Higher red meat intake links to greater diabetes odds in large US study

Large-scale U.S. population data suggest that red meat consumption patterns may be associated with diabetes prevalence, whereas alternative protein choices may confer metabolic advantages.

Study: Association Between Red Meat Intake and Diabetes: A Cross-Sectional Analysis of a Nationally Representative Sample of U.S. Adults (NHANES 2003–2016). Image Credit: PeopleImages / Shutterstock

In a recent study published in the British Journal of Nutrition, researchers examined the associations between total, processed, and unprocessed red meat intake and diabetes and evaluated the statistical substitution effects of alternative protein sources for red meat.

Background: Red Meat Consumption and Diabetes Burden

More than 37 million Americans have diabetes, and nearly 2 million new cases of diabetes each year are diagnosed in the United States (U.S.). The growing burden of diabetes has drawn greater attention to modifiable lifestyle factors, particularly dietary habits.

Red meat, especially processed varieties such as sausages and deli meats, is widely consumed in the U.S. diet and has been linked to increased risks of cardiovascular disease, certain cancers, and premature mortality.

While there are inconsistencies in its association with diabetes prevalence, particularly in nationally representative populations, uncertainty remains about how red meat compares with other protein sources that may influence diabetes outcomes.

A better understanding of these relationships will help guide appropriate dietary guidelines and effective public health strategies, for which more research is needed.

NHANES Cross-Sectional Design and Statistical Substitution Modeling

This cross-sectional observational analysis used data from the National Health and Nutrition Examination Survey (NHANES) 2003-2016, conducted by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC). NHANES employs multistage probability sampling designed to represent the non-institutionalized U.S. population. After excluding individuals with missing data, implausible energy consumption, or unknown diabetes status, 34,737 adults aged 18 years and older were included.

Dietary intake was assessed using up to two 24-hour dietary recalls collected with the United States Department of Agriculture (USDA) Automated Multiple-Pass Method, which may not fully reflect usual long-term dietary intake and may introduce measurement error or recall bias. Red meat was classified as processed (e.g., sausages, frankfurters, luncheon meats) or unprocessed (e.g., beef, pork, lamb), and intake was measured in servings per day.

Diabetes was defined by self-report of physician diagnosis, fasting plasma glucose ≥126 mg/dL, glycated hemoglobin (HbA1c) ≥6.5%, or use of antidiabetic medications. Covariates included age, sex, race/ethnicity, education, poverty income ratio (PIR), body mass index (BMI), smoking, alcohol intake, physical activity measured in metabolic equivalent of task (MET) minutes per week, psychological distress, food insecurity, and overall dietary factors.

Researchers used logistic regression models to estimate adjusted odds ratios (ORs) and 95% confidence intervals (CIs). Substitution analyses were conducted to assess the impact of reducing red meat intake by half a serving per day through modeled dietary replacements, rather than observed behavioral changes.

Associations Between Red Meat Intake and Diabetes Odds

Among 34,737 participants with a mean age of 45.8 years, 10.5% had diabetes. Individuals in the highest quintile of total red meat intake consumed a median of 5.72 ounces per day and were more likely to be younger, male, non-Hispanic White, married, and physically active. They also had higher BMI and higher intake of calories, alcohol, and sugary drinks than those in the lowest quintile.

After adjusting for demographics, lifestyle, socioeconomic status, and dietary habits (including fruit, vegetables, poultry, fish, eggs, dairy, plant-based protein, whole grains, coffee, sugary drinks, and BMI), red meat consumption was significantly associated with an increased odds of diabetes, but this association does not establish a causal relationship.

Compared to participants with the lowest red meat intake, those in the highest quintile of total red meat intake had increased odds of diabetes, OR 1.49; 95% CIs 1.22-1.81; P-trend <0.001. Similar findings were observed for processed red meat intake, OR 1.47; 95% CIs 1.17-1.84; P-trend = 0.001, and unprocessed red meat intake, OR 1.24; 95% CIs 1.06-1.44; P-trend = 0.006.

Overall, these results indicate that the health consequences of dietary patterns depend on both the item being reduced and the item being substituted. In repeated analyses, each additional serving of total red meat per day corresponded to 16% higher odds of diabetes, OR 1.16; 95% CIs 1.07-1.25. Likewise, approximately 10% higher odds were associated with each serving of processed or unprocessed red meat consumed. No significant interaction was observed between red meat consumption and PIR, food insecurity, sex, age, or psychological distress.

Substitution Effects of Alternative Protein Sources

Analysis of potential substitutions highlighted the practical effects of dietary patterns on the risk of diabetes. Participants who replaced one serving of total red meat per day with plant-based protein sources, such as nuts, seeds, legumes, and soy products, had 14% lower odds of diabetes, OR 0.86; 95% CIs 0.79-0.94.

Similarly, participants who substituted poultry, dairy, or whole grains for total red meat experienced reductions in the odds of diabetes of 11%-12%. Replacing red meat with plant-based protein sources was statistically associated with lower odds of diabetes, although causality cannot be established.

Interpretation, Biological Mechanisms, and Limitations

In this nationally representative sample of U.S. adults, higher intakes of total, processed, and unprocessed red meat were independently associated with an increased risk of diabetes.

The associations persisted after accounting for BMI and overall diet quality. Replacing red meat with plant-based protein sources, poultry, dairy products, and whole grains was associated with a lower odds of diabetes, but not with a proven risk reduction. 

Plant-based proteins showed the strongest statistical association in substitution models. These results illustrate that diet is a potentially modifiable factor associated with diabetes prevalence, although the cross-sectional design precludes conclusions regarding prevention or causation.

Potential biological explanations discussed in the research include the roles of saturated fat, heme iron, sodium, nitrates, and inflammation-related metabolic pathways. However, these mechanisms were not directly tested in this study.

Because this was a cross-sectional observational study, reverse causation and residual confounding remain possible, and the study could not distinguish between type 1 and type 2 diabetes.