Circulating Biomarkers of Dairy Fat and Risk of Incident Diabetes Mellitus Among Men and Women in the United States in Two Large Prospective Cohorts

Abstract

Background: In prospective studies, the relationship of self-reported consumption of dairy foods with risk of diabetes mellitus is inconsistent. Few studies have assessed dairy fat, using circulating biomarkers, and incident diabetes mellitus. We tested the hypothesis that circulating fatty acid biomarkers of dairy fat, 15:0, 17:0, and t-16:1n-7, are associated with lower incident diabetes mellitus.

Methods and results: Among 3333 adults aged 30 to 75 years and free of prevalent diabetes mellitus at baseline, total plasma and erythrocyte fatty acids were measured in blood collected in 1989 to 1990 (Nurses' Health Study) and 1993 to 1994 (Health Professionals Follow-Up Study). Incident diabetes mellitus through 2010 was confirmed by a validated supplementary questionnaire based on symptoms, diagnostic tests, and medications. Risk was assessed by using Cox proportional hazards, with cohort findings combined by meta-analysis. During mean±standard deviation follow-up of 15.2±5.6 years, 277 new cases of diabetes mellitus were diagnosed. In pooled multivariate analyses adjusting for demographics, metabolic risk factors, lifestyle, diet, and other circulating fatty acids, individuals with higher plasma 15:0 had a 44% lower risk of diabetes mellitus (quartiles 4 versus 1, hazard ratio, 0.56; 95% confidence interval, 0.37-0.86; P-trend=0.01); higher plasma 17:0, 43% lower risk (hazard ratio, 0.57; 95% confidence interval, 0.39-0.83; P-trend=0.01); and higher t-16:1n-7, 52% lower risk (hazard ratio, 0.48; 95% confidence interval, 0.33-0.70; P-trend <0.001). Findings were similar for erythrocyte 15:0, 17:0, and t-16:1n-7, although with broader confidence intervals that only achieved statistical significance for 17:0.

Conclusions: In 2 prospective cohorts, higher plasma dairy fatty acid concentrations were associated with lower incident diabetes mellitus. Results were similar for erythrocyte 17:0. Our findings highlight the need to better understand the potential health effects of dairy fat, and the dietary and metabolic determinants of these fatty acids.

Keywords: biomarkers; dairy; diabetes mellitus; fatty acid.

© 2016 American Heart Association, Inc.

Figures

Figure 1

Semi-parametric multivariable-adjusted associations of plasma 15:0, 17:0, t-16:1n-7, and 14:0 with incident diabetes among 3,333 US men and women in two separate cohorts, evaluated using restricted cubic splines with covariate adjustments as in Table 2. Solid and dashed lines represent hazard ratios (HRs) and 95% confidence intervals, respectively; dotted vertical lines represent 21 knots. P-values for linear association are shown. P-values for nonlinearity were nonsignificant in all analyses; the automatic selection process did not identify any significant spline variables.