Table 1.

Prospective cohort studies published between 2008 and August 2018 that investigated the associations between dietary fat, fatty acids, and food sources of fat and risk for diabetes

Reference	Region	Study objective	Population	Dietary assessment	Adjustments	Time	Results	Risk
Kaushik (2009) [12]	USA	Investigate the association between dietary long-chain FA and incidence of T2D	195,204 American adults, 24–78 years of age, from three prospective cohorts (Nurses’ Health Study (NHS), NHS 2, and the Health Professionals Follow-Up Study)	Validated FFQ	Smoking, alcohol consumption, physical activity, family history of diabetes, BMI, saturated fat, trans-fat, linolenic acid, linoleic acid, caffeine, cereal fiber, glycemic index, calories	14–18 years	Highest quintile vs. lowest quintile of long-chain FA (RR = 1.24; 95% CI 1.09, 1.40) was associated with T2D. Association existed across all quintiles	↑
Brostow et al. (2011) [13]	Singapore	Investigate the associations between total n-3 FA, EPA, DHA, non-marine α-linolenic acid, n-6 FA, and n-6:n-3 FA ratio and T2D risk	43,176 Chinese men and women living in Singapore, 45–74 years of age, free from diabetes at baseline, from the Singapore Chinese Health Study	Semi-quantitative FFQ	Fully adjusted model: age, sex, interview year, dialect, hypertensive status, smoking, alcohol frequency, education, BMI, physical activity, hypertension, dietary factors	5.7 years	Highest quintile vs. lowest quintile of total n-3 PUFA (HR = 0.78; 95% CI 0.65, 0.94) and non-marine α-linolenic acid (HR = 0.79; 95% CI 0.67, 0.93) were inversely associated with validated self-reported incidence of diabetes	↓
							EPA, DHA, n-6 PUFA, and n-6:n-3 PUFA ratio were not associated with T2D	↔
Djoussé et al. (2011) [14]	USA	Investigate the association between n-3 FA and fish consumption and T2D risk	36,328 women, 54.6 year of age on average, free from diabetes at baseline, from the Women’s Health Study	Validated FFQ	Age, BMI, parental history of diabetes, smoking, exercise, alcohol, menopausal status, red meat intake, quintiles of energy intake, linoleic acid, α-linolenic acid, dietary magnesium, trans and saturated fats, cereal fiber, glycemic index	12.4 years	Highest quintile vs. lowest quintile of marine n-3 FA (HR = 1.44; 95% CI 1.25, 1.65) were positively associated with validated self-reported incidence of diabetes	↑
							Plant-based n-3 FA were not associated with incident diabetes	↔
Margolis et al. (2011) [15]	USA	Investigate the association between low-fat dairy consumption and incident T2D	82,076 multiethnic postmenopausal women 50–79 years of age, free from diabetes at baseline, from the Women’s Health Initiative Observational Study	Semi-quantitative FFQ	Age, race/ethnicity, total energy intake, income, education, smoking, alcohol consumption, use of postmenopausal hormone therapy, physical activity, family history of diabetes, BMI, blood pressure, dietary factors	8 years	Highest quintile vs. lowest quintile of low-fat dairy consumption (RR = 0.65; 95% CI 0.44, 0.96) was inversely associated with risk of T2D	↓
							High yogurt consumption (≥ 2×/week) (RR = 0.46; 95% CI 0.31, 0.68) was inversely associated with risk of T2D	↓
							High-fat dairy consumption was not associated with risk of diabetes	↔
Villegas et al. (2011) [16]	China	Investigate the associations between fish, shellfish, and long chain n-3 FA and the risk of T2D	51,963 Chinese men and 64,193 Chinese women, 40–74 years of age, free from diabetes at baseline, from the Shanghai Men’s Health Study and the Shanghai Women’s Health Study	In-person interview using a validated FFQ	Age, energy intake, waist-to-hip ratio, BMI, smoking, alcohol consumption, physical activity, income level, educational level, occupation, family history of diabetes, hypertension, and dietary pattern	Approximately 5–10 years	Highest quintile vs. lowest quintile of shellfish (HR = 0.86; 95% CI 0.76, 0.99) and long-chain n-3 FA intakes (HR = 0.84; 95% CI 0.74, 0.95) were inversely associated with T2D in women. Inverse associations existed across all quintiles	↓
Alhazmi et al. (2013) [17]	Australia	Investigate the association between macronutrient intake and T2D risk	8370 Australian women, 45–50 years of age, free from diabetes at baseline, from the Australian Longitudinal Study of Women’s Health	Validated FFQ called the Dietary Questionnaire for Epidemiological Studies	Fully adjusted model: lifestyle, sociodemographic factors, other fat types, fiber, energy	6 years	Highest quintile vs. lowest quintile of MUFA (RR = 1.64; 95% CI 1.06, 2.54), total n-3 PUFA (RR = 1.55; 95% CI 1.03, 2.32), α-linolenic acid (RR = 1.84; 95% CI 1.25, 2.71), and total n-6 PUFA (RR1.60; 95% CI 1.03, 2.48) were positively associated with validated self-reported incidence of T2D	↑
							No associations between total dietary carbohydrate, protein or fat with T2D	↔
Louie et al. (2013) [18]	Australia	Investigate the associations between baseline consumption of dairy products and incidence of metabolic syndrome and T2D	1807 (included in study of metabolic syndrome) and 1824 (included in study of T2D) Australian men and women 49 years of age and older, from the Blue Mountains Eye Study	Validated semi-quantitative FFQ	Fully adjusted model: age, sex, smoking status, physical activity, dietary glycemic load, fiber from vegetables, total energy intake, family history of T2D, systolic blood pressure, BMI, HDL-cholesterol, total cholesterol, triglyceride, calcium	10 years	Highest quartile vs. lowest quartile of regular-fat dairy consumption (OR = 0.41; 95% CI 0.23, 0.71) was inversely associated with risk of metabolic syndrome	↑
							Total dairy consumption was not associated with risk of metabolic syndrome or T2D	↔
Ericson et al. (2015) [19]	Sweden	Investigate the associations between dietary fat and its food sources and T2D risk	26,930 men and women, 45–74 years of age, free from diabetes at baseline, from the Malmö Diet and Cancer Cohort	Interview-based modified diet-history method that combined diet recall, FFQ, and dietary assessment interview	Energy, age, sex, method version, season, leisure time, physical activity, smoking, alcohol intake, education, BMI	14 years	Highest quintile vs. lowest quintile of high-fat dairy consumption (HR = 0.77; 95% CI 0.68, 0.87) was inversely associated with incidence of T2D	↓
							Highest quintile vs. lowest quintile of meat consumption was associated with increased risk for T2D (HR = 1.36; 95% CI 1.20, 1.55)	↑
							Highest quintile vs. lowest quintile of saturated FA with 4–10 carbons (HR = 0.83; 95% CI 0.74, 0.93), lauric acid (12:0) (HR = 0.84; 95% CI 0.75, 0.95, and myristic acid (14:0) (HR = 0.83; 95% CI 0.74, 0.94) were associated with decreased risk for T2D	↓
							Total dietary fat was not associated with T2D	↔
Guasch-Ferré et al. (2015) [20]	USA	Investigate the association between olive oil consumption and incident T2D	59,930 women 37–65 years of age, free from diabetes at baseline, from the Nurses’ Health Study (NHS) I and 85,157 women, 26–45 years of age, free from diabetes at baseline, from the NHS II	Validated FFQ	Ethnicity, ancestry, smoking status, alcohol, physical activity, family history of diabetes, history of hypertension, history of hypercholesterolemia, multivitamin use, postmenopausal status, menopausal hormone use, quintiles of Alternative Healthy Eating Index, total energy intake	22 years	>1 tablespoon of olive oil (>8 g) per day vs. those who never consumed olive oil (HR = 0.90; 95% CI 0.82, 0.99) was inversely associated with risk for T2D	↓
							Substituting 1 tablespoon olive oil per day for margarine, butter, or mayonnaise = 5%, 8%, and 15% lower risk of T2D, respectively	↓
Guasch-Ferré et al. (2017) [21]	Spain	Investigate the associations between total fat, subtypes of dietary fat, and food sources rich in saturated FA and incidence of T2D	3349 Spanish men, 55–80 years of age and Spanish women 60–80 years of age, free from diabetes at baseline, from the PREvención con DIeta MEDiterránea (PREDIMED) study	Validated semi-quantitative FFQ	Fully adjusted model: age, sex, BMI, smoking status, educational status, leisure-time physical activity, baseline hypertension or the use of antihypertensive medication, total energy intake, alcohol intake, quartiles of fiber, protein intake, dietary cholesterol, quartiles of the other subtypes of fat, hypercholesterolemia, lipid lowering drugs, fasting plasma glucose at baseline	4.3 years	Highest quartile vs. lowest quartile of saturated and animal fat consumption (HR = 2.19: 95% CI 1.28, 3.73) were associated with risk for T2D	↑
							Total dietary fat, MUFA, PUFA, trans FA were not associated with T2D	↔

BMI body mass index, CI confidence interval, DHA docosahexaenoic acid, EPA eicosapentaenoic acid, FA fatty acid, FFQ food frequency questionnaire, HR hazard ratio, MUFA monounsaturated fatty acid, OR odds ratio, PUFA polyunsaturated fatty acid, RR relative risk, T2D type 2 diabetes