0.90), and 0.47 (0.21, 1.04), respectively, and showed strong inverse associations (P-trend = 0.039).
- In a sensitivity analysis using absolute cutoffs for food groups, the fully adjusted HRs were 0.85 (95% CI: 0.65, 1.11) for moderate adherence and 0.70 (95% CI: 0.51, 0.95) for high adherence, with a significant inverse linear trend (P = 0.003).
Other Important Findings
- Older participants, those with lower educational levels, high blood cholesterol at baseline, and those who were more physically active showed better adherence to the provegetarian FP.
- Participants with diabetes, current smokers, and those with a higher BMI showed lesser adherence to this FP.
- The consumption of the 7 plant-derived food groups increased monotonically across increasing categories of the provegetarian FP, whereas the consumption of the 5 food groups from animal sources monotonically decreased.
- The most striking differences were observed for the consumption of nuts.
- The intakes of total energy, carbohydrate, polyunsaturated fat, and vitamin C, folate, and fiber were greater in the groups with better adherence to the provegetarian FP.
- The intakes of protein, total fat, saturated fat, and heme iron were inversely associated with adherence to this FP.
Limitations Noted in the Document
- The study population was elderly and at high risk of cardiovascular disease, limiting generalizability to younger or healthier individuals.
- The provegetarian diet score may have lacked accuracy due to measurement errors in dietary assessment.
- Not every plant food is necessarily healthy (refined cereals and potatoes), and the study population consumed primarily refined white bread.
- The study’s assessment of fish consumption was a negative component in the score, but the literature does not consistently support the association between fish consumption and all-cause mortality.
Conclusion
The study provides evidence to support the concept that a diet emphasizing plant-derived foods, with reductions in animal-derived foods, is linked to a reduced risk of death among older individuals at high cardiovascular risk. The findings support the idea that dietary recommendations should focus on the consumption of foods, rather than on single nutrients, to improve health outcomes. The simplicity and translational potential of this approach could encourage a positive impact on public health, encouraging a more realistic and manageable approach to healthier eating habits.
IFFS Team Summary
- 7000 subject, mean age 67, all at high risk for cardiovascular disease, followed for 5 years
- scored on how much their diet was plant based (even if some foods were chips and french fries), and deducted points for all animal foods
- those with higher plant based consumption had 40% less mortality, even though they were not fully vegetarian
- omnivorous food consumption with low meat and high plant foods decreases mortality in high risk people
- 23,000 German subjects, aged 35 to 65 years from the European Prospective Investigation Into Cancer and Nutrition-Potsdam study – mean 7.8 year follow-up
- 4 major health factors identified: never smoking, BMI < 30, 3.5 h/wk or more of physical activity, good diet (high in fruits, vegetables, and whole-grain bread and low meat consumption)
- all factors equally weighted
- adjusted for age, sex, educational status, and occupational status
- Those with all 4 factors at baseline had a 78% less chance of acquiring chronic disease defined as Diabetes (93% less chance), Myocardial Infarction (81% less chance), Stroke (50% less chance), Cancer (36% less chance)
- As compared to subjects without any of the the 4 factors
- (also in High Protein Low Carb Diet section)