Replacing the consumption of red meat with other major dietary protein sources and risk of type 2 diabetes mellitus: a prospective cohort study

Abstract

Background: Greater consumption of red meat has been associated with a higher risk of type 2 diabetes mellitus (T2DM). A decreased intake of red meat and simultaneous increased intake of other high-protein foods may be associated with a lower risk of T2DM. These analyses of specific food replacements for red meat may provide more accurate dietary advice.

Objective: We examined the association between a decrease in intake of red meat accompanied by an increase in other major dietary protein sources and risk of T2DM.

Methods: We prospectively followed 27,634 males in the Health Professionals Follow-up Study, 46,023 females in the Nurses' Health Study, and 75,196 females in the Nurses' Health Study II. Diet was assessed by a validated FFQ and updated every 4 y. Cox proportional hazards models adjusted for T2DM risk factors were used to model the food replacements. We calculated HRs and 95% CIs for the T2DM risk associated with replacements of 1 daily serving of red meat with another protein source.

Results: During 2,113,245 person-years of follow-up, we identified 8763 incident T2DM cases from 1990 to 2013. In the pooled analyses, a decrease in total red meat intake during a 4-y period replaced with another common protein food was associated with a lower risk of T2DM in the subsequent 4-y period. The HR (95% CI) per 1 serving/d was 0.82 (0.75, 0.90) for poultry, 0.87 (0.77, 0.98) for seafood, 0.82 (0.78, 0.86) for low-fat dairy, 0.82 (0.77, 0.86) for high-fat dairy, 0.90 (0.81, 0.99) for eggs, 0.89 (0.82, 0.98) for legumes, and 0.83 (0.78, 0.89) for nuts. The associations were present for both unprocessed and processed red meat, although stronger for the replacement of processed red meat.

Conclusions: Replacing red meat consumption with other protein sources was associated with a lower risk of T2DM.

Keywords: cohort study; diet change; follow-up study; food substitution; protein replacement; red meat; type 2 diabetes mellitus.

FIGURE 1

HR (95% CI) for type 2 diabetes mellitus associated with 1 daily serving decreased intake of unprocessed or processed red meat and concomitant increased intake of another major dietary protein source in pooled analyses of all 3 cohorts (n = 148,853). Note: the Cox proportional hazards models included all protein foods simultaneously and were adjusted for age, calendar time, and calories (initial and change, both quintiles), marital status (with spouse, yes or no), race (white, African American, Asian/other), family history of diabetes (yes, no), history of hypertension (yes, no), history of hypercholesterolemia (yes, no), BMI (<20.0, 20.0 to <23.0, 23.0 to <25.0, 25.0 to <30.0, ≥30.0 kg/m²) alcohol intake, both as initial intake (0, 0.1–4.9, 5–14.9, 15–29.9, ≥30 g/d) and change (quintiles), modified AHEI (initial and change, both quintiles), smoking status change (never to never, current to past, past to current, never to current, past to past, current to current), physical activity (metabolic equivalents initially and change, both quintiles) and for women initial menopausal status and use of postmenopausal hormones (premenopausal, postmenopausal + never hormone use, postmenopausal + past hormone use, postmenopausal + current hormone use), initial intake of red meat, poultry, seafood, low-fat dairy, high-fat dairy, eggs, legumes, and nuts (servings/d). 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. The results across the 3 cohorts were pooled using an inverse variance weighted, fixed-effect meta-analysis. AHEI, Alternative Healthy Eating Index.