Abstract
Background
Dietary guidelines recommend substituting animal protein with plant protein, however, the ideal ratio of plant-to-animal protein (P:A) remains unknown.
Objectives
We aimed to evaluate associations between the P:A ratio and incident cardiovascular disease (CVD), coronary artery disease (CAD), and stroke in 3 cohorts.
Methods
Multivariable-adjusted Cox proportional hazard models were used to estimate hazard ratios (HRs) for CVD outcomes among 70,918 females in the Nurses’ Health Study (NHS) (1984–2016), 89,205 females in the NHSII (1991–2017) and 42,740 males from the Health Professionals Follow-up Study (1986–2016). The P:A ratio was based on percent energy from plant and animal protein and assessed using food frequency questionnaires every 4 y.
Results
During 30 y of follow-up, 16,118 incident CVD cases occurred. In the pooled multivariable-adjusted models, participants had a lower risk of total CVD [HR: 0.81; 95% confidence interval (CI): 0.76, 0.87; P trend < 0.001], CAD (HR: 0.73; 95% CI: 0.67, 0.79; P trend < 0.001), but not stroke (HR: 0.98; 95% CI: 0.88, 1.09; P trend = 0.71), when comparing highest to lowest deciles of the P:A ratio (ratio: ∼0.76 compared with ∼0.24). Dose–response analyses showed evidence of linear and nonlinear relationships for CVD and CAD, with more marked risk reductions early in the dose-response curve. Lower risk of CVD (HR: 0.72; 95% CI: 0.64, 0.82) and CAD (HR: 0.64; 95% CI: 0.55, 0.75) were also observed with higher ratios and protein density (20.8% energy) combined. Substitution analyses indicated that replacing red and processed meat with several plant protein sources showed the greatest cardiovascular benefit.
Conclusions
In cohort studies of United States adults, a higher plant-to-animal protein ratio is associated with lower risks of CVD and CAD, but not stroke. Furthermore, a higher ratio combined with higher protein density showed the most cardiovascular benefit.
Keywords: protein ratio, plant protein, animal protein, cardiovascular disease, prospective cohort study
Introduction
Cardiovascular disease (CVD) remains one of the leading causes of death in the United States [1]. Clinical practice guidelines internationally recommend replacing some animal protein with plant protein in the diet to prevent CVD [2,3]. Dietary guidelines, such as Canada’s Food Guide, also recommend choosing protein foods from plants more often [4]. Other guidelines, such as the EAT-Lancet Commission, likewise advise consuming more protein from plant sources for planetary health [5]. Despite these recommendations, most protein comes from animal sources in developed countries. For example, the average ratio of plant-to-animal protein in the diet is ∼1:3 (0.33) in the United States [6,7]. Healthy dietary patterns frequently encouraged by guidelines may contain a range of animal and plant protein intake, however, the optimal ratio in the diet for CVD prevention is unknown [8]. The optimal ratio of these protein types has not been extensively evaluated and there are no recommendations for the relative contribution from these sources of protein, therefore, assessing ratios of the plant-to-animal protein will help refine public health guidance (as a percentage of protein coming from plants and animals) on protein sources. We, therefore, examined the association between deciles of plant-to-animal protein ratios with CVD in 3 large United States prospective cohort studies, including dose–response relationships. As higher protein diets have previously been thought to raise CVD risk [9], we also examined the joint effects of the plant-to-animal protein ratio with protein density (the percentage of total calories derived from protein) to explore CVD risk associations with higher protein intake. Lastly, as 2 diets with the same plant-to-animal protein ratio can include different sources of protein, we used statistical substitution models to estimate risk of CVD when the top animal protein sources in the diet were replaced with plant protein sources.
Methods
Study design and population
This analysis was conducted in 3 ongoing prospective cohort studies: the Nurses’ Health Study (NHS), NHSII, and Health Professionals Follow-up Study (HPFS) [10,11]. The NHS was initiated in 1976, with 121,700 female nurses aged 30–55 y. The NHSII started in 1989 and includes 116,429 female nurses aged 25–42 y. The HPFS began in 1986, recruiting 51,529 male health professionals aged 40–75 y. In all 3 cohorts, participants were followed up on lifestyle, medical history, and other health-related factors every 2–4 y, with a response rate of ∼90% over total person-time until 2016–2017. Participants who had cancer, CVD, those who did not complete food frequency questionnaires (FFQs) or met the FFQ exclusion criteria (missing dietary data or a reported energy intake of <600 or >3500 kcal/d for females and <800 or >4200 kcal/d for males) at baseline in each cohort (1984 for NHS, 1991 for NHSII and 1986 for HPFS) were excluded. Ratio outliers (<1% or >99% percentile), including those who consume no animal protein, were also excluded. After exclusions, 70,918 females from NHS, 89,205 females from NHSII, and 42,740 males from HPFS remained (see participant flow chart in Supplemental Figure 1). The study protocol was approved by the Institutional Review Boards of Brigham and Women’s Hospital and the Harvard T.H. Chan School of Public Health and informed consent was implied by the return of the cohort questionnaires.
Dietary assessment
A validated semiquantitative FFQ was administered every 4 y to determine dietary intake, starting in 1984 in NHS, 1991 in NHSII, and 1986 in HPFS [[12], [13], [14]]. Participants were asked how often, on average, they had consumed specific foods in the preceding year. Total protein intake (g/d) was derived from the responses to the FFQs by multiplying the amount and frequency of protein in a questionnaire item and summing across all foods that contained protein. Similarly, this approach was used for animal and plant protein sources. Grams of protein were calculated based on the United States Department of Agriculture (USDA) and the Harvard University Food Composition Database [15]. The plant-to-animal protein ratios were calculated using percentage of calories from plant and animal protein for ease of interpretation and to help account for calorie consumption. Higher ratios indicate higher plant protein consumption. Supplemental Tables 1–3 summarize the top food contributors to animal and plant protein (in g/d). Previous validation studies in these cohorts showed a Pearson correlation coefficient of 0.61 for total protein (g/d) when comparing the FFQ to dietary records [16]. The correlation coefficients ranged from 0.48 to 0.74 for the animal protein sources and from 0.30 to 0.45 for the plant protein sources [17,18]. In primary analyses, the cumulative average of dietary intake at each cycle was used to best represent long-term diet and to help reduce measurement error as the primary exposure [19]. As participants may have altered their diet after diagnosis of a major illness, the dietary variables were not updated when participants reported a diagnosis of coronary revascularization, diabetes, angina, or cancer.
Cardiovascular outcomes
CVD was defined as a composite of incident nonfatal myocardial infarction (MI), fatal coronary artery disease (CAD), and fatal and nonfatal stroke. Nonfatal MI was identified by physicians according to the World Health Organization criteria [20] including diagnostic electrocardiographic changes or elevated cardiac enzymes and nonfatal stroke was identified according to the National Survey of Stroke criteria [21]. Death was confirmed by relatives, postal authorities, or the National Death Index [22]. Fatal CAD and stroke were determined by autopsy records or death certificate/other evidence. Cause of death was defined according to the International Classification of Diseases-8th Revision.
Covariates
Participants provided information on their medical history, family history, lifestyle, BMI (kg/m2), reproductive factors, and medication use through questionnaires every 2–4 y. Information on total energy intake and alcohol consumption was obtained from the FFQ. To assess overall diet quality, we calculated a modified 2010 Alternative Healthy Eating Index (AHEI) [23], excluding alcohol and foods already included as part of the plant-to-animal protein ratio.
Statistical analyses
Person-years of follow-up were calculated from the return date of the first FFQ until the date of CVD diagnosis, death, or end of follow-up (June 2016 in NHS, June 2017 in NHSII, and January 2016 in HPFS), whichever came first. Cox regressions with time-varying covariates were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of CVD outcomes comparing high to low deciles of the plant-to-animal protein ratio (deciles were examined to assess a greater range of ratios). The ratio was also analyzed as a continuous variable, per 1-SD increase. To assess linear trends, we used the median value of the ratio within each decile as a continuous variable. Restricted cubic spline plots with 4 knots were used for curve fitting of the association between the ratio and total CVD, CAD, stroke, and stroke subtypes (ischemic and hemorrhagic). P values for nonlinearity and a linear relation were determined from the likelihood ratio tests. For the joint analyses, the ratios were combined into 3 categories: 1) low, deciles 1–3, 2) medium, deciles 4–7, and 3) high, deciles 8–10. Protein density (i.e., percent total energy intake from protein), was categorized as tertiles [lower (median 15.8%), medium (18.2%), and higher density (20.8%)]. All analyses were stratified by age and follow-up intervals, and multivariable models were adjusted for race, smoking, menopausal status, postmenopausal hormone use (in females), oral contraceptive use (in females), multivitamin use, regular aspirin use, physical activity, family history of MI, family history of diabetes, marital status, BMI, total energy intake, alcohol intake, socioeconomic status (census-tract median family income, median home value, and percentage with a college degree), the modified AHEI and baseline hypercholesterolemia (includes cholesterol-lowering medications), hypertension (includes blood pressure-lowering medications), and diabetes status. Analyses were performed separately in each cohort and pooled HRs were obtained by combining data from the 3 cohorts.
We conducted several sensitivity analyses. Instead of using the cumulative average diet, the baseline ratio and the most recent measure of the ratio (simple update) were used. Second, the cumulative average ratio was continuously updated until the end of follow-up rather than stop updating when an intermediate disease occurred (hypercholesterolemia or medication, hypertension or medication, and diabetes). Third, the primary models were adjusted for updated diagnoses of intermediate diseases (hypercholesterolemia or medication, hypertension or medication, and diabetes). Fourth, we adjusted for additional dietary variables, including carbohydrate, fiber, saturated fat, MUFAs, PUFAs, and trans fat (g/d) instead of the modified AHEI. Effect modification was also examined according to sex (female/male), baseline hypercholesterolemia status (yes/no), baseline hypertension status (yes/no), physical activity (below or above median), BMI category (<30/≥30), smoking status (ever/never), age (<60/≥60), AHEI (below or above median) and multivitamin use (yes/no) and were evaluated using the Wald test on cross-product terms based on a 1-SD increase in the ratio and the stratification variables.
We used a leave-one-nutrient-out model to evaluate the association between the isocaloric substitution of 3% of energy from animal protein for plant protein and CVD risk [24]. We also used statistical substitution models for the top animal protein sources in the diet replaced with plant protein sources to provide a better understanding of dietary recommendations when considering plant and animal protein sources [24]. For each substitution of 1 protein food item (servings/d) for another, we exponentiated the difference between the β-coefficients of the 2 foods to estimate the HR, and we used the variances and covariance of the 2 food items to estimate the 95% CI. Further analyses included estimating 2 additional ratios: 1) where more unhealthy animal protein sources (red and processed meat) were removed and 2) where more unhealthy plant protein sources (refined grains) were removed from the ratios. We also examined deciles of total protein, animal protein, and plant protein density (as a percentage of energy) with each of the CVD outcomes. Lastly, we examined joint effects of plant and animal protein density (lower, medium, and higher categories) with each outcome. The proportional hazards assumption was evaluated with a likelihood ratio test comparing the model with and without an interaction term between age and the ratio, and the tests did not indicate a violation in any cohort. The P values and 95% CIs presented in this report were not adjusted for multiplicity, and therefore inferences drawn from these statistics may require cautious interpretations. Analyses were performed with the SAS statistical package 9.4 (SAS Institute). Statistical tests were 2-sided, and P values of <0.05 were considered statistically significant.
Results
Descriptive characteristics
During 30 y of follow-up, 16,118 incident CVD cases, including 10,187 CAD cases and 6137 stroke cases, were documented. The median plant-to-animal protein ratio increased from ∼0.36 (1:3) to ∼0.50 (1:2) over follow-up (Figure 1). The cumulative average ratio ranged from ∼0.15 to 1.84. The median calorie density from total protein remained consistent over follow-up (∼17%–19%), with animal protein decreasing from ∼13% to ∼11% of total caloric intake, and plant protein increasing from ∼5% to ∼6% of total caloric intake (Supplemental Figures 2–4). Poultry, red meat, and dairy were consistently the top contributors to animal protein across cohorts (Supplemental Tables 1–3). Refined grains, whole grains, potatoes, nuts, and beans were the top contributors to plant protein (Supplemental Tables 1–3). Table 1 shows the baseline characteristics by cohort according to deciles of the ratio. Participants with a higher ratio had a lower BMI, were more physically active, were less likely to smoke, and had higher intakes of fiber and carbohydrates, a lower saturated fat intake, and a higher AHEI score, among others.
FIGURE 1.
Trends in the median plant-to-animal protein ratio in the 3 prospective cohort studies. Secular population trends of plant-to-animal protein ratio indicate changing from ∼1:3 to ∼1:2 across all cohorts over time. NHS, Nurses’ Health Study; HPFS, Health Professionals Follow-up Study.
TABLE 1.
Baseline characteristics of participants according to select deciles of the plant-to-animal protein ratio.
| Variable | Decile 1 | Decile 3 | Decile 5 | Decile 8 | Decile 10 |
|---|---|---|---|---|---|
| NHS (1984) | |||||
| No. of participants | 7091 | 7095 | 7073 | 7097 | 7091 |
| Mean dietary protein ratio | 0.21 | 0.29 | 0.36 | 0.47 | 0.73 |
| Age, mean (SD), y | 50.3 (7.0) | 50.0 (7.1) | 50.0 (7.2) | 50.2 (7.2) | 51.1 (7.3) |
| Non-Hispanic White race/ethnicity1 | 97.7 | 97.7 | 98.0 | 97.8 | 97.2 |
| BMI, mean (SD) | 25.7 (5.0) | 25.3 (4.8) | 25.1 (4.7) | 24.7 (4.5) | 24.1 (4.4) |
| Physical activity, mean (SD), MET-h/wk | 13.4 (20.6) | 13.5 (17.6) | 14.1 (18.7) | 14.3 (20.2) | 15.1 (22.6) |
| Smoker, current | 28.7 | 24.7 | 24.1 | 21.6 | 22.1 |
| Alcohol, mean (SD), g/d | 7.5 (12.0) | 7.1 (11.3) | 7.2 (11.5) | 6.5 (10.8) | 6.0 (10.5) |
| Diabetes | 2.5 | 2.8 | 2.3 | 2.0 | 1.8 |
| Hypertension | 20.1 | 18.7 | 17.9 | 16.1 | 15.7 |
| Hypercholesterolemia | 6.2 | 6.3 | 6.1 | 6.0 | 6.5 |
| Family history of MI | 25.5 | 25.6 | 25.6 | 25.7 | 23.3 |
| Total energy intake, mean (SD), kcal/d | 1607 (490) | 1722 (507) | 1768 (522) | 1800 (534) | 1800 (570) |
| Multivitamin use | 36.9 | 36.5 | 36.3 | 36.9 | 38.1 |
| Aspirin use | 70.2 | 71.6 | 73.7 | 71.5 | 68.5 |
| Carbohydrates, mean (SD), % energy | 39.9 (7.1) | 43.9 (6.3) | 46.2 (6.2) | 49.5 (6.1) | 53.2 (7.2) |
| Saturated fat, mean (SD), % energy | 13.8 (2.7) | 13.1 (2.3) | 12.6 (2.3) | 11.9 (2.2) | 10.9 (2.4) |
| Monounsaturated fat, mean (SD), % energy | 13.5 (2.5) | 13.0 (2.3) | 12.7 (2.3) | 12.3 (2.2) | 11.7 (2.6) |
| Polyunsaturated fat, mean (SD), % energy | 6.2 (1.7) | 6.5 (1.6) | 6.6 (1.6) | 6.8 (1.7) | 7.2 (2.0) |
| Trans fat, mean (SD), % energy | 1.9 (0.6) | 1.9 (0.6) | 1.9 (0.6) | 1.9 (0.6) | 1.9 (0.7) |
| Protein, mean (SD), % energy | 21.1 (3.1) | 19.0 (2.5) | 17.8 (2.4) | 16.2 (2.2) | 14.6 (2.4) |
| Fiber, mean (SD), g/d | 13.6 (3.9) | 15.3 (3.9) | 16.2 (4.1) | 17.6 (4.5) | 19.6 (5.7) |
| Animal protein, mean (SD), g/d | 69.1 (10.3) | 59.0 (7.62) | 52.8 (7.0) | 44.5 (6.2) | 34.3(6.3) |
| Plant protein, mean (SD), g/d | 14.7 (2.2) | 17.2 (2.3) | 18.7 (2.5) | 20.9 (2.9) | 24.4 (4.3) |
| AHEI score, mean (SD) | 46.8 (10.6) | 47.0 (10.2) | 47.7 (10.3) | 49.2 (10.7) | 52.0 (11.8) |
| Vitamin B12, mean (SD), mcg/d | 13.4 (13.9) | 12.8 (26.3) | 11.8 (22.0) | 10.9 (23.1) | 9.2 (16.0) |
| Iron, mean (SD), mg/d | 16.7 (16.1) | 16.9 (16.3) | 17.1 (15.2) | 17.2 (14.8) | 17.9 (15.8) |
| Calcium, mean (SD), mg/d | 970 (484) | 914 (440) | 867 (394) | 839 (392) | 825 (424) |
| EPA, mean (SD), g/d | 0.08 (0.08) | 0.07 (0.06) | 0.06 (0.05) | 0.05 (0.04) | 0.04 (0.04) |
| DHA, mean (SD), g/d | 0.17 (0.13) | 0.14 (0.10) | 0.13 (0.09) | 0.11 (0.07) | 0.09 (0.07) |
| Zinc, mean (SD), mg/d | 16.9 (15.4) | 15.8 (14.1) | 14.8 (12.6) | 14.3 (13.0) | 14.5 (15.3) |
| Vitamin D, mean (SD), IU/d | 369 (276) | 329 (242) | 311 (236) | 288 (228) | 277 (255) |
| Refined grains, servings/d | 1.3 (1.0) | 1.6 (1.2) | 1.8 (1.3) | 2.0 (1.5) | 2.2 (1.7) |
| Legumes, servings/d | 0.2 (0.2) | 0.2 (0.2) | 0.2 (0.2) | 0.2 (0.2) | 0.2 (0.2) |
| Whole grains, servings/d | 0.7 (0.7) | 0.9 (0.9) | 1.0 (1.0) | 1.3 (1.2) | 1.6 (1.5) |
| Nuts, servings/d | 0.1 (0.1) | 0.15 (0.2) | 0.18 (0.2) | 0.3 (0.3) | 0.4 (0.6) |
| Dairy, servings/d | 2.3 (1.6) | 2.2 (1.4) | 2.0 (1.3) | 1.8 (1.2) | 1.5 (1.1) |
| Red meat, servings/d | 1.4 (0.8) | 1.3 (0.7) | 1.2 (0.7) | 1.0 (0.6) | 0.7 (0.5) |
| Poultry, servings/d | 0.6 (0.5) | 0.6 (0.4) | 0.5 (0.4) | 0.45 (0.3) | 0.4 (0.3) |
| Animal protein, % energy | 17.4 (2.7) | 14.7 (1.9) | 13.1 (1.8) | 11.0 (1.5) | 8.5 (1.6) |
| Plant protein, % energy | 3.7 (0.6) | 4.3 (0.6) | 4.7 (0.6) | 5.2 (0.7) | 6.1 (1.1) |
| NHSII (1991) | |||||
| No. of participants | 8919 | 8920 | 8920 | 8920 | 8919 |
| Mean dietary protein ratio | 0.21 | 0.28 | 0.33 | 0.44 | 0.77 |
| Age, mean (SD), y | 35.9 (4.7) | 35.9 (4.7) | 36.1 (4.7) | 36.3 (4.6) | 36.5 (4.6) |
| Non-Hispanic White race/ethnicity1 | 95.6 | 96.8 | 96.6 | 96.6 | 95.4 |
| BMI, mean (SD) | 25.7 (5.8) | 24.0 (5.4) | 24.7 (5.3) | 24.2 (5.1) | 23.2 (4.6) |
| Physical activity, mean (SD), MET-h/wk | 19.2 (25.5) | 19.7 (25.4) | 19.6 (24.8) | 21.4 (28.2) | 26.4 (34.2) |
| Smoker, current | 14.7 | 12.6 | 12.1 | 10.9 | 1.0 |
| Alcohol, mean (SD), g/d | 2.79 (6.07) | 3.03 (6.02) | 3.05 (6) | 3.25 (5.98) | 3.37 (6.41) |
| Diabetes | 1.2 | 1.0 | 0.9 | 1.0 | 0.8 |
| Hypertension | 7.8 | 6.9 | 6.0 | 6.0 | 4.9 |
| Hypercholesterolemia | 16.1 | 14.4 | 14.3 | 14.4 | 13.7 |
| Family history of MI | 21.5 | 20.7 | 20.0 | 20.0 | 19.7 |
| Total energy intake, mean (SD), kcal/d | 1685 (515) | 1776 (524) | 1827 (550) | 1853 (566) | 1780 (577) |
| Multivitamin use | 42.6 | 43.0 | 44.6 | 44.2 | 46.9 |
| Aspirin use | 11.7 | 11.1 | 11.3 | 11.4 | 11.4 |
| Carbohydrates, mean (SD), % energy | 43.5 (6.5) | 47.1 (5.9) | 49.3 (5.8) | 52.4 (5.9) | 57.6 (6.9) |
| Saturated fat, mean (SD), % energy | 12.5 (2.4) | 11.8 (2.2) | 11.4 (2.1) | 10.7 (2.2) | 12.5 (2.4) |
| Monounsaturated fat, mean (SD), % energy | 12.7 (2.5) | 12.3 (2.3) | 12.1 (2.3) | 11.8 (2.3) | 10.8 (2.7) |
| Polyunsaturated fat, mean (SD), % energy | 5.5 (1.4) | 5.6 (1.3) | 5.6 (1.3) | 5.7 (1.4) | 5.7 (1.6) |
| Trans fat, mean (SD), % energy | 0.7 (0.3) | 0.7 (0.3) | 0.7 (0.3) | 0.7 (0.3) | 0.7 (0.3) |
| Protein, mean (SD), % energy | 22.8 (3.1) | 20.7 (2.7) | 19.5 (2.5) | 17.8 (2.4) | 15.6 (2.4) |
| Fiber, mean (SD), g/d | 14.7 (3.8) | 16.8 (4.2) | 17.9 (4.3) | 19.7 (4.9) | 23.3 (6.9) |
| Animal protein, mean (SD), g/d | 83.9 (11.6) | 72.3 (9.1) | 65.3 (8.3) | 55.2 (7.3) | 40.2 (8.6) |
| Plant protein, mean (SD), g/d | 17.4 (2.4) | 20.2 (2.6) | 21.9 (2.8) | 24.5 (3.2) | 29.5 (5.1) |
| Vitamin B12, mean (SD), mcg/d | 11.3 (12.0) | 10.3 (11.8) | 9.7 (11.4) | 8.8 (10.2) | 8.9 (19.7) |
| Iron, mean (SD), mg/d | 23.3 (25.5) | 23.6 (24.1) | 24.2 (23.7) | 24.2 (23.0) | 26.7 (26.5) |
| Calcium, mean (SD), mg/d | 1126 (478) | 1059 (438) | 1019 (419) | 957 (399) | 966 (439) |
| EPA, mean (SD), g/d | 0.07 (0.08) | 0.06 (0.07) | 0.06 (0.06) | 0.05 (0.06) | 0.05 (0.06) |
| DHA, mean (SD), g/d | 0.18 (0.14) | 0.16 (0.12) | 0.15 (0.1) | 0.13 (0.1) | 0.11 (0.1) |
| Zinc, mean (SD), mg/d | 17.0 (11.2) | 16.0 (10.6) | 15.6 (10.3) | 14.8 (10.9) | 14.8 (14.1) |
| Vitamin D, mean (SD), IU/d | 446 (285) | 406 (257) | 389 (249) | 358 (249) | 356 (278) |
| AHEI score, mean (SD) | 45.1 (10.0) | 46.2 (10.1) | 46.9 (9.9) | 48.7 (10.7) | 54.4 (11.8) |
| Refined grains, servings/d | 1.3 (0.8) | 1.6 (0.9) | 1.8 (1.0) | 2.0 (1.2) | 2.2 (1.5) |
| Legumes, servings/d | 0.15 (0.2) | 0.17 (0.2) | 0.17 (0.2) | 0.17 (0.2) | 0.19 (0.3) |
| Whole grains, servings/d | 0.8 (0.7) | 1.0 (0.9) | 1.2 (1.0) | 1.5 (1.2) | 1.9 (1.5) |
| Nuts, servings/d | 0.1 (0.1) | 0.12 (0.1) | 0.15 (0.2) | 0.2 (0.2) | 0.3 (0.4) |
| Dairy, servings/d | 2.7 (1.7) | 2.5 (1.5) | 2.4 (1.4) | 2.1 (1.3) | 1.8 (1.2) |
| Red meat, servings/d | 1.2 (0.8) | 1.1 (0.7) | 1.0 (0.6) | 0.9 (0.5) | 0.5 (0.4) |
| Poultry, servings/d | 0.9 (0.6) | 0.8 (0.4) | 0.7 (04) | 0.6 (0.4) | 0.4 (0.3) |
| Animal protein, % energy | 18.9 (2.7) | 16.2 (2.1) | 15.3 (2.0) | 12.3 (1.7) | 9.0 (1.9) |
| Plant protein, % energy | 3.9 (0.6) | 4.5 (0.6) | 4.7 (0.6) | 5.5 (0.7) | 6.6 (1.1) |
| HPFS (1986) | |||||
| No. of participants | 4274 | 4274 | 4287 | 4274 | 4274 |
| Mean dietary protein ratio | 0.20 | 0.29 | 0.35 | 0.47 | 0.79 |
| Age, mean (SD), y | 53.6 (9.4) | 53.3 (9.5) | 53.0 (9.4) | 52.9 (9.6) | 53.7 (9.7) |
| Non-Hispanic White race/ethnicity1 | 95.1 | 95.6 | 94.9 | 95.2 | 93.0 |
| BMI, mean (SD) | 26.2 (3.4) | 25.8 (3.3) | 25.5 (3.5) | 25.2 (3.2) | 24.7 (3.1) |
| Physical activity, mean (SD), MET-h/wk | 17.1 (22.3) | 18.2 (22.8) | 19.8 (24.2) | 21.5 (24.9) | 25.5 (30.1) |
| Smoker, current | 12.2 | 10.3 | 9.0 | 7.2 | 6.5 |
| Alcohol, mean (SD), g/d | 11.2 (15.4) | 11.8 (15.7) | 11.9 (15.4) | 11.3 (15.1) | 9.7 (14.4) |
| Diabetes | 3.4 | 2.7 | 2.5 | 2.1 | 2.1 |
| Hypertension | 22.3 | 21.5 | 19.9 | 18.7 | 17.9 |
| Hypercholesterolemia | 9.9 | 10.3 | 10.2 | 10.4 | 11.5 |
| Family history of MI | 15.1 | 15.1 | 15.0 | 14.9 | 14.1 |
| Total energy intake, mean (SD), kcal/d | 1875 (594) | 1976 (605) | 2009 (605) | 2021 (623) | 2032 (660) |
| Multivitamin use | 38.8 | 40.1 | 42.3 | 43.9 | 46.4 |
| Aspirin use | 26.4 | 27.1 | 27.3 | 27.0 | 24.2 |
| Carbohydrates, mean (SD), % energy | 39.8 (7.1) | 44.1 (6.6) | 46.4 (6.4) | 49.9 (6.7) | 54.9 (8.9) |
| Saturated fat, mean (SD), % energy | 12.7 (2.8) | 11.7 (2.4) | 11.1 (2.4) | 10.3 (2.4) | 8.9 (2.7) |
| Monounsaturated fat, mean (SD), % energy | 13.2 (2.6) | 12.6 (2.4) | 12.3 (2.3) | 11.9 (2.6) | 11.2 (3.6) |
| Polyunsaturated fat, mean (SD), % energy | 5.7 (1.4) | 5.8 (1.4) | 5.9 (1.4) | 6.0 (1.5) | 6.4 (2.2) |
| Trans fat, mean (SD), % energy | 0.6 (0.2) | 0.6 (0.2) | 0.6 (0.2) | 0.6 (0.2) | 0.5 (0.3) |
| Protein, mean (SD), % energy | 21.86 (3.2) | 19.66 (2.7) | 18.61 (2.6) | 17.1 (2.4) | 15.48 (2.6) |
| Fiber, mean (SD), g/d | 15.0 (5.9) | 18.4 (6.6) | 20.4 (7.4) | 22.6 (8.2) | 27.7 (12.6) |
| Animal protein, mean (SD), g/d | 89.5 (13.2) | 75.8 (10.2) | 68.6 (9.4) | 57.9 (8.2) | 43.5 (8.9) |
| Plant protein, mean (SD), g/d | 18.3 (2.8) | 21.8 (3.0) | 24.0 (3.3) | 27.2 (3.8) | 33.6 (6.2) |
| Vitamin B12, mean (SD), mcg/d | 14.9 (30.4) | 13.4 (15.8) | 13.0 (18.8) | 11.5 (12.9) | 10.2 (15.0) |
| Iron, mean (SD), mg/d | 17.8 (14.3) | 18.8 (14.5) | 19.3 (14.4) | 20.0 (14.3) | 22.3 (16.8) |
| Calcium, mean (SD), mg/d | 982 (514) | 909 (420) | 887 (389) | 858 (387) | 880 (421) |
| EPA, mean (SD), g/d | 0.13 (0.14) | 0.12 (0.1) | 0.11 (0.1) | 0.1 (0.08) | 0.08 (0.08) |
| DHA, mean (SD), g/d | 0.24 (0.2) | 0.21 (0.15) | 0.2 (0.15) | 0.17 (0.12) | 0.14 (0.12) |
| Zinc, mean (SD), mg/d | 22.0 (24.5) | 21.2 (23.0) | 20.8 (22.2) | 20.2 (21.4) | 21.6 (25.4) |
| Vitamin D, mean (SD), IU/d | 469 (343) | 413 (291) | 406 (296) | 383 (306) | 367 (316) |
| AHEI score, mean (SD) | 48.2 (10.7) | 50.1 (10.6) | 52.0 (10.8) | 54.4 (11.2) | 59.0 (12.0) |
| Refined grains, servings/d | 1.2 (1.0) | 1.4 (1.1) | 1.5 (1.2) | 1.7 (1.3) | 1.7 (1.5) |
| Legumes, servings/d | 0.3 (0.3) | 0.4 (0.3) | 0.4 (0.3) | 0.5 (0.4) | 0.6 (0.5) |
| Whole grains, servings/d | 0.9 (0.8) | 1.2 (1.1) | 1.4 (1.2) | 1.8 (1.5) | 2.5 (2.1) |
| Nuts, servings/d | 0.2 (0.2) | 0.25 (0.3) | 0.3 (0.3) | 0.4 (0.5) | 0.8 (1) |
| Dairy, servings/d | 2.4 (1.9) | 2.2 (1.5) | 2.0 (1.4) | 1.7 (1.2) | 1.5 (1.1) |
| Red meat, servings/d | 1.6 (1.0) | 1.4 (0.9) | 1.3 (0.8) | 1.0 (0.7) | 0.6 (0.6) |
| Poultry, servings/d | 0.8 (0.6) | 0.6 (0.4) | 0.6 (0.4) | 0.5 (0.3) | 0.4 (0.3) |
| Animal protein, % energy | 18.2 (2.8) | 15.3 (2.1) | 13.8 (1.9) | 11.7 (1.7) | 8.7 (1.8) |
| Plant protein, % energy | 3.7 (0.6) | 4.4 (0.6) | 4.8 (0.8) | 5.5 (0.8) | 6.7 (1.3) |
Abbreviations: AHEI, alternate healthy eating index; HPFS, HPFS, Health Professionals Follow-up Study; MI, myocardial infarction, MET, metabolic equivalent task; NHS, Nurses’ Health Study.
Values are mean ± SD or percentage and are standardized to the age distribution of the study population.
Further details are provided in Supplemental Table 13.
Plant-to-animal protein ratio and risk of cardiovascular outcomes
Table 2 shows the results of the pooled findings for deciles of the plant-to-animal protein ratio for total CVD, CAD, and stroke; findings by each cohort are provided in Supplemental Tables 4–6. Overall, a higher ratio was associated with a lower risk of total CVD and CAD, but not stroke. Findings were generally similar in direction and magnitude across the 3 cohorts, although the findings were not significant in the NHSII, potentially due to fewer cases in this cohort. In the pooled analyses, the most adjusted multivariable model 3 HR (95% CIs) for total CVD, comparing the highest decile to the lowest decile (ratio of ∼0.24 compared with ∼0.76), was 0.81 (0.76, 0.87; P trend < 0.001, see Table 2). The pooled HRs (95% CIs) in model 3 for CAD was 0.73 (0.67, 0.79; P trend < 0.001, see Table 2) and for stroke was 0.98 (0.88, 1.09; P trend = 0.71, see Table 2), comparing extreme deciles. The inclusion of intermediate risk factors in the models consistently attenuated these associations by ∼5% (comparing models 2 to 3).
TABLE 2.
Pooled associations of deciles of the plant-to-animal protein ratio with the cardiovascular disease outcomes.
| Decile 1 HR (95% CI) | Decile 2 HR (95% CI) | Decile 3 HR (95% CI) | Decile 4 HR (95% CI) | Decile 5 HR (95% CI) | Decline 6 HR (95% CI) | Decile 7 HR (95% CI) | Decile 8 HR (95% CI) | Decile 9 HR (95% CI) | Decile 10 HR (95% CI) | P trend | HR (95% CI) per 1-SD | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Median ratio | 0.24 (1:4.2) | 0.30 (1:3.3) | 0.34 (1:2.9) | 0.37 (1:2.7) | 0.40 (1:2.5) | 0.43 (1:2.3) | 0.46 (1:2.2) | 0.50 (1:2.0) | 0.57 (1:1.8) | 0.76 (1:1.3) | 1-SD (0.18) | |
| Cardiovascular disease | ||||||||||||
| Cases/person-years | 2053/536,687 | 1814/539,977 | 1573/541,073 | 1575/541,623 | 1536/542,719 | 1549/542,619 | 1530/542,810 | 1455/543,302 | 1513/542,630 | 1520/542,558 | ||
| Age-adjusted model 1 | 1.00 (ref) | 0.90 (0.84, 0.95) | 0.78 (0.73, 0.83) | 0.77 (0.73, 0.83) | 0.74 (0.69, 0.79) | 0.74 (0.69, 0.79) | 0.71 (0.67, 0.76) | 0.66 (0.62, 0.71) | 0.67 (0.63, 0.72) | 0.64 (0.60, 0.69) | <0.001 | 0.89 (0.87, 0.90) |
| Multivariable model 2 | 1.00 (ref) | 0.93 (0.88, 0.99) | 0.84 (0.79, 0.90) | 0.85 (0.80, 0.91) | 0.82 (0.77, 0.88) | 0.84 (0.78, 0.90) | 0.81 (0.76, 0.87) | 0.77 (0.72, 0.82) | 0.78 (0.73, 0.83) | 0.76 (0.71, 0.81) | <0.001 | 0.93 (0.91, 0.95) |
| Multivariable model 3 | 1.00 (ref) | 0.96 (0.90, 1.02) | 0.87 (0.82, 0.93) | 0.89 (0.83, 0.95) | 0.87 (0.81, 0.93) | 0.89 (0.83, 0.95) | 0.86 (0.81, 0.92) | 0.82 (0.76, 0.87) | 0.83 (0.78, 0.89) | 0.81 (0.76, 0.87) | <0.001 | 0.95 (0.93, 0.96) |
| Coronary artery disease | ||||||||||||
| Cases/person-years | 1390/537,184 | 1209/540,444 | 1027/541,516 | 1013/542,067 | 939/543,198 | 931/543,114 | 959/543,232 | 920/543,724 | 911/543,065 | 888/543,062 | ||
| Age-adjusted model 1 | 1.00 (ref) | 0.88 (0.81, 0.95) | 0.75 (0.69, 0.82) | 0.74 (0.68, 0.80) | 0.67 (0.62, 0.73) | 0.66 (0.61, 0.72) | 0.66 (0.61, 0.72) | 0.62 (0.57, 0.68) | 0.60 (0.55, 0.65) | 0.56 (0.52, 0.61) | <0.001 | 0.85 (0.83, 0.86) |
| Multivariable model 2 | 1.00 (ref) | 0.92 (0.86, 1.00) | 0.82 (0.76, 0.89) | 0.82 (0.76, 0.89) | 0.76 (0.70, 0.83) | 0.76 (0.70, 0.83) | 0.77 (0.71, 0.84) | 0.74 (0.68, 0.80) | 0.71 (0.66, 0.78) | 0.68 (0.62, 0.74) | <0.001 | 0.89 (0.87, 0.91) |
| Multivariable model 3 | 1.00 (ref) | 0.95 (0.88, 1.03) | 0.86 (0.79, 0.93) | 0.87 (0.80, 0.94) | 0.81 (0.74, 0.88) | 0.81 (0.75, 0.88) | 0.83 (0.76, 0.90) | 0.79 (0.73, 0.86) | 0.77 (0.70, 0.83) | 0.73 (0.67, 0.79) | <0.001 | 0.91 (0.89, 0.93) |
| Stroke | ||||||||||||
| Cases/person-years | 688/537,355 | 632/540,619 | 569/541,629 | 588/542,198 | 611/543,244 | 635/543,152 | 592/543,430 | 550/543,820 | 621/543,159 | 651/543,053 | ||
| Age-adjusted model 1 | 1.00 (ref) | 0.93 (0.84, 1.04) | 0.84 (0.76, 0.94) | 0.86 (0.77, 0.96) | 0.88 (0.79, 0.98) | 0.90 (0.81, 1.01) | 0.82 (0.74, 0.92) | 0.74 (0.67, 0.83) | 0.82 (0.73, 0.91) | 0.81 (0.73, 0.90) | <0.001 | 0.95 (0.93, 0.98) |
| Multivariable model 2 | 1.00 (ref) | 0.96 (0.86, 1.07) | 0.89 (0.80, 1.00) | 0.92 (0.82, 1.03) | 0.95 (0.85, 1.06) | 0.99 (0.88, 1.10) | 0.90 (0.81, 1.01) | 0.83 (0.74, 0.93) | 0.92 (0.82, 1.03) | 0.93 (0.84, 1.04) | 0.17 | 0.99 (0.96, 1.01) |
| Multivariable model 3 | 1.00 (ref) | 0.98 (0.88, 1.09) | 0.92 (0.82, 1.03) | 0.96 (0.85, 1.07) | 0.99 (0.88, 1.10) | 1.03 (0.92, 1.15) | 0.95 (0.85, 1.06) | 0.87 (0.78, 0.97) | 0.97 (0.87, 1.08) | 0.98 (0.88, 1.09) | 0.71 | 1.00 (0.98, 1.03) |
Abbreviations: AHEI, alternate healthy eating index; NHS, Nurses’ Health Study.
Hazard ratios (HRs) were determined using Cox regression models. Analysis was conducted after combining all 3 cohorts. The plant-to-animal protein ratio was calculated based on cumulatively averaged nutrient density from all preceding food frequency questionnaires up to each follow-up interval. Decile 10 reflects the highest amount of plant protein to animal protein, Decile 1 reflects the lowest. Models were stratified by cohort, age in months, and follow-up period. No corrections for multiple tests were applied. Model 1 was adjusted for age (months). Multivariable model 2 was adjusted for age and race (White and non-White), smoking (never; past; and current: 1–14, 15–24, and ≥25 cigarettes/d), menopausal status, and postmenopausal hormone use (premenopausal, never/past users of hormone therapy, and current users of hormone therapy; only in NHS and NHSII), oral contraceptive use (never, past, and current; only in NHSII), multivitamin use (no/yes), regular aspirin use (no/yes), physical activity (<3, 3–9, 9–18, 18–27, 27–42, and ≥42 metabolic equivalents/wk), family history of myocardial infarction (no/yes), family history of diabetes (no/yes), marital status (married, widowed, and divorced/separated), BMI (<23, 23–24.9, 25–29,9, 30–34.9, and ≥35), alcohol intake, total energy intake, modified AHEI score (including vegetables, fruits, sugar-sweetened beverages, trans fatty acids (TFAs), PUFAs, and sodium), and socioeconomic status (in quintiles). Multivariable model 3 was further adjusted for baseline hypercholesterolemia (no/yes), hypertension (no/yes), and diabetes (no/yes).
In addition, a 1-SD increase in the ratio (∼0.18) was associated with a lower risk of total CVD (pooled HR: 0.95; 95% CIs: 0.93, 0.96), CAD (pooled HR: 0.91; 95% CI: 0.89, 0.93), but not stroke (pooled HR: 1.00; 95% CI: 0.98, 1.03) in the most adjusted model 3 (see Tables 2). Decile findings for ischemic and hemorrhagic strokes were not significant (Supplemental Table 7).
Figure 2 shows the pooled dose–response relationships of the plant-to-animal protein ratio with risk of cardiovascular outcomes. Total CVD (P linearity < 0.001 and P curvature < 0.001) and CAD (P linearity < 0.001 and P curvature < 0.001) showed evidence of linear and nonlinear relationships with marked risk reductions observed early in the dose-response curve followed by more gradual reductions. There was no evidence of linear or nonlinear associations for stroke (P linearity = 0.86; P curvature = 0.08). Supplemental Figure 5 shows the pooled dose–response relationships for stroke subtypes, and like total stroke, there was no evidence of linear or nonlinear associations for ischemic (P linearity = 0.33; P curvature = 0.16;) and hemorrhagic (P linearity = 0.90; P curvature = 0.37) stroke.
FIGURE 2.
Dose–response relationship of the plant-to-animal protein ratio with risk of cardiovascular outcomes. (A) Total cardiovascular disease, (B) coronary artery disease, and (C) stroke. Analysis was conducted after combining all 3 cohorts. Multivariable model was adjusted for age, race, smoking, menopausal status and postmenopausal hormone use, oral contraceptive use, multivitamin use, regular aspirin use, physical activity, family history of myocardial infarction, family history of diabetes, marital status, BMI, alcohol intake, total energy intake, modified AHEI score, socioeconomic status, baseline hypercholesterolemia, hypertension, and diabetes. Dose–response relationships were determined using restricted cubic splines. Solid lines represent hazard ratios and dotted lines represent 95% confidence intervals.
Joint analyses
In the joint analysis by protein density (Figure 3), the inverse associations with CVD (HR: 0.72; 95% CI: 0.64, 0.82) and CAD (HR: 0.64; 95% CI: 0.55, 0.75) were stronger when higher ratios (>0.50) were combined with higher protein density (median 20.8%). There were no significant associations with stroke. Additionally, we conducted a test for interaction between the 3 protein density groups and 1-SD increase in the ratio. The associations with CVD (P interaction < 0.01) and CAD (P interaction < 0.01) were significantly stronger among individuals with higher protein density (Supplemental Table 8).
FIGURE 3.
Joint associations of the plant-to-animal protein ratio and protein density with cardiovascular outcomes. Hazard ratios (HRs) were determined using Cox regression models. Analysis was conducted after combining all 3 cohorts. The reference group was the lower ratios and lower protein density group. Multivariable model was adjusted for the same covariates as model 3 in Table 2. No corrections for multiple tests were applied.
Sensitivity and subgroup analyses
In the 5 sensitivity analyses (baseline ratio, most recent ratio, continuously updating after intermediate disease diagnoses, adjusting for updated diagnoses of intermediate disease, and adjusting for additional dietary variables), results remained largely consistent comparing pooled highest to lowest ratio deciles (Supplemental Table 9). The associations were consistent across most subgroups and risk of cardiovascular outcomes, however, the association between the plant-to-animal protein ratio and CVD risk was significantly stronger among participants with hypercholesterolemia (P interaction < 0.01) and ever smokers (P interaction < 0.01) (Figure 4).
FIGURE 4.
Subgroup analyses for cardiovascular outcomes for 1-SD increase in the plant-to-animal protein ratio. Hazard ratios (HRs) were determined using Cox regression models. Pooled multivariable-adjusted HRs (indicated by the blue dots) and 95% CIs (indicated by black horizontal lines) of all 3 cohorts are shown. The same covariates were adjusted as model 3 in Table 2, with the exception of not adjusting for the categorical covariate when it was used as a strata. No corrections for multiple tests were applied. AHEI, Alternative Healthy Eating Index; CAD, coronary artery disease; CI, confidence interval; CVD, cardiovascular disease.
Substitution analyses
Replacing 3% energy from animal protein with the same amount of energy from plant protein was associated with an 18% and 24% lower risk of CVD and CAD, respectively, with no association for stroke (Figure 5). We also modeled the replacement of 1 serving/d of the top 3 animal protein sources for plant protein sources. For CVD, replacing poultry with nuts, red and processed meat with all 4 plant protein sources, and dairy for nuts was associated with a lower risk. For CAD, replacing poultry with whole grains and nuts, replacing red and processed meat for all 4 plant protein sources, and replacing dairy with nuts was associated with a lower risk. For stroke, replacing red and processed meat with refined grains and potatoes, whole grains, and nuts was associated with a lower risk.
FIGURE 5.
Substitution analyses for 3% energy replacement of animal for plant protein and top animal protein sources for plant protein sources and the cardiovascular outcomes. Pooled multivariable-adjusted HRs (indicated by the colored dots) and 95% CIs (indicated by the horizontal lines) of all 3 cohorts are shown. HR (95% CI) for each cardiovascular outcome associated with replacing 3% energy from animal protein with plant protein and 1 daily serving of poultry, red meat, and dairy replaced with refined grains and potatoes, whole grains, nuts, and legumes are shown. The Cox proportional hazards models included all protein foods simultaneously and models were stratified by cohort, age in moths, and follow-up period and adjusted for race, smoking, menopausal status and postmenopausal hormone use, oral contraceptive use, multivitamin use, regular aspirin use, physical activity, family history of myocardial infarction, family history of diabetes, marital status, BMI, alcohol intake, total energy intake, modified AHEI score, socioeconomic status, and baseline hypercholesterolemia, hypertension and diabetes. For the energy substitution, the leave-one-out approach to examine the isocaloric substitution of animal for plant protein and included protein density, carbohydrate density, and fat density in the models instead of the modified AHEI. For each substitution of 1 food item for another, we exponentiated the difference between the β-coefficients of the 2 foods to estimate the HR, and we used the variances and covariance of the 2 food items to estimate the 95% CI. No corrections for multiple tests were applied. Red meat included unprocessed and processed sources. AHEI, Alternative Healthy Eating Index; CI, confidence interval.
Additional analyses
Results remained relatively similar when analyzing ratios without protein from refined grains, however, the findings were attenuated, but still significant for CVD and CAD, when protein from red and processed meat were removed from the ratios (Supplemental Tables 10 and 11). For the analysis of deciles of total protein, animal protein, and plant protein density, higher total protein density was associated with a higher risk of CVD and CAD, whereas animal protein density was associated with a higher risk of CAD. Higher plant protein density was associated with a lower risk of CVD and CAD, with stroke findings trending toward lower risk (Supplemental Table 12). In the joint analysis of plant and animal protein density (Supplemental Figure 6), inverse associations with CVD and CAD were only observed in the higher plant protein density category (6% energy) and medium (13% energy) and lower animal protein density (10.5% energy) categories. For stroke, an inverse association was observed in the high plant protein density and medium animal protein density categories.
Discussion
In these 3 large prospective cohort studies, a higher plant-to-animal protein ratio (∼0.76 compared with ∼0.24) was associated with a 19% lower risk of CVD and a 27% lower risk of CAD but was not associated with stroke or stroke subtypes. These findings remained consistent across several subgroups and sensitivity analyses. Dose–response relationships showed that risk reduction became more gradual with higher plant-to-animal protein ratios for CVD and that the optimal ratio may be different for CAD and stroke. Furthermore, joint analyses suggested that higher protein density alongside higher plant-to-animal protein ratios may also be important for CVD risk reduction. Finally, substitution analyses indicated that the observed inverse associations are likely driven by the replacement of mainly red and processed meat, as well as some poultry, with several plant protein sources, but particularly with nuts.
We are not aware of other studies examining plant-to-animal protein ratios with CVD risk. However, a previous study found that a higher animal-to-plant protein ratio was associated with a higher waist circumference and fasting blood glucose in the Adventist Health Study-2 Calibration Study [25], which can impact CVD risk. Other research on plant-to-animal protein ratios in the literature has been related to modeling nutrient adequacy and health risks. One study estimated that diets with 25%–70% of protein coming from plants were compatible with minimal health risk and that nutrient adequacy was only compromised when plant protein increased above 80% [26], highlighting that fortified foods and supplements would be needed above this cutoff, particularly for nutrients such as calcium, eicosapentaenoic acid, docosahexaenoic acid, vitamin B12, and iron, among others. We examined the intake of these nutrients according to deciles of the ratio and found that although some nutrients were lower in decile 10 (∼76% plant protein), most nutrients met the recommended dietary allowance (RDA). These nutrient intakes did, however, include fortified foods and supplements.
When examining dose–response relationships, risk reductions for CVD start to plateau around ∼0.5, however, for CAD, risk continues to decrease at high ratios (≤1.84). In contrast, risk reductions for stroke plateau at ∼0.5. Analyses of plant protein have shown a reduced risk of CAD and cardiovascular mortality [27], particularly in substitution analyses [28], whereas the results with stroke have been less consistent [29,30]. Studies have also shown that some animal protein sources may be beneficial for stroke prevention, such as fish [31]. In our substitution analyses, replacing red and processed meat with most plant protein sources was associated with a lower risk of stroke, however, substituting poultry and dairy with plant protein did not demonstrate a significant association with stroke. In the current analysis, plant protein included refined grains (ranging from 23% to 36% of protein intake across cohort years), which may negatively impact stroke risk compared with whole grains, nuts, and legumes; however, the findings remained similar after adjusting for diet quality and when removing refined grains from the ratio. Further possible explanations for the discrepant findings between CAD and stroke may be that plant protein has a larger impact on cholesterol levels than blood pressure, which is a greater risk factor for stroke [32], and that there were fewer stroke cases than CAD cases. In addition, the stroke subtype may be important. Previous research has shown that red meat has been associated with a higher risk of ischemic stroke, but a lower risk of hemorrhagic stroke [33], whereas the opposite has been shown in vegetarians (associated with lower ischemic stroke and higher hemorrhagic stroke) [34]. In our study, neither ischemic stroke nor hemorrhagic stroke findings were significant. Further research is needed to explore the role of different protein sources and the optimal plant-to-animal protein ratio in the diet for stroke prevention, including stroke subtypes.
We also observed that higher ratios (>0.50) combined with higher protein density (20.8%) presented the greatest cardiovascular benefits, highlighting that more protein should be consumed when increasing plant protein intake. These findings are in line with previous studies that have shown that lower carbohydrate diets focused on plant sources of protein and fat (rather than animal sources) are associated with lower mortality risk [35]. Protein density may also have a connection with diet quality. We examined higher diet quality with increasing plant protein intake in these cohorts; however, we did not observe a significant interaction between protein ratios and overall diet quality in relation to CVD risk. Achieving this protein density with higher ratios will require individuals to focus on more protein-dense plant sources such as legumes, however, further investigation of protein sources combined with protein density and CVD risk is needed.
There are several mechanisms in which a higher plant-to-animal protein ratio in the diet may be associated with lower CVD risk. Plant protein sources, as a protein package, are lower in saturated fat and higher in unsaturated fat, carbohydrates, and fiber than animal protein, all of which may provide cardiovascular benefits [8]. In our sensitivity analyses, we further adjusted for these dietary components and although the findings were slightly attenuated, the results were still significant, suggesting other factors may be contributing to the lower CVD risk. Other factors may include the amino acid profiles of plant and animal proteins [36]. For example, the main protein fraction, 7S globulin, found in legumes has been shown to inhibit hepatic apolipoprotein B synthesis [37]. Plant protein is also higher in arginine, which may be beneficial for blood pressure [38], and lower in branched-chain amino acids (BCAAs), which have been associated with higher CVD risk [39]. Furthermore, other nonindispensable amino acids higher in plant protein, such as glycine, may also play a role in preventing CVD risk [40]. Other possible mechanisms include less heme iron and trimethylamine N-oxide production and higher bioactive and polyphenol content from plant protein [8,[41], [42], [43]]. Metabolomics analyses of animal and plant protein-rich diets have also suggested several of these pathways as plausible mediators in the associations between lower CVD risk with increasing plant protein intake, including BCAAs, trimethylamine N-oxide, acylcarnitines, and glycine, among others [44]. Lastly, the relationship between a higher plant-to-animal protein ratio and CVD risk may also be mediated through cholesterol levels, blood pressure, and diabetes incidence. Our analyses after adjustment for these mediators (including medications) consistently attenuated the effect estimates by ∼5%, nevertheless, the findings remained significant. Moreover, in our subgroup analyses, the lower CVD risk with a higher plant-to-animal protein ratio was stronger in those with hypercholesterolemia, further highlighting higher plant protein intake may be important for improving lipoprotein profiles.
Implications
This study provides additional evidence to support replacing animal protein sources, particularly red and processed meat, with plant protein sources in the diet to reduce CVD risk, mainly CAD. We were unable to determine the optimal ratio for CVD risk reduction in the current analysis due to the potentially different dose–response relationships for CAD and stroke. However, risk for both CVD and stroke started to plateau or potentially increase at ∼0.5, emphasizing that a ratio of at least 1:2 may provide cardiovascular benefits compared with ratios with more animal protein, although this ratio may be much higher for CAD prevention (a ratio of 0.76 or higher).
Another implication from our analyses is related to plant protein sources. Our primary analysis included plant protein from all sources, including refined grains, whole grains, nuts, and legumes, therefore, we included these sources in the substitution analyses. However, even though replacing red meat and poultry with less healthy plant sources showed a lower risk of CVD, from a public health and guideline recommendation, animal protein, particularly red and processed meat, should optimally be replaced with legumes and nuts, as shown in a recent systematic review and meta-analysis [45] and as recommended by dietary and clinical practice guidelines [[2], [3], [4]]. This caveat is particularly important given several analyses have shown that saturated fat in the diet should be replaced with PUFAs, MUFAs, and carbohydrates from whole plant sources (i.e., complex carbohydrates) and not refined carbohydrates to lower CVD risk [46]. Furthermore, differences in energy, protein density, and nutrient profile (including potential shortfall nutrients) must be kept in mind when replacing animal with plant protein sources [47]. For example, although most nutrients met the RDA across the deciles, there were decreasing trends in nutrients such as calcium, ω-3 fatty acids, zinc, vitamin B12, and vitamin D, with increasing plant protein intake. These findings highlight the importance of plant protein sources and the need to focus on higher quality foods (legumes and nuts over refined grains), as nutrient and protein quality would improve with these higher quality sources [48,49]. Fortified foods and supplements may also be needed, particularly when plant protein intake is >80% [26], or if individuals are not meeting their calorie needs. Our finding that replacing animal protein sources with nuts would be the most beneficial choice for CVD risk reduction additionally raises an important question of nutrient adequacy and energy intake, particularly considering the animal protein source to be replaced. Nuts are recommended by several dietary and clinical practice guidelines globally to lower low-density lipoprotein cholesterol, improve overall lipoprotein profile, and decrease overall CVD risk [50,51], and would benefit from being part of the protein package alongside other protein sources, as depicted in the plate from Canada’s Food Guide [4].
Strengths and limitations
Strengths of our study include the large sample size, long duration of follow-up, and repeated measures of diet that allowed us to assess cumulative ratios over time. We also comprehensively assessed our findings through several sensitivity, subgroup, and substitution analyses. Limitations of our study include the number of participants excluded from the analysis, reliance on self-reported dietary intake, and the observational study design; therefore, measurement errors and residual confounding cannot be ruled out. In addition, the plant protein sources had lower validity coefficients which may reflect more measurement errors and greater attenuation of the associations for these protein sources. For our substitution analysis of plant and animal protein sources, we analyzed the substitutions in servings/d for ease of interpretation, however, modeling substitutions of servings of foods with different energy content and adjusting for total energy intake, introduces residual energy, making the substitution analysis not entirely isocaloric [52]. Finally, our study included health professionals who are mostly White; therefore, the generalizability of our findings requires confirmation in more diverse populations.
In conclusion, these findings suggest that a higher plant-to-animal protein ratio is associated with a lower risk of CVD, particularly CAD. The findings also indicate that a ratio of ≥0.5 should be considered for CVD risk reduction, although this ratio may be higher for CAD prevention (≥0.76), and that plant protein consumption combined with higher protein density may provide additional cardiovascular benefit. Most of this benefit appears to be from replacing red and processed meat with several plant protein sources. Overall, further research is needed to clarify the optimal ratio in the diet for the prevention of CVD. Examining the ratio in populations who consume more healthy plant protein sources, as well as more plant protein in general, will help us better understand the optimal ratio in the diet for CAD and stroke prevention.
Acknowledgments
We thank the participants and staff of the Nurses’ Health Studies I and II and the Health Professionals Follow-up Study for their participation and valuable contributions. Aspects of this work were presented in abstract form at the American Society of Nutrition Annual Conference in Boston, MA (July 22–25, 2023).
Author contributions
The authors’ contributions were as follows – AJG, FBH: had full access to all of the data in the study and took responsibility for the integrity of the data and the accuracy of the data analysis; AJG, FBH, EBR, WCW, JEM: concept and design; AJG, FW, AJT, EBR, WCW, QS, JEM, FBH: acquisition, analysis, or interpretation of data. AJG: drafting of the manuscript; and all authors critically revised of the manuscript for important intellectual content.
Conflict of interest
AJG has received travel support and/or honoraria from the Lawson Centre Nutrition Digitial Series, Vinasoy and the British Nutrition Society. DJAJ has received research grants from Saskatchewan & Alberta Pulse Growers Associations, the Agricultural Bioproducts Innovation Program through the Pulse Re-search Network, the Advanced Foods and Material Network, Loblaw Companies Ltd., Unilever Canada and Netherlands, Barilla, the Almond Board of California, Agriculture and Agri-food Canada, Pulse Canada, Kellogg’s Company, Canada, Quaker Oats, Canada, Procter & Gamble Technical Centre Ltd., Bayer Consumer Care, Springfield, NJ, Pepsi/Quaker, International Nut & Dried Fruit Council (INC), Soy Foods Association of North America, the Coca-Cola Company (investigator initiated, unrestricted grant), Solae, Haine Celestial, the Sanitarium Company, Orafti, the International Tree Nut Council Nutrition Research and Education Foundation, the Peanut Institute, Soy Nutrition Institute (SNI), the Canola and Flax Councils of Canada, the Calorie Control Council, the Canadian Institutes of Health Research (CIHR), the Canada Foundation for Innovation (CFI), and the Ontario Research Fund (ORF). He has received in-kind supplies for trials as research support from the Almond Board of California, Walnut Council of California, the Peanut Institute, Barilla, Unilever, Unico, Primo, Loblaw Companies, Quaker (Pepsico), Pristine Gourmet, Bunge Limited, Kellogg Canada, WhiteWave Foods. He has been on the speaker’s panel, served on the scientific advisory board and/or received travel support and/or honoraria from the Lawson Centre Nutrition Digital Series, Nutritional Fundamentals for Health (NFH)-Nutramedica, Saint Barnabas Medical Center, The University of Chicago, 2020 China Glycemic Index (GI) International Conference, Atlantic Pain Conference, Academy of Life Long Learning, the Almond Board of California, Canadian Agriculture Policy Institute, Loblaw Companies Ltd, the Griffin Hospital (for the development of the NuVal scoring system), the Coca-Cola Company, Epicure, Danone, Diet Quality Photo Navigation (DQPN), Better Therapeutics (FareWell), Verywell, True Health Initiative (THI), Heali AI Corp, Institute of Food Technologists (IFT), Soy Nutrition Institute (SNI), Herbalife Nutrition Institute (HNI), Saskatchewan & Alberta Pulse Growers Associations, Sanitarium Company, Orafti, the International Tree Nut Council Nutrition Research and Education Foundation, the Peanut Institute, Herbalife International, Pacific Health Laboratories, Barilla, Metagenics, Bayer Consumer Care, Unilever Canada and Netherlands, Solae, Kellogg, Quaker Oats, Procter & Gamble, Abbott Laboratories, Dean Foods, the California Strawberry Commission, Haine Celestial, PepsiCo, the Alpro Foundation, Pioneer Hi-Bred International, DuPont Nutrition and Health, Spherix Consulting and WhiteWave Foods, the Advanced Foods and Material Network, the Canola and Flax Councils of Canada, Agri-Culture and Agri-Food Canada, the Canadian Agri-Food Policy Institute, Pulse Canada, the Soy Foods Association of North America, the Nutrition Foundation of Italy (NFI), Nutra-Source Diagnostics, the McDougall Program, the Toronto Knowledge Translation Group (St. Michael’s Hospital), the Canadian College of Naturopathic Medicine, The Hospital for Sick Children, the Canadian Nutrition Society (CNS), the American Society of Nutrition (ASN), Arizona State University, Paolo Sorbini Foundation, and the Institute of Nutrition, Metabolism and Diabetes. He received an honorarium from the United States Department of Agriculture to present the 2013 W.O. Atwater Memorial Lecture. He received the 2013 Award for Excellence in Research from the International Nut and Dried Fruit Council. He received funding and travel support from the Canadian Society of Endocrinology and Metabolism to produce mini cases for the Canadian Diabetes Association (CDA). He is a member of the International Carbo-hydrate Quality Consortium (ICQC). His wife, Alexandra L Jenkins, is a director and partner of INQUIS Clinical Research for the Food Industry, his 2 daughters, Wendy Jenkins and Amy Jenkins, have published a vegetarian book that promotes the use of the foods described here, The Portfolio Diet for Cardiovascular Risk Reduction (Academic Press/Elsevier 2020 ISBN:978-0-12-810510-8) and his sister, Caroline Brydson, received funding through a grant from the St. Michael’s Hospital Foundation to develop a cookbook for one of his studies. He is also a vegan. KJM has received funding from the United States Highbush Blueberry Council. All the other authors report no conflicts of interest.
Funding
AJG was supported by a Canadian Institutes of Health Research (CIHR) Postdoctoral Fellowship and Toronto 3D Postdoctoral Fellowship Top-Up Award. AJT was supported by a CIHR Postdoctoral Fellowship. FW is supported by an American Heart Association Postdoctoral Fellowship. The Nurses’ Health Studies and Health Professional Follow-up Studies are supported by National Institutes of Health grants UM1 CA186107, R01 CA49449, R01 HL034594, U01 HL145386, R01 HL088521, U01 CA176726, R01 CA49449, U01 CA167552, R01 HL60712, and R01 HL35464.
Data availability
Because of participant confidentiality and privacy concerns, data cannot be shared publicly, and requests to access Nurses’ Health Studies/Health Professionals Follow-up Study data must be submitted in writing. Further information including the procedures to obtain and access data from the Nurses’ Health Studies and Health Professionals’ Follow-up Study is described at https://www.nurseshealthstudy.org/ researchers (contact e-mail: nhsaccess@channing.harvard.edu) and https://sites.sph.harvard.edu/hpfs/for-collaborators/.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ajcnut.2024.09.006.
Contributor Information
Andrea J Glenn, Email: a.glenn@nyu.edu.
Frank B Hu, Email: fhu@hsph.harvard.edu.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Mohebi R., Chen C., Ibrahim N.E., McCarthy C.P., Gaggin H.K., Singer D.E., et al. Cardiovascular disease projections in the United States based on the 2020 census estimates. J. Am. Coll. Cardiol. 2022;80(6):565–578. doi: 10.1016/j.jacc.2022.05.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pearson G.J., Thanassoulis G., Anderson T.J., Barry A.R., Couture P., Dayan N., et al. 2021 Canadian Cardiovascular Society Guidelines for the management of dyslipidemia for the prevention of cardiovascular disease in adults. Can. J. Cardiol. 2021;37(8):1129–1150. doi: 10.1016/j.cjca.2021.03.016. [DOI] [PubMed] [Google Scholar]
- 3.Lichtenstein A.H., Appel L.J., Vadiveloo M., Hu F.B., Kris-Etherton P.M., Rebholz C.M., et al. 2021 dietary guidance to improve cardiovascular health: a scientific statement from the American Heart Association. Circulation. 2021;144(23):e472–4e87. doi: 10.1161/CIR.0000000000001031. [DOI] [PubMed] [Google Scholar]
- 4.Health Canada Canada’s Food Guide [Internet] 2019 https://food-guide.canada.ca/en/ [date updated September 30 2024; date cited February 27, 2024]. Available from: [Google Scholar]
- 5.Willett W., Rockström J., Loken B., Springmann M., Lang T., Vermeulen S., et al. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. Lancet. 2019;393(10170):447–492. doi: 10.1016/S0140-6736(18)31788-4. [DOI] [PubMed] [Google Scholar]
- 6.Shan Z., Rehm C.D., Rogers G., Ruan M., Wang D.D., Hu F.B., et al. Trends in dietary carbohydrate, protein, and fat intake and diet quality among US adults, 1999–2016. JAMA. 2019;322(12):1178–1187. doi: 10.1001/jama.2019.13771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Smit E., Nieto F.J., Crespo C.J., Mitchell P. Estimates of animal and plant protein intake in US adults: results from the Third National Health and Nutrition Examination Survey, 1988–1991. J. Am. Diet. Assoc. 1999;99(7):813–820. doi: 10.1016/S0002-8223(99)00193-5. [DOI] [PubMed] [Google Scholar]
- 8.Richter C.K., Skulas-Ray A.C., Champagne C.M., Kris-Etherton P.M. Plant protein and animal proteins: do they differentially affect cardiovascular disease risk? Adv. Nutr. 2015;6(6):712–728. doi: 10.3945/an.115.009654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhang X., Sergin I., Evans T.D., Jeong S.-J., Rodriguez-Velez A., Kapoor D., et al. High-protein diets increase cardiovascular risk by activating macrophage mTOR to suppress mitophagy. Nat. Metabol. 2020;2(1):110–125. doi: 10.1038/s42255-019-0162-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Colditz G.A., Manson J.E., Hankinson S.E. The Nurses’ Health Study: 20-year contribution to the understanding of health among women. J. Womens Health. 1997;6(1):49–62. doi: 10.1089/jwh.1997.6.49. [DOI] [PubMed] [Google Scholar]
- 11.Rimm E.B., Giovannucci E.L., Willett W.C., Colditz G.A., Ascherio A., Rosner B., et al. Prospective study of alcohol consumption and risk of coronary disease in men. Lancet. 1991;338(8765):464–468. doi: 10.1016/0140-6736(91)90542-w. [DOI] [PubMed] [Google Scholar]
- 12.Willett W.C., Sampson L., Stampfer M.J., Rosner B., Bain C., Witschi J., et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am. J. Epidemiol. 1985;122(1):51–65. doi: 10.1093/oxfordjournals.aje.a114086. [DOI] [PubMed] [Google Scholar]
- 13.Rimm E.B., Giovannucci E.L., Stampfer M.J., Colditz G.A., Litin L.B., Willett W.C. Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am. J. Epidemiol. 1992;135(10):1114–1126. doi: 10.1093/oxfordjournals.aje.a116211. [discussion: 27–36] [DOI] [PubMed] [Google Scholar]
- 14.Yuan C., Spiegelman D., Rimm E.B., Rosner B.A., Stampfer M.J., Barnett J.B., et al. Relative validity of nutrient intakes assessed by questionnaire, 24-hour recalls, and diet records as compared with urinary recovery and plasma concentration biomarkers: findings for women. Am. J. Epidemiol. 2018;187(5):1051–1063. doi: 10.1093/aje/kwx328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Harvard Nutrient Database. https://hsph.harvard.edu/nutrition-questionnaire-service-center/nutrient-tables-download-page [Internet]. [date modified January 1, 2022; date cited February 27, 2024]. Available from:
- 16.Yuan C., Spiegelman D., Rimm E.B., Rosner B.A., Stampfer M.J., Barnett J.B., et al. Validity of a dietary questionnaire assessed by comparison with multiple weighed dietary records or 24-hour recalls. Am. J. Epidemiol. 2017;185(7):570–584. doi: 10.1093/aje/kww104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hu F.B., Rimm E., Smith-Warner S.A., Feskanich D., Stampfer M.J., Ascherio A., et al. Reproducibility and validity of dietary patterns assessed with a food-frequency questionnaire. Am. J. Clin. Nutr. 1999;69(2):243–249. doi: 10.1093/ajcn/69.2.243. [DOI] [PubMed] [Google Scholar]
- 18.Salvini S., Hunter D.J., Sampson L., Stampfer M.J., Colditz G.A., Rosner B., et al. Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption. Int. J. Epidemiol. 1989;18(4):858–867. doi: 10.1093/ije/18.4.858. [DOI] [PubMed] [Google Scholar]
- 19.Hu F.B., Stampfer M.J., Rimm E., Ascherio A., Rosner B.A., Spiegelman D., et al. Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Am. J. Epidemiol. 1999;149(6):531–540. doi: 10.1093/oxfordjournals.aje.a009849. [DOI] [PubMed] [Google Scholar]
- 20.Mendis S., Thygesen K., Kuulasmaa K., Giampaoli S., Mähönen M., Ngu Blackett K., et al. World Health Organization definition of myocardial infarction: 2008–09 revision. Int. J. Epidemiol. 2011;40(1):139–146. doi: 10.1093/ije/dyq165. [DOI] [PubMed] [Google Scholar]
- 21.Walker A.E., Robins M., Weinfeld F.D. The National Survey of Stroke. Clinical findings. Stroke. 1981;12(2 Pt 2 Suppl 1):I13–I44. [PubMed] [Google Scholar]
- 22.Stampfer M.J., Willett W.C., Speizer F.E., Dysert D.C., Lipnick R., Rosner B., et al. Test of the National Death Index. Am. J. Epidemiol. 1984;119(5):837–839. doi: 10.1093/oxfordjournals.aje.a113804. [DOI] [PubMed] [Google Scholar]
- 23.Chiuve S.E., Fung T.T., Rimm E.B., Hu F.B., McCullough M.L., Wang M., et al. Alternative dietary indices both strongly predict risk of chronic disease. J. Nutr. 2012;142(6):1009–1018. doi: 10.3945/jn.111.157222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Song M., Giovannucci E. Substitution analysis in nutritional epidemiology: proceed with caution. Eur. J. Epidemiol. 2018;33(2):137–140. doi: 10.1007/s10654-018-0371-2. [DOI] [PubMed] [Google Scholar]
- 25.Azemati B., Rajaram S., Jaceldo-Siegl K., Haddad E.H., Shavlik D., Fraser G.E. Dietary animal to plant protein ratio is associated with risk factors of metabolic syndrome in participants of the AHS-2 Calibration Study. Nutrients. 2021;13(12):4296. doi: 10.3390/nu13124296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Fouillet H., Dussiot A., Perraud E., Wang J., Huneau J.F., Kesse-Guyot E., et al. Plant to animal protein ratio in the diet: nutrient adequacy, long-term health and environmental pressure. Front. Nutr. 2023;10 doi: 10.3389/fnut.2023.1178121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tharrey M., Mariotti F., Mashchak A., Barbillon P., Delattre M., Fraser G.E. Patterns of plant and animal protein intake are strongly associated with cardiovascular mortality: the Adventist Health Study-2 cohort. Int. J. Epidemiol. 2018;47(5):1603–1612. doi: 10.1093/ije/dyy030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Al-Shaar L., Satija A., Wang D.D., Rimm E.B., Smith-Warner S.A., Stampfer M.J., et al. Red meat intake and risk of coronary heart disease among US men: prospective cohort study. BMJ. 2020;371 doi: 10.1136/bmj.m4141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Preis S.R., Stampfer M.J., Spiegelman D., Willett W.C., Rimm E.B. Lack of association between dietary protein intake and risk of stroke among middle-aged men. Am. J. Clin. Nutr. 2010;91(1):39–45. doi: 10.3945/ajcn.2009.28060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bernstein A.M., Pan A., Rexrode K.M., Stampfer M., Hu F.B., Mozaffarian D., et al. Dietary protein sources and the risk of stroke in men and women. Stroke. 2012;43(3):637–644. doi: 10.1161/STROKEAHA.111.633404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zhang Z., Xu G., Yang F., Zhu W., Liu X. Quantitative analysis of dietary protein intake and stroke risk. Neurology. 2014;83(1):19–25. doi: 10.1212/WNL.0000000000000551. [DOI] [PubMed] [Google Scholar]
- 32.O'Donnell M.J., Xavier D., Liu L., Zhang H., Chin S.L., Rao-Melacini P., et al. Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet. 2010;376(9735):112–123. doi: 10.1016/S0140-6736(10)60834-3. [DOI] [PubMed] [Google Scholar]
- 33.Collaborators G. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2024;403(10440):2162–2203. doi: 10.1016/S0140-6736(24)00933-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Tong T.Y.N., Appleby P.N., Bradbury K.E., Perez-Cornago A., Travis R.C., Clarke R., et al. Risks of ischaemic heart disease and stroke in meat eaters, fish eaters, and vegetarians over 18 years of follow-up: results from the prospective EPIC-Oxford study. BMJ. 2019;366:l4897. doi: 10.1136/bmj.l4897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Seidelmann S.B., Claggett B., Cheng S., Henglin M., Shah A., Steffen L.M., et al. Dietary carbohydrate intake and mortality: a prospective cohort study and meta-analysis. Lancet Public Health. 2018;3(9):e419–e428. doi: 10.1016/S2468-2667(18)30135-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Mariotti F. Animal and plant protein sources and cardiometabolic health. Adv. Nutr. 2019;10:S351–S366. doi: 10.1093/advances/nmy110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lovati M.R., Manzoni C., Gianazza E., Arnoldi A., Kurowska E., Carroll K.K., et al. Soy protein peptides regulate cholesterol homeostasis in Hep G2 cells. J. Nutr. 2000;130(10):2543–2549. doi: 10.1093/jn/130.10.2543. [DOI] [PubMed] [Google Scholar]
- 38.Appel L.J. The effects of protein intake on blood pressure and cardiovascular disease. Curr. Opin. Lipidol. 2003;14(1):55–59. doi: 10.1097/00041433-200302000-00010. [DOI] [PubMed] [Google Scholar]
- 39.Yang R., Dong J., Zhao H., Li H., Guo H., Wang S., et al. Association of branched-chain amino acids with carotid intima-media thickness and coronary artery disease risk factors. PLOS ONE. 2014;9(6) doi: 10.1371/journal.pone.0099598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Tharrey M., Mariotti F., Mashchak A., Barbillon P., Delattre M., Huneau J.-F., et al. Patterns of amino acid intake are strongly associated with cardiovascular mortality, independently of the sources of protein. Int. J. Epidemiol. 2020;49(1):312–321. doi: 10.1093/ije/dyz194. [DOI] [PubMed] [Google Scholar]
- 41.Fang X., An P., Wang H., Wang X., Shen X., Li X., et al. Dietary intake of heme iron and risk of cardiovascular disease: a dose–response meta-analysis of prospective cohort studies. Nutr. Metab. Cardiovasc. Dis. 2015;25(1):24–35. doi: 10.1016/j.numecd.2014.09.002. [DOI] [PubMed] [Google Scholar]
- 42.Heianza Y., Ma W., DiDonato J.A., Sun Q., Rimm E.B., Hu F.B., et al. Long-term changes in gut microbial metabolite trimethylamine n-oxide and coronary heart disease risk. J. Am. Coll. Cardiol. 2020;75(7):763–772. doi: 10.1016/j.jacc.2019.11.060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wang F., Glenn A.J., Tessier A.-J., Mei Z., Haslam D.E., Guasch-Ferré M., et al. Integration of epidemiological and blood biomarker analysis links haem iron intake to increased type 2 diabetes risk. Nat. Metabolism. 2024;6:1807–1818. doi: 10.1038/s42255-024-01109-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Lépine G., Fouillet H., Rémond D., Huneau J.F., Mariotti F., Polakof S. A scoping review: metabolomics signatures associated with animal and plant protein intake and their potential relation with cardiometabolic risk. Adv. Nutr. 2021;12(6):2112–2131. doi: 10.1093/advances/nmab073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Neuenschwander M., Stadelmaier J., Eble J., Grummich K., Szczerba E., Kiesswetter E., et al. Substitution of animal-based with plant-based foods on cardiometabolic health and all-cause mortality: a systematic review and meta-analysis of prospective studies. BMC Med. 2023;21(1):404. doi: 10.1186/s12916-023-03093-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.WHO Guidelines Approved by the Guidelines Review Committee . World Health Organization; Geneva: 2023. Saturated Fatty Acid and Trans-Fatty Acid Intake for Adults and Children: WHO Guideline. [PubMed] [Google Scholar]
- 47.Messina M., Duncan A.M., Glenn A.J., Mariotti F. Perspective: plant-based meat alternatives can help facilitate and maintain a lower animal to plant protein intake ratio. Adv. Nutr. 2023;14(3):392–405. doi: 10.1016/j.advnut.2023.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Marinangeli C.P.F., Miller K., Fulgoni V.L. Effect of increasing plant protein intake on protein quality and nutrient intake of US adults. Appl. Physiol. Nutr. Metab. 2023;48(1):49–61. doi: 10.1139/apnm-2022-0054. [DOI] [PubMed] [Google Scholar]
- 49.Mitchell D.C., Marinangeli C.P.F., Pigat S., Bompola F., Campbell J., Pan Y., et al. Pulse intake improves nutrient density among US adult consumers. Nutrients. 2021;13(8):2668. doi: 10.3390/nu13082668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Glenn A.J., Aune D., Freisling H., Mohammadifard N., Kendall C.W.C., Salas-Salvadó J., et al. Nuts and cardiovascular disease outcomes: a review of the evidence and future directions. Nutrients. 2023;15(4):911. doi: 10.3390/nu15040911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Guasch-Ferré M., Tessier A.J., Petersen K.S., Sapp P.A., Tapsell L.C., Salas-Salvadó J., et al. Effects of nut consumption on blood lipids and lipoproteins: a comprehensive literature update. Nutrients. 2023;15(3):596. doi: 10.3390/nu15030596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Ibsen D.B., Laursen A.S.D., Würtz A.M.L., Dahm C.C., Rimm E.B., Parner E.T., et al. Food substitution models for nutritional epidemiology. Am. J. Clin. Nutr. 2021;113(2):294–303. doi: 10.1093/ajcn/nqaa315. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Because of participant confidentiality and privacy concerns, data cannot be shared publicly, and requests to access Nurses’ Health Studies/Health Professionals Follow-up Study data must be submitted in writing. Further information including the procedures to obtain and access data from the Nurses’ Health Studies and Health Professionals’ Follow-up Study is described at https://www.nurseshealthstudy.org/ researchers (contact e-mail: nhsaccess@channing.harvard.edu) and https://sites.sph.harvard.edu/hpfs/for-collaborators/.