In order to compare protein sources with the different outcomes, we evaluated the association of substituting vegetable protein for animal protein (a change that would be expected if one were following a vegetarian diet)
Only the association with CHD mortality was statistically significant, and resulted in a 30% decreased risk among women in the highest compared to lowest category of vegetable protein intake, representing a substitution of 2.4% of total energy from vegetable for animal protein (95% CI, 0.51-0.98; P for trend, 0.02).
While the use of nutrient values was necessary to evaluate our hypotheses of protein for carbohydrate substitutions on various outcomes, realistically most individuals interchange foods when implementing dietary changes. Therefore, we also examined the association of different food sources of protein in place of carbohydrate-rich foods, standardized as servings per 1,000 calories, while holding constant total energy, dietary fats and other covariates.