Table 1. Characteristics of the cohort studies of dairy products intake and type 2 diabetes mellitus (T2DM).
| Author,y | populati | Country | Men | Age, y | Follow | Subjects | Dietary | Dairy quantity | Relative risk | Assessment | Adjustment |
| on | (%) | -up,y | (cases) | Assessment | (high vs. low intake) | of T2DM | |||||
| Sluijs,2012 | (EPIC- | 8 countries | 50% | 52 | 16 | 24,475 | FFQ | Total dairy (628.9 g vs.79.7 g) | 0.97 (0.82,1.15) | Self reporting, primary | Center, age, sex, BMI, educational |
| InterAct) | in Europe | (10,694) | 24-h dietary | Milk (486.1 g vs. 0.3 g) | 1.08 (0.90,1.31) | care registers, | level, smoking, physical activity, | ||||
| recall data | Yogurt (190.4 g vs. 8.4 g) | 0.89 (0.77,1.03) | secondary care | alcohol intake, fruit plus vegetables, | |||||||
| Cheese (73.7 g vs. 3.2 g) | 0.83 (0.7,0.98) | registers, medication | red meat, processed meat, sugar | ||||||||
| use(drug registers), | |||||||||||
| hospital admissions, | |||||||||||
| Fermented dairy (220.7 g vs. 40.4 g) | 0.85 (0.73,0.99) | and mortality data | sweetened soft drinks, coffee, cereals, | ||||||||
| Grantham, | AusDiab | Australia | 45% | 52 | 5 | 5,582 | 121-item FFQ | Total dairy (408 g vs. 346 g) | 0.71 (0.48,1.05) | 75 g OGTT | Age, sex, energy intake, family |
| 2012 | (209) | Low-fat milk (375 g vs. 200 g) | 0.65 (0.44,0.94) | history of diabetes, education level, | |||||||
| Full-fat milk (375 g vs. 200 g) | 1.18 (0.78,1.79) | level of physical activity, smoking, | |||||||||
| Yogurt (73 g vs. 3 g) | 1.14 (0.78,1.67) | TAG, HDL cholesterol, systolic blood | |||||||||
| Cheese (20 g vs. 6 g) | 0.78 (0.48,1.15) | pressure, waist circumference and hip | |||||||||
| Louie, 2012 | BMES | Australia | 42% | 63.5 | 10 | 1,824 | 145-item FFQ | Total dairy (3.1 vs. 0.5) | 1.50 (0.47,4.77) | Self-reporting, taking | Age, sex, smoking, physical activity, |
| (145) | Low-fat dairy (2.1 vs. 0) | 1.09 (0.57,2.09) | medication for T2DM, | dietary glycemic load, fibre, total | |||||||
| Full-fat dairy (1.9 vs. 0.1) | 0.87(0.48,1.58) | fasting blood glucose | energy intake and family history of | ||||||||
| >7.0 mmol/L | type 2 diabetes, calcium. | ||||||||||
| Struijk, | Inter99 | Denmark | 47.5% | 30–60 | 5 | 5,232 | FFQ | Total dairy (578 g vs. 47 g) | 0.96 (0.58,1.58) | 75 g OGTT | Age, gender and intervention group, |
| 2012 | (214) | Low-fat dairy (536 g vs. 57 g) | 0.85 (0.52,1.40) | diabetes family history, education | |||||||
| Full-fat dairy (89 g vs. 4 g) | 0.94 (0.56,1.58) | level, physical activity smoking | |||||||||
| Milk (546 g vs. 16 g) | 0.95 (0.58,1.57) | status, alcohol intake, wholegrain | |||||||||
| Cheese (49 g vs. 4 g) | 0.78 (0.47,1.29) | cereal, meat, fish, coffee, tea, fruit, | |||||||||
| Fermented dairy (260 g vs. 13 g) | 0.86 (0.50,1.47) | vegetables, energy intake, change in | |||||||||
| diet waist circumference | |||||||||||
| Soedamah | Whitehall II | England | 72% | 56 | 9.8 | 4,186 | 114-item FFQ | Total dairy (575 g vs. 246 g) | 1.30 (0.95,1.77) | Self-reporting, and | Age, ethnicity and employment grade, |
| Muthu, 2012 | (273) | Low-fat dairy (458 g vs. 28 g) | 0.98 (0.73,1.31) | 75 g OGTT | smoking, alcohol intake, BMI, | ||||||
| Full-fat dairy (182 g vs. 27 g) | 1.23 (0.91,1.67) | physical activity and family history of | |||||||||
| Yogurt (117 g vs. 0 g) | 1.04 (0.87,1.58) | CHD/hypertension, fruit and | |||||||||
| Milk (441 g vs. 147 g) | 0.97 (0.71,1.31) | vegetables, bread, meat, fish, coffee, | |||||||||
| Fermented dairy (105 g vs. 17 g) | 1.17 (0.87,1.58) | tea and total energy intake. | |||||||||
| Cheese (31 g vs. 6 g) | 1.20 (0.88,1.64) | ||||||||||
| Margolis, | WHI-OS | USA | 0 | 50–79 | 8 | 82,076 | 122-item FFQ | Total dairy (3.4 vs.0.5) | 0.93 (0.83,1.04) | Self-reporting | Age, race/ethnicity, total energy |
| 2011 | (3946) | Low-fat dairy (2.8 vs.0.05) | 0.65 (0.44,0.96) | confirmed by review | intake, income, education, smoking, | ||||||
| Full-fat dairy (1.3 vs.0.06) | 0.80 (0.65,0.99) | of medical records | alcohol intake, family history of | ||||||||
| Yogurt (≥2/wk vs.<1/mo) | 0.46 (0.31,0.68) | diabetes, postmenopausal hormone | |||||||||
| therapy, blood pressure, BMI, | |||||||||||
| physical activity, dietary glycemic | |||||||||||
| load, dietary total fat, dietary total | |||||||||||
| fiber, magnesium | |||||||||||
| Malik, 2011 | NHS II cohort | USA | 0 | 34–53 | 8 | 37,038 | 133-item FFQ | Total dairy (2.14 vs. 0.62) | 0.75 (0.55,1.02) | self-reporting | Age, BMI, total energy intake, family |
| (550) | Low-fat dairy (1.44 vs.0.18) | 0.74 (0.54,1.01) | confirmed by | history of diabetes, smoking, physical | |||||||
| Full-fat dairy (1.14 vs. 0.19) | 0.72 (0.53,0.99) | review of medical | activity, alcohol, oral contraceptive | ||||||||
| records | use, hormone replacement therapy. | ||||||||||
| Polyunsaturated, saturated fat, glycemic | |||||||||||
| load, fiber, trans fat, processed meat, | |||||||||||
| carbonated soft drinks, fruit drinks,coffee | |||||||||||
| Kirii, 2009 | JPHC cohort | Japan | 57% | 40–69 | 5 | 59,796 | FFQ | Total dairy (≥300 g vs. <50 g) | Males 1.18 (0.90,1.56) | Self-reporting, | Age, area, BMI, family history of |
| (1,114) | Females 0.71(0.51,0.98) | Validity verified by | diabetes mellitus, smoking, alcohol | ||||||||
| Milk (≥200 g vs. <50 g) | Males 1.02(0.85,1.24) | medical record data | intake hypertension, exercise, coffee, | ||||||||
| Females 0.87(0.70,1.09) | and plasma glucose | magnesium, total energy | |||||||||
| Cheese (≥5 g vs. 0 g) | Males 0.88 (0.64,1.21) | random samples. | |||||||||
| Females 1.12(0.80,1.57) | |||||||||||
| Yogurt (≥60 g vs. 0 g) | Males 1.01 (0.75,1.36) | ||||||||||
| Females 0.77(0.58,1.01) | |||||||||||
| Villegas, | SWHS cohort | China | 0 | 51 | 6.9 | 64,191 | FFQ | Milk (250 vs. 0) | 0.60 (0.41,0.88) | Self-reporting, | Age, energy intake, BMI, waist-hip |
| 2009 | (2,270) | fasting glucose and | ratio, smoking status, alcohol | ||||||||
| OGTT | consumption, physical activity, | ||||||||||
| income level, education level, | |||||||||||
| occupation, and hypertension. | |||||||||||
| Elwood | Caerphilly | UK | 100 | 45–59 | 20 | 640 | FFQ and 7- | Milk | 0.57 (0.20,1.63) | Self-reporting | Age, smoking, BMI and social class |
| 2007 | prospective | (41) | day weighed | ||||||||
| study | intake | ||||||||||
| Liu, 2006 | WHS cohort | USA | 0 | 55 | 10 | 37,183 | 131-item FFQ | Total dairy (>2.9 vs. <0.85) | 0.68 (0.52,0.89) | Diagnostic criteria | Age, total energy intake, |
| (1063) | Low-fat dairy (>2.0 vs. ≤0.27) | 0.69 (0.522,0.91) | of ADA, based on | randomized-treatment assignment, | |||||||
| Full-fat dairy (>1.33 vs. <0.2) | 0.99 (0.82,1.20) | self-reporting, 3 | family history of diabetes, smoking, | ||||||||
| Yogurt (≥2/wk vs.<1/mo) | 0.82 (0.70,0.97) | complementary | BMI, hypercholesterolemia, | ||||||||
| Whole milk (≥2/wk vs.<1/mo) | 1.04 (0.84,1.30) | approaches to | hypertension, physical activity | ||||||||
| Skim milk (≥2/wk vs.<1/mo) | 0.92 (0.78,1.09) | validate the cases | hormones, alcohol consumption, fiber, | ||||||||
| Cottage cheese (≥2/wk vs<1/mo) | 0.86 (0.71,1.05) | total fat, and dietary glycemic load, | |||||||||
| Ice cream (≥2/wk vs.<1/mo) | 0.88 (0.74,1.05) | calcium, vitamin D, and magnesium. | |||||||||
| Other cheese (≥2/wk vs.<1/mo) | 0.80 (0.64,1.01) | ||||||||||
| Van Dam, | Black Women’s | USA | 0 | 21–69 | 8 | 41,186 | FFQ | Total dairy (2.53 vs. 0.07) | 0.93 (0.75,1.15) | Self-reporting, | Age, total energy intake, BMI, |
| 2006 | Health Study | (1,964) | Low-fat dairy (1.22 vs. 0) | 0.87 (0.76,1.00) | validity verification | smoking physical activity, alcohol, | |||||
| Full-fat dairy (1.33 vs. 0.07) | 1.03 (0.88,1.20) | of a random sample | family history of diabetes, education | ||||||||
| level, coffee, sugar-sweetened soft | |||||||||||
| drink, processed meat, red meat, | |||||||||||
| . | calcium or magnesium intake | ||||||||||
| Pittas, 2006 | NHS cohort | US | 0 | 30–55 | 20 | 83,779 | FFQ | Total dairy (3.9 vs. 0.9) | 0.79 (0.70,0.90) | Criteria by | Age, BMI, hypertension, family |
| (4,843) | National Diabetes | history of diabetes, smoking, physical | |||||||||
| Data Group and | activity, caffeine, alcohol, and state of | ||||||||||
| ADA.self-reporting | residence, fat (saturated, | ||||||||||
| polyunsaturated, or trans), cereal fiber, | |||||||||||
| glycemic load, magnesium, and retinol | |||||||||||
| Choi, 2005 | HPFS cohort | USA | 100 | 43–75 | 12 | 41,254 | FFQ | Total dairy (≥2.9 vs. <0.9) | 0.75 (0.61,0.93) | Criteria by | Age, total energy intake, family history |
| (1243) | Low-fat dairy (>1.58 vs. <0.14) | 0.74 (0.60,0.91) | National Diabetes | of diabetes, smoking, BMI, | |||||||
| Full-fat dairy (>1.72 vs.<0.38) | 0.82 (0.66,1.02) | Data Group. Based | hypercholesterolemia, hypertension, | ||||||||
| Yogurt (≥2/wk vs. <1/mo) | 0.83 (0.66,1.06) | on self-reporting. | physical activity, alcohol, fiber, | ||||||||
| Whole milk (≥2/wk vs. <1/mo) | 1.19 (1.00,1.43) | Validity verified | trans-fat polyunsaturated to saturated | ||||||||
| Low-Fat milk (≥2/wk vs. <1/mo) | 0.78 (0.63,0.97) | with medical | fat, glycemic load | ||||||||
| Cottage cheese (≥2/wk vs. <1/mo) | 0.96 (0.80,1.17) | records in a sample | |||||||||
| Other cheese (≥2/wk vs. <1/mo) | 0.88 (0.67,1.16) | of 71 participants. | |||||||||
| Ice cream (≥2/wk vs. <1/mo) | 0.78 (0.64,0.95) | ||||||||||
| Montonen, | Finnish Mobile | Finland | 50 | 40–69 | 23 | 4304 | dietary history | Regular dairy (>305 vs. <39) | 0.81(0.62–1.08) | from the Social | Adjusted for age, sex, body mass |
| 2005 | Clinic Health | (383) | interview | Low fat dairy (>0 vs. 0) | 0.90(0.60,1.36) | Insurance | index, energy intake, smoking, family | ||||
| Examination | Whole milk (>878 vs. <326) | 1.06(0.75,1.50) | Institution’s | history of diabetes, and geographic | |||||||
| Survey | nationwide register | area | |||||||||
| of persons | |||||||||||
| receiving drug | |||||||||||
| reimbursement | |||||||||||
| Ericson, | Malmö Diet and | Sweden | 57 | 8 | 23 531 | 148-FFQ | Total dairy women (6.0 vs. 1.8) | 0·88 (0·70, 1·09) | Self report, and | sex, smoking status, alcohol | |
| 2013 | Cancer cohort | (837) | Total dairy men (6.3 vs. 1.8) | 1·20 (0·98, 1·47) | verified with an | consumption, leisure-time physical | |||||
| inquiry to the | activity, BMI, waist-to-hip ratio, | ||||||||||
| treating physician, | hypertension, history of high blood lipid | ||||||||||
| local cancer | levels at baseline, education, vitamin | ||||||||||
| registries | supplementation, non-consumption of the | ||||||||||
| respective food group, total energy intake | |||||||||||
| (kJ/day). |
FFQ, food-frequency questionnaire; OGGT, oral glucose tolerance test; BMI, body mass index TAG,triglyceride; HDL, High density lipoproteins; ADA, American Diabetes Association CHD, coronary heart disease; EPIC-InterAct European Prospective Investigation into Cancer and Nutrition cohort; AusDiab, Australian Diabetes Obesity and Lifestyle Study; BMES, Blue Mountains Eye Study; WHI-OS, Women’s Health Initiative observational study; NHS, The Nurses’ Health Study; JPHC, Japan Public Health Center-based Prospective Study; SWHS, Shanghai Women’s Health Study; WHS, Women’s Health Study; HPFS, Health Professionals Follow-up Study.