Table 1.

Randomised controlled trials (RCT) demonstrating an effect of overall dairy consumption on markers of metabolic health and cardiovascular disease (CVD) risk.

Author	Country	Study Design	Population	Age (Years)	Intervention	Duration	Main Findings
Vasilopoulou et al. 2020 [47]	UK	Crossover	n = 54 (31 male; 23 female), with risk of CVD	25–70	2 arm: (A) monounsaturated fatty acid (MUFA)-modified dairy—340 g UHT milk, 45 g cheese, 25.1 g butter. (B) Control—340 g ultrahigh temperature (UHT) pasteurised milk, 45 g cheese, 25.1 g butter	Two 12-week periods separated by an 8-week washout period	No significant change from baseline in serum total cholesterol (TC) between diets. Group A had a significant beneficial effect in terms of attenuation of the rise of the low-density lipoprotein (LDL) cholesterol. No changes in high-density lipoprotein (HDL) cholesterol between diets. The LDL:HDL ratio decreased significantly after group A, and increased after the control. No significant differences were observed for indexes of insulin sensitivity/resistance. Fasting plasma nitrite concentrations increased after the modified diet, yet decreased after the control.
Markey et al. 2017 [49]	UK	Crossover	n = 54 (31 male; 23 female), with risk of CVD	25–70	2 arm: (A) MUFA-modified dairy—340 g UHT milk, 45 g cheese, 25.1 g butter. (B) Control—340 g UHT milk, 45 g cheese, 25.1 g butter.	Two 12-week periods, separated by an 8-week washout period	Group A showed a smaller increase in saturated fatty acids (SFA) and greater increase in MUFA intake when compared with the control.
Rosqvist et al. 2015 [3]	Sweden	Parallel	n = 57 (gender split not stated), overweight or obese	20–70	2 arm: (A) milk-fat globule membrane (MFGM) group—100 mL whipping cream (40%fat)/d, 100 mL fat-free milk (0.1% fat)/d and 1 scone/d (baked with wheat flour, water, sodium chloride and baking powder). (B) Fat-free milk, 100 mL, (0.1% fat)/d and 1 scone/d (baked with wheat flour, water, butter oil (98.7% fat), sodium chloride, baking powder and milk protein isolate).	8 weeks	Control diet increased TC, LDL, apolipoprotein B:apolipoprotein A-I ratio and non-HDL plasma lipids, whereas the MFGM diet did not. HDL, triglyceride, sitosterol, lathosterol, campesterol and proprotein convertase subtilisin/kexin type 9 concentrations and fatty acid compositions did not differ between groups.
Benatar et al. 2013 [50]	New Zealand	Parallel	n = 180 (54 male; 126 female), healthy volunteers	>18	3 arm: (A) increased dairy—an extra two-to-three servings per day, and to change to high-fat milk and dairy solids. (B) Habitual dairy intake remains unchanged. (C) Decreased dairy were asked to eliminate all possible sources of dairy.	1 month	No significant change in LDL or HDL, triglycerides, systolic or diastolic BP, C-reactive protein, glucose or insulin across groups. There was a small increase in weight in group A.
Nestel et al. 2013 [51]	Australia	Crossover	n = 12 (gender not specified) overweight or obese	40–70	3 arm: (A) low-fat dairy, 1% fat milk (400 mL/d) and 1% fat yoghurt (200 g/d). (B) Full-fat dairy (fermented), cheddar cheese (85 g/d) and full-cream yoghurt (three servings, 600 g/d). (C) Full-fat dairy (non-fermented), butter (30 g/d) and cream (70 mL/d) and small amounts of ice-cream.	Two 3-week periods, for group B + C (full fat diets). Group A diet (low fat) was consumed twice—between and at the end of the full-fat dairy dietary periods, for a duration of 2 weeks.	Lowest LDL and HDL concentrations were observed in group A, but plasma Triacylglycerol (TAG) concentrations did not differ significantly across the groups. Concentrations of plasma sphingomyelin and IL-6 were significantly higher after the non-fermented dairy diet (group C) than the low-fat dairy diet
Crichton et al. 2012 [52]	Australia	Crossover	n = 61 (18 male; 43 female), overweight or obese	18–75	2 arm: (A) 4 servings of reduced-fat dairy/d, 1 serving = 250 mL milk, 175–200 g yoghurt and 190 g custard. (B) Control—1 serving of dairy/d, reflecting habitual intake.	Two 6-month periods, no washout period	No significant changes in resting metabolic rate or total energy expenditure, systolic and diastolic BP, fasting blood glucose, TC, HDL or LDL, triglycerides or hs-CRP (high sensitivity C-reactive protein). Additionally no differences between groups for waist circumference (WC), body weight and fat mass.
Palacios et al. 2011 [53]	Puerto Rico	Parallel	n = 25 (5 male; 20 female), obese	22–50	3 arm: (A) 4 servings of dairy/d (low-fat milk, low-fat cheese and low-fat yoghurt), with a dairy-calcium intake goal of 1200–1300 mg/d. (B) calcium supplement (600 mg/d, calcium carbonate). (C) Control—habitual diet with placebo tablet.	21 weeks	No significant group effects were observed for anthropometric measurements or serum lipids such as TC, HDL, LDL and TAG levels. Although TAG levels decreased by 18% in group A (high dairy).
Stancliffe et al. 2011 [54]	USA	Parallel	n = 40 (19 male; 21 female), overweight or obese with metabolic syndrome	37.0 ± 9.9	2 arm: (A) low-dairy diet—0.5 servings/d and provided with 3 servings/d of non-dairy foods that is low-sodium luncheon meats, soy-based luncheon meat substitutes, packaged fruit cups, granola bars and peanut butter crackers. (B) adequate dairy diet—3.5 servings/d, of which 2/3 servings were milk and/or yoghurt.	12 weeks	Group A decreased malondialdehyde and oxidised LDL. Inflammatory markers were suppressed with intake of AD, with decreases in TNF-a (Tumor Necrosis Factor-a); decreases in IL-6 (Interleukin-6) and monocyte chemoattractant protein 1 and an increase in adiponectin. Group B exerted no effect on oxidative or inflammatory markers. Group A significantly reduced waist circumference and trunk fat but group B exerted no effects.
van Meijl and Mensink, 2010 [55]	Netherlands	Crossover	n = 35 (10 male; 25 female), overweight or obese	18–70	2 arm: (A) 500 mL low-fat milk and 150 g low-fat yoghurt per day. (B) Control—600 mL fruit juice and 43 g fruit biscuits per day	Two 8-week periods separated by a 2-week washout period	Plasma concentrations of TNF-a decreased, and soluble TNF-a receptor-1 increased after low-fat dairy consumption compared to the control. s-TNFR-2 also increased. Low-fat dairy consumption had no effect on IL-6, monocyte chemoattractant protein-1, intracellular adhesion molecule-1 and vascular cell adhesion molecule-1 concentrations. Lipid profiles were not analysed.
Zemel et al. 2010 [46]	USA	Crossover	n = 20 (14 male; 6 female), overweight or obese	Average—31 ± 10.3	2 arm: (A) dairy smoothie, 3 times/d, with non-fat dry milk as the protein source, and containing 350 mg calcium per smoothie. (B) Soy smoothie, 3 times/d, with soy protein isolate as the protein source and 50 mg calcium per smoothie.	Two 4-week periods separated by a 4-week washout period	Group A resulted in significant suppression of oxidative stress and lower inflammatory markers; tumour necrosis factor-a; IL-6; monocyte chemoattractant protein-1 and increased adiponectin. Group B exerted no significant effects. Lipid profiles were not analysed.
Wennersberg et al. 2009 [56]	Norway	Parallel	n = 121 (41 male; 80 female)	30–65	2 arm: (A) milk group, 3–5 portions of dairy/d. Portion = 200 g milk, 200–250 g yoghurt or sour milk, 75 g cream or crème fraıche, 15–40 g cheese, 3–10 g butter or butter-containing spreads, 50 mL cottage cheese, and ice-cream occasionally. (B) Control—habitual daily diet.	6 months	No significant differences between changes in body weight or body composition, BP, markers of inflammation, endothelial function, adiponectin or oxidative stress in group A and B. There was a modest unfavourable increase in serum TC concentrations in the group A.
van Meijl and Mensink, 2009 [57]	Netherlands	Crossover	n = 35 (10 males; 25 females), overweight or obese	18–70	2 arm: (A) 500 mL low-fat milk and 150 g low-fat yoghurt per day. (B) Control—600 mL fruit juice and 43 g fruit biscuits per day	Two 8-week periods separated by a 2-week washout period	In group A, systolic BP significantly decreased compared with the control, but diastolic BP did not reach significance. Decreases in HDL and apo A-1 concentrations were also observed in group A. Serum TC, LDL, apo B, TAG, non-esterified fatty acids, glucose, insulin, C-reactive protein and plasminogen activator inhibitor-1 remained unchanged.
Tricon et al. 2006 [58]	UK	Crossover	n = 32 males, healthy volunteers	34–60	2 arm: (A) 500 mL UHT full-fat milk, 12.5 g butter and 36.3 g cheese per day, naturally enriched with CLA (conjugated linoleic acid). (B) Control, 500 mL UHT full-fat milk, 12.5 g butter and 28 g cheese per day.	Two 6 week periods, separated by a 7 week washout period	Diet A did not significantly affect body weight, inflammatory markers, insulin, glucose, TAG, or TC, LDL and HDL cholesterol but resulted in a small increase in the LDL:HDL ratio. The modified dairy products changed LDL fatty acid composition but had no significant effect on LDL particle size or the susceptibility of LDL to oxidation.
Zemel et al. 2005 [59]	America	Parallel	Study 1.n = 34 (11 male; 23 female), Study 2. n = 29 (4 male; 25 female), obese	26–55	Study 1. (A) Dairy group—3 servings of dairy/d, at least one serving to be milk (B) Control—low dairy, 0–1 servings of low-fat dairy/d. Study 2. (A) Dairy group—3 servings of dairy/d, at least one serving to be milk and a 500-kcal/d deficit (B) Control—low dairy, 0–1 servings of low-fat dairy/d and 500 kcal/d deficit diet.	24 weeks (both studies)	Study 1. Body weight remained stable for both groups. Group A resulted in decreases in total body fat, trunk fat, insulin and BP and an increase in lean mass and no significant changes in the control group. Study 2. Both diets produced significant weight and fat loss. Weight and fat loss within group A were 2-fold higher, and loss of lean body mass significantly reduced compared with the control. There were no effects on circulating lipids in either group.
Tholstrup et al. 2004 [60]	Denmark	Crossover	n = 14 males, healthy volunteers	20–31	3 arms: (A) 1.5 L of whole milk/10 MJ (54 g of fat and 1779 mg calcium per 10 MJ). (B) Butter 64 g/10 MJ (54 g of fat and 10 mg calcium per 10 MJ) (C) 205 g of hard cheese, “Samsø”, 45% fat of dry weight, i.e., 26% fat/10 MJ (1989 mg calcium/10 MJ).	Three 3-week periods separated by a 4-week washout period	Fasting LDL concentrations were significantly higher after butter than cheese diet, with a borderline significant difference in TC after the experimental periods. Postprandial glucose showed a higher response after cheese diet compared to milk diet. No differences were found between groups for HDL, Very Low-Density Lipoprotein (VLDL), apo A-1 and apo B concentrations.