Abstract
IMPORTANCE Although emphasis has recently been placed on the importance of high-protein diets to overall health, a comprehensive analysis of long-term cause-specific mortality in association with the intake of plant protein and animal protein has not been reported. OBJECTIVE To examine the associations between overall mortality and cause-specific mortality and plant protein intake. DESIGN, SETTING, AND PARTICIPANTS This prospective cohort study analyzed data from 416 104 men and women in the US National Institutes of Health-AARP Diet and Health Study from 1995 to 2011. Data were analyzed from October 2018 through April 2020. EXPOSURES Validated baseline food frequency questionnaire dietary information, including intake of plant protein and animal protein. MAIN OUTCOMES AND MEASURES Hazard ratios and 16-year absolute risk differences for overall mortality and cause-specific mortality. RESULTS The final analytic cohort included 237 036 men (57%) and 179 068 women. Their overall median (SD) ages were 62.2 (5.4) years for men and 62.0 (5.4) years for women. Based on 6 009 748 person-years of observation, 77 614 deaths (18.7%; 49 297 men and 28 317 women) were analyzed. Adjusting for several important clinical and other risk factors, greater dietary plant protein intake was associated with reduced overall mortality in both sexes (hazard ratio per 1 SD was 0.95 [95% CI, 0.94-0.97] for men and 0.95 [95% CI, 0.93-0.96] for women; adjusted absolute risk difference per 1 SD was -0.36% [95% CI, -0.48% to -0.25%] for men and -0.33% [95% CI, -0.48% to -0.21%] for women; hazard ratio per 10 g/1000 kcal was 0.88 [95% CI, 0.84-0.91] for men and 0.86 [95% CI, 0.82-0.90] for women; adjusted absolute risk difference per 10 g/1000 kcal was -0.95% [95% CI, -1.3% to -0.68%] for men and -0.86% [95% CI, -1.3% to -0.55%] for women; all P < .001). The association between plant protein intake and overall mortality was similar across the subgroups of smoking status, diabetes, fruit consumption, vitamin supplement use, and self-reported health status. Replacement of 3% energy from animal protein with plant protein was inversely associated with overall mortality (risk decreased 10% in both men and women) and cardiovascular disease mortality (11% lower risk in men and 12% lower risk in women). In particular, the lower overall mortality was attributable primarily to substitution of plant protein for egg protein (24% lower risk in men and 21% lower risk in women) and red meat protein (13% lower risk in men and 15% lower risk in women). CONCLUSIONS AND RELEVANCE In this large prospective cohort, higher plant protein intake was associated with small reductions in risk of overall and cardiovascular disease mortality. Our findings provide evidence that dietary modification in choice of protein sources may influence health and longevity.
Generated Summary
This prospective cohort study analyzed data from 416,104 men and women in the US National Institutes of Health-AARP Diet and Health Study from 1995 to 2011 to examine the associations between overall mortality and cause-specific mortality and plant protein intake. Data were analyzed from October 2018 through April 2020. The final analytic cohort included 237,036 men (57%) and 179,068 women. Their overall median (SD) ages were 62.2 (5.4) years for men and 62.0 (5.4) years for women. Greater dietary plant protein intake was associated with reduced overall mortality in both sexes. Adjusting for several important clinical and other risk factors, greater dietary plant protein intake was associated with reduced overall mortality in both sexes. Replacement of 3% energy from animal protein with plant protein was inversely associated with overall mortality and cardiovascular disease mortality. The study examined associations of plant protein intake with overall mortality and specific causes of death, including cancer, cardiovascular disease (CVD), heart disease, stroke, respiratory disease, infection, injury/accident, and other causes. Dietary information, including intake of plant protein and animal protein, was obtained from validated baseline food frequency questionnaires.
Key Findings & Statistics
- The final analytic cohort included 237,036 men (57%) and 179,068 women.
- Their overall median (SD) ages were 62.2 (5.4) years for men and 62.0 (5.4) years for women.
- Based on 6,009,748 person-years of observation, 77,614 deaths (18.7%; 49,297 men and 28,317 women) were analyzed.
- Greater dietary plant protein intake was associated with reduced overall mortality in both sexes (hazard ratio per 1 SD was 0.95 [95% CI, 0.94-0.97] for men and 0.95 [95% CI, 0.93-0.96] for women; adjusted absolute risk difference per 1 SD was -0.36% [-0.48% to -0.25%] for men and -0.33% [-0.48% to -0.21%] for women; hazard ratio per 10 g/1000 kcal was 0.88 [0.84-0.91] for men and 0.86 [0.82-0.90] for women; adjusted absolute risk difference per 10 g/1000 kcal was -0.95% [-1.3% to -0.68%] for men and -0.86% [-1.3% to -0.55%] for women; all P < .001).
- Replacement of 3% energy from animal protein with plant protein was inversely associated with overall mortality (risk decreased 10% in both men and women) and cardiovascular disease mortality (11% lower risk in men and 12% lower risk in women).
- In men, the HRs were 0.95 (95% CI, 0.94-0.97) per 1 SD and 0.88 (95% CI, 0.84-0.91) per 10 g/1000 kcal increment, with adjusted ARDs (at 16 years of follow-up) of -0.36% (-0.48% to -0.25%) per 1 SD and -0.95% (-1.3% to -0.68%) per 10 g/1000 kcal increment (P < .001).
- Corresponding adjusted HRs in women were 0.95 (95% CI, 0.93-0.96) per 1 SD and 0.86 (95% CI, 0.82-0.90) per 10 g/1000 kcal increment, with adjusted ARDs of -0.33% (-0.48% to -0.21%) per 1 SD and -0.86% (-1.3% to -0.55%) per 10 g/1000 kcal increment (P < .001).
- In men, the HRs for CVD were 0.88 (0.86 to 0.90) with a p-value of <.001. In women the HRs were 0.93 (0.90 to 0.97) with a p-value of <.001.
- In men, the HRs for stroke were 0.88 (0.82 to 0.94) with a p-value of <.001. In women the HRs were 0.88 (0.81 to 0.96) with a p-value of .003.
- The median percentage of daily energy intake from total protein was 15.3%, and the median percentage of daily dietary protein intake was 40% from plant sources and 60% from animal protein sources (including 19% from dairy foods).
- The median plant protein intake was 26.9 g/d (14.4 g/1000 kcal per day) for men and 21.6 g/d (14.9 g/1000 kcal per day) for women.
- The adjusted ARDs at 16 years of follow-up for CVD mortality were -0.13% (-0.22% to -0.05%) per 1-SD intake increment and -0.35% per (-0.59% to -0.14%) per 10 g/1000 kcal increment in men, and -0.13% (-0.22% to -0.053%) per 1 SD and -0.35% (-0.57% to -0.15%) per 10 g/1000 kcal increment in women.
- Plant protein intake was inversely associated with mortality from all CVD, stroke, and “other causes combined,” representing risk reductions of 5% to 12% per 1-SD intake increment and 12% to 30% per 10 g/1000 kcal intake increment in men, and 7% to 12% per 1 SD and 17% to 29% per 10 g/1000 kcal in women (all P < .003).
- Significant association with lower overall mortality (men: HR, 0.90; 95% CI, 0.88-0.93; women: HR, 0.90; 95% CI, 0.87-0.93; all P < .001), and CVD mortality (men: HR, 0.89; 95% CI, 0.85-0.94; women: HR, 0.88; 95% CI, 0.82-0.94; P < .001).
Other Important Findings
- The association between plant protein intake and overall mortality was similar across subgroups of smoking status, diabetes, fruit consumption, vitamin supplement use, and self-reported health status.
- The lower overall mortality was attributable primarily to substitution of plant protein for egg protein and red meat protein.
- Plant protein intake was significantly inversely associated with age-adjusted mortality from all causes in both men and women.
- Additional adjustment for consumption of sugar-sweetened beverages and total carbohydrates as well as for median household income did not materially change the association estimates.
- For men, the HRs for heart disease were 0.97 (0.94 to 0.99) with a p-value of .02. For women, the HRs for heart disease were 0.96 (0.92 to 1.00) with a p-value of .04.
- Examining the substitution of 3% energy from plant protein for specific sources of animal protein revealed that the mortality reduction appeared strongest with substitution for egg protein, with 15% to 39% lower risk in men and 17% to 34% lower risk in women for overall, cancer, CVD, heart disease, and respiratory disease mortality.
- Substitution of 3% energy from plant protein for red meat protein intake showed stronger mortality reductions of 11% to 21% in men and 11% to 25% in women for overall, CVD, heart disease, and stroke mortality in both sexes, and cancer and respiratory disease mortality in women.
- A similar change from dairy protein to plant protein was associated with lower overall and CVD mortality in men and women, and heart disease and stroke mortality in men, with overall mortality HRs of 0.92 in both men and women (P < .001).
- However, substituting of 3% energy from plant protein for white meat protein was not associated with risk of mortality.
- The overall mortality association with a 1-SD increment of plant protein intake was largely similar across population subsets based on smoking status, diabetes, fruit consumption, vitamin supplement use, and self-reported health status in both men and women, and on use of postmenopausal hormone therapy for women.
- By contrast, significant differences were observed in subgroups of BMI and years of follow-up (in both sexes), alcohol consumption (men), and age and vegetable consumption (women) (Figure).
- We observed stronger inverse associations for those with BMI of 18.5 to 25 (men: HR, 0.94; 95% CI, 0.91-0.96; women: HR, 0.92; 95% CI, 0.89-0.95), follow-up time of less than 5 years (men: HR, 0.92; 95% CI, 0.90-0.95; women: HR, 0.90; 95% CI, 0.87-0.93), consumption of 3 drinks or more daily (HR, 0.89; 95% CI, 0.85-0.93) for men, and for women who were younger than 60 years of age (HR, 0.90; 95% CI, 0.87-0.94) or who reported lower dietary vegetable consumption (HR, 0.90; 95% CI, 0.88-0.93).
- We examined mortality associations of substituting 3% of energy from plant protein for the equivalent amount of animal protein of various sources and shows a significant association with lower overall mortality (men: HR, 0.90; 95% CI, 0.88-0.93; women: HR, 0.90; 95% CI, 0.87-0.93; all P < .001), and CVD mortality (men: HR, 0.89; 95% CI, 0.85-0.94; women: HR, 0.88; 95% CI, 0.82-0.94; P < .001).
Limitations Noted in the Document
- The study’s findings may not be generalizable to other racial/ethnic populations because the vast majority of the study participants were non-Hispanic white.
- The study used a food frequency questionnaire to assess intake of plant protein and animal protein, which has inherent measurement error that would underestimate associations.
- Our analysis was based on a single dietary assessment at baseline, with possible unknown dietary changes during the follow-up period.
- The study’s findings are unlikely to be materially altered by the additional events, the present analysis was conducted with the currently available follow-up.
- Cannot preclude the possibility of residual confounding.
Conclusion
In this large prospective cohort, higher plant protein intake was associated with small but significant associations between higher intake of plant protein and lower overall and CVD mortality, with prominent inverse associations observed for replacement of egg protein and red meat protein with plant protein, particularly for plant protein derived from bread, cereal, and pasta. The findings from this and previous studies provide evidence that dietary modifications in choice of protein sources may promote health and longevity. The authors state that “In this large prospective cohort, higher plant protein intake was associated with reduced risk of overall mortality, with men and women experiencing (respectively) 12% and 14% lower mortality per 10 g/1000 kcal intake increment (5% lower mortality per 1-SD increment).” The inverse association was apparent for CVD and stroke mortality in both sexes, was independent of several risk factors, and was evident in most other cohort subgroups. Replacement of 3% energy from various animal protein sources with plant protein was associated with 10% decreased overall mortality in both sexes. Of note, substitution analyses suggested that replacement of egg protein and red meat protein with plant protein resulted in the most prominent protective associations for overall mortality, representing 24% and 21% lower risk for men and women, respectively, for egg protein replacement, and 13% and 15% lower risk for men and women for red meat protein replacement. The effect sizes of these risk estimates were small.
IFFS Team Summary
- 237 K men and 179 k women from US National Institutes of Health–AARP Diet and Health Study
- Median age 62 years for both men and women
- Avg 16 years follow up
- 77k deaths analyzed
- Median age 62 years for both men and women
- Avg 16 years follow up
- 77k deaths analyzed
- Higher plant protein intake was associated with reduced risk of overall mortality
- Per 10g plant protein for every 1000 k cal consumed (I.e. evert 25g for plant protein in a 2500 k cal diet)
- 12% lower mortality for men
- 14% lower mortality for women
- Per 10g plant protein for every 1000 k cal consumed (I.e. evert 25g for plant protein in a 2500 k cal diet)
- 12% lower mortality for men
- 14% lower mortality for women
- Every 3% of energy from animal protein that was replaced with plant protein
- Resulted in 10% mortality reduction for men and women
- Resulted in 10% mortality reduction for men and women
- These results were stronger for:
- Reducing protein from egg consumption (24% lower risk in men, 21% lower in women)
- Reducing protein from red meat consumption (13% lower risk in men, 15% lower in women)
- Reducing protein from egg consumption (24% lower risk in men, 21% lower in women)
- Reducing protein from red meat consumption (13% lower risk in men, 15% lower in women)
- Highest benefits for Overall Mortality and Cardiovascular Disease
- Some decrease was also seen for Cancer, Stroke, and Infection related mortality
- Deceased mortality seen from injuries and accidents
- Could this be related to decreased falls in those with better brain health?
- Could this be related to decreased falls in those with better brain health?
- Even smokers had some benefits (though diet could not compensate for smoking habits)