Table 4. Risk of Overall and Cause-Specific Mortality Associated With Substitution of 3% Dietary Energy Intake From Plant Protein for Various Animal Protein Sources Among 237 036 Men and 179 068 Womena.
Animal protein source | Men | Women | ||
---|---|---|---|---|
HR (95% CI)b | P value | HR (95% CI)b | P value | |
Total animal protein | ||||
Overall | 0.90 (0.88-0.93) | <.001c | 0.90 (0.87-0.93) | <.001c |
Cancer | 0.97 (0.92-1.02) | .19 | 0.93 (0.88-0.99) | .02 |
CVD | 0.89 (0.85-0.94) | <.001c | 0.88 (0.82-0.94) | <.001c |
Heart disease | 0.91 (0.86-0.96) | <.001c | 0.90 (0.84-0.98) | .01 |
Stroke | 0.78 (0.68-0.90) | <.001c | 0.81 (0.70-0.94) | .007 |
Respiratory disease | 0.92 (0.82-1.03) | .14 | 0.92 (0.82-1.04) | .18 |
Infection | 0.99 (0.85-1.14) | .86 | 0.81 (0.68-0.97) | .02 |
Injury and accident | 0.82 (0.71-0.95) | .008 | 0.89 (0.72-1.11) | .30 |
Other causes combined | 0.83 (0.78-0.88) | <.001c | 0.90 (0.83-0.97) | .005 |
Red meat protein | ||||
Overall | 0.87 (0.85-0.90) | <.001c | 0.85 (0.81-0.88) | <.001c |
Cancer | 0.93 (0.88-0.98) | .004 | 0.89 (0.83-0.95) | <.001c |
CVD | 0.88 (0.83-0.93) | <.001c | 0.82 (0.76-0.89) | <.001c |
Heart disease | 0.89 (0.84-0.94) | <.001c | 0.84 (0.77-0.92) | <.001c |
Stroke | 0.79 (0.68-0.91) | .001c | 0.75 (0.63-0.89) | <.001c |
Respiratory disease | 0.83 (0.74-0.94) | .003 | 0.80 (0.70-0.91) | .001c |
Infection | 0.94 (0.81-1.10) | .47 | 0.75 (0.62-0.92) | .006 |
Injury and accident | 0.80 (0.69-0.94) | .005 | 0.90 (0.71-1.16) | .42 |
Other causes combined | 0.80 (0.75-0.86) | <.001c | 0.85 (0.78-0.92) | <.001c |
White meat protein | ||||
Overall | 0.98 (0.94-1.01) | .11 | 0.95 (0.91-0.98) | .005 |
Cancer | 1.02 (0.97-1.08) | .40 | 0.93 (0.87-0.99) | .02 |
CVD | 0.95 (0.90-1.01) | .07 | 0.94 (0.87-1.02) | .12 |
Heart disease | 0.97 (0.91-1.03) | .27 | 0.97 (0.89-1.05) | .41 |
Stroke | 0.83 (0.71-0.96) | .01 | 0.90 (0.76-1.06) | .20 |
Respiratory disease | 1.17 (1.03-1.33) | .02 | 1.06 (0.93-1.21) | .40 |
Infection | 1.06 (0.90-1.25) | .52 | 0.88 (0.72-1.07) | .20 |
Injury and accident | 0.91 (0.77-1.07) | .23 | 0.99 (0.78-1.26) | .96 |
Other causes combined | 0.90 (0.84-0.97) | .004 | 0.90 (0.84-0.97) | .004 |
Dairy protein | ||||
Overall | 0.92 (0.89-0.95) | <.001c | 0.92 (0.89-0.95) | <.001c |
Cancer | 1.01 (0.96-1.06) | .73 | 0.97 (0.91-1.03) | .31 |
CVD | 0.89 (0.84-0.94) | <.001c | 0.88 (0.82-0.95) | <.001c |
Heart disease | 0.91 (0.86-0.97) | .003c | 0.92 (0.84-0.99) | .03 |
Stroke | 0.77 (0.66-0.89) | <.001c | 0.80 (0.69-0.94) | .006 |
Respiratory disease | 0.94 (0.83-1.06) | .31 | 0.97 (0.85-1.10) | .60 |
Infection | 0.98 (0.84-1.15) | .79 | 0.80 (0.67-0.97) | .02 |
Injury and accident | 0.83 (0.71-0.98) | .02 | 0.88 (0.70-1.10) | .25 |
Other causes combined | 0.83 (0.78-0.89) | <.001c | 0.90 (0.83-0.97) | .008 |
Egg protein | ||||
Overall | 0.76 (0.72-0.80) | <.001c | 0.79 (0.73-0.85) | <.001c |
Cancer | 0.85 (0.78-0.93) | <.001c | 0.83 (0.73-0.93) | .002c |
CVD | 0.74 (0.67-0.82) | <.001c | 0.72 (0.63-0.83) | <.001c |
Heart disease | 0.76 (0.69-0.85) | <.001c | 0.72 (0.62-0.85) | <.001c |
Stroke | 0.67 (0.52-0.88) | .003 | 0.75 (0.55-1.03) | .08 |
Respiratory disease | 0.61 (0.50-0.74) | <.001c | 0.66 (0.53-0.83) | <.001c |
Infection | 0.94 (0.71-1.25) | .66 | 0.80 (0.55-1.17) | .25 |
Injury and accident | 0.71 (0.53-0.94) | .02 | 0.67 (0.43-1.05) | .08 |
Other causes combined | 0.71 (0.63-0.80) | <.001c | 0.92 (0.78-1.08) | .28 |
Abbreviations: CVD, cardiovascular disease; HR, hazard ratio.
Plant, animal and total protein were modeled as percentages of energy, using energy from the specific protein divided by total energy intake, and the leave-1-out model was used as the substitution approach: to evaluate the substitution of 3% energy from plant protein for an individual component of animal protein (red meat protein, white meat protein, dairy protein and egg protein), energy from plant protein, total protein, and other 3 components of animal protein were simultaneously included in the model. The HRs represent the relative risk of overall mortality with 3% energy from plant protein substituted for the excluded animal protein energy.
Adjusted as indicated in note c of Table 2, and for women, the risk estimates were additionally adjusted for postmenopausal hormone replacement therapy (yes vs no).
Achieved Bonferroni-corrected threshold (.05/18 = .0028).