• Dairy;
  • metabolic syndrome;
  • milk;
  • systematic review


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

A growing body of observational research suggests that dairy consumption may have a beneficial effect on the metabolic syndrome (MetS). MetS is a clustering of cardiometabolic risk factors within an individual that carries with it an increased risk of developing cardiovascular disease. A systematic search of electronic databases identified cross-sectional studies (n = 10) and prospective cohort studies (n = 3) that assessed dairy intake in relation to MetS. The quality of the included studies was assessed based on study methodology, measurement and reporting of dietary intake, use of standardized MetS diagnostic criteria and statistical analysis. Dairy intake was inversely associated with incidence or prevalence of MetS in seven out of 13 studies. Three studies found no association between dairy and MetS. Three studies reported mixed relationships between specific dairy foods and MetS. The majority of studies suggested a potential benefit of dairy consumption on the risk of having MetS, but methodological differences, potential biases and other limitations in the studies conducted prevent conclusions to be drawn. Future randomized controlled trials are needed to confirm the effect of dairy consumption on MetS.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

An accumulating body of research suggests that the consumption of dairy foods may influence the development of metabolic syndrome (MetS). MetS is characterized by a clustering of cardiometabolic risk factors within an individual, namely abdominal obesity, hypertension, and dyslipidemia (hypertriglyceridemia and low high-density lipoprotein (HDL) cholesterol) (1). With an increasing number of risk factors clustering within an individual, the perturbation of each is greater (2) which together contributes to a 1.7-fold increase in the risk for developing CVD (3) and a 5-fold increase in the risk of developing type 2 diabetes mellitus (4). Modifiable lifestyle and dietary behaviours have been identified as playing a fundamental role in both the development and subsequent course of MetS (5). Weight loss to combat abdominal obesity is one of the key recommendations for management of MetS (5). Consistent with this recommendation, a growing body of cross-sectional and prospective research has shown associations between higher dairy consumption and lower prevalence of obesity, including abdominal obesity, and a lower prevalence of MetS. The best evidence for elucidating the effects of dairy intake on MetS will come from randomized, clinical intervention studies but there is a paucity of such evidence available at present.

Because of the growing importance placed on identifying dietary factors that may impact on MetS status and the increasing number of studies within this area, it is important to provide a synthesis of the evidence currently available, and to consider this evidence within the context of the methodological quality of the studies. A number of reviews have previously been published on the relationships between dairy intake and MetS which have addressed physiological mechanisms underlying effects of dairy on cardiometabolic health (6,7), and relationships between dairy and individual MetS components (8,9), but none have specifically used quality appraisal methods to identify potential methodological differences and limitations that might explain the apparent heterogeneity of findings. Therefore, the aims of this systematic review were to: (i) summarize and examine studies assessing the relationship between dairy intake and MetS; (ii) evaluate the quality of identified studies and (iii) use this appraisal to interrogate the heterogeneity in findings. Considerations for future research are also discussed.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

Literature selection

A search for relevant literature was undertaken on MEDLINE (1950 to July 2009), Web of Science (1968 to July 2009), CINAHL (1981 to July 2009) and EMBASE (1980 to July 2009), to identify observational studies which examined the association between dairy intake and MetS (prevalence or incidence), and for any randomized controlled trials investigating the effect of dairy intake on MetS.

The search strategy consisted of two key search terms: dairy (dairy, milk, cheese, yogurt as isolated words and in combination with food, product, intake, consumption), in combination with MetS (metabolic syndrome, insulin resistance syndrome, cardiometabolic health). Searches were limited to papers that involved adult participants and were written in English. The reference lists of all included studies and any relevant published reviews were inspected to identify additional papers for possible inclusion.

Inclusion and exclusion criteria

All cross-sectional, prospective cohort and intervention studies that reported an association between dairy intake (or intake of individual dairy products) and MetS were considered for inclusion. Letters, conference proceedings, abstracts of dissertations and reviews were not included but were screened for additional references that met the inclusion criteria. Based on perusal of the abstracts, studies were included if they reported on MetS as a global indicator of vascular risk, and excluded if they examined only one or more of the individual components of MetS but not overall MetS status. Studies were not limited by the criteria used to define MetS. Papers also had to include a quantitative measure of dairy food consumption, either as total dairy or intake of individual dairy products. If it was unclear whether the study met the inclusion criteria, the full text of the article was retrieved.

Data extraction and synthesis

Data were extracted on study design, participant characteristics, dairy intake measures, MetS status, statistical methods and results. This information was summarized to allow comparative analysis and quality assessment.

Quality assessment

The included studies were assessed on their quality according to the reporting of the (i) study design and method; (ii) study attrition; (iii) measurement of dairy intake; (iv) measurement of MetS and (v) statistical analysis. Sixteen criteria were chosen and adapted from a detailed checklist developed for observational longitudinal studies (10), including the six areas of potential study bias recommended to be used in any quality appraisal component of systematic reviews (11) (Appendix 1). Each criterion were scored as yes (1), no (0) or partly (0.5), based on information provided in the paper. The scores were totalled to give an overall indication of study quality.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

Included papers

This literature search yielded a total of 133 citations. After removing 53 duplicates, 80 papers remained. The title and abstract of these papers were screened to identify potentially relevant papers for full review. The full text was obtained for 26 papers deemed to be potentially relevant. Screening of reference lists identified an additional two relevant papers. From this total of 28 publications, 13 studies were eligible to be included.

The included studies consisted of 10 cross-sectional studies (12–21), and three prospective studies (22–24). Table 1 describes the design, cohort, country, sample characteristics and exclusion criteria of the included primary studies, grouped by study design and listed by quality appraisal score in descending order. The studies varied considerably in their classification of both MetS and dairy intake. The prevalence of MetS and the diagnostic criteria used in each study are shown in Table 2. A description of how dairy intake was measured and the type of dairy product examined (including quantity and fat content) is summarized in Table 3. Mean dairy intakes in each study sample (when provided) are also shown. Table 4 presents the results of all included studies, including confounding variables that were considered.

Table 1.  Description of studies reporting associations between dairy consumption and metabolic syndrome
StudyCohort FUD (prospective)Age (year)*SexnExclusion criteriaQuality score, /16
  • *

    Age is at baseline for prospective studies.

  • Quality score was calculated on 16 criteria based on the reporting of the study design and method, study attrition, measurement of dairy intake, measurement of MetS, and statistical analysis.

  • ARIC, Atherosclerosis Risk in Communities; CARDIA, Coronary Artery Risk Development in Young Adults; CVD, cardiovascular disease; DESIR, Data from a Epidemiological Study on Insulin Resistance Syndrome; F, female; FUD: follow-up duration; M, male; MetS, metabolic syndrome; MONICA, Monitoring of trends and determinants in cardiovascular disease; NHANES, National Health and Nutrition Examination Survey; TLGS, Tehran Lipid and Glucose Study; WHS, Women's Health Study.

Azadbakht et al. (12)TLGS, Tehran18–74M/F827CVD, diabetes, stroke, energy intakes deemed too low or high14
Ruidavets et al. (17)MONICA, France45–64M912Analyses repeated excluding subjects with high levels of physical activity, hypertension, dyslipidemia, current smokers, energy intake deemed too low13
Snijder et al. (19)Hoorn Study, Netherlands50–75M/F1896Analyses repeated excluding subjects with known diabetes, CVD, taking lipid-lowering or antihypertensive medications13
Beydoun et al. (13)NHANES, USA>18M/F4519Nil13
Yoo et al. (20)Bogalusa Heart Study, USA19–38M/F1181Energy intakes deemed too low or high12.5
Shin et al. (18)Recruited from a Cancer screening centre, Korea>30M5337Cancer, type 2 diabetes, myocardial infarction, heart attack, cerebral infarction12.5
Liu et al. (15)WHS, female health professionals, USA>45F10 066CVD, cancer, diabetes, never used postmenopausal hormones11
Mennen et al. (16)DESIR, recruited from 10 Health Centres, France30–64M/F4976Diabetes; analyses repeated excluding subjects with known CVD, hypertension, hyperlipidaemia9.5
Lawlor et al. (14)British Women's Heart and Health Study, UK60–79F4024Type 1 diabetes9.5
Analyses repeated excluding full fat milk drinkers
Elwood et al. (21)Caerphilly Cohort, UK45–59M2375Diabetes at baseline7.5
FUD: 20 years
Pereira et al. (23)CARDIA, USA FUD: 10 years18–30M/F3157Energy intakes deemed too low or high, pregnant at baseline or within 180 days of year 10 follow-up, medications that affect blood lipids, MetS at baseline13.5
Lutsey et al. (22)ARIC, USA45–64M/F9514MetS or CVD at baseline, energy intakes deemed too low or high13
FUD: 9 years
Snijder et al. (24)Hoorn Study, Netherlands50–75M/F1124MetS at baseline10.5
FUD: 6.4 years
Table 2.  Metabolic syndrome diagnostic criteria and prevalence
StudyMetS criteriaModifications to criteria% reported with MetS
  1. ATP III NCEP, Adult Treatment Panel III of the National Cholesterol Education Program; BMI, body mass index; BP, blood pressure; DBP, diastolic blood pressure; F, female; HDL, high-density lipoprotein; M, male; MetS, metabolic syndrome; SBP, systolic blood pressure; WHO, World Health Organization; WHR, waist-hip ratio.

Azadbakht et al. (12)ATP III NCEPNil24.2%
Ruidavets et al. (17)ATP III NCEPNil23.5%
Snijder et al. (19)ATP III NCEPNil30.4%
Beydoun et al. (13)ATP III NCEPNil25.8%
Yoo et al. (20)ATP III NCEPNil12.0%
Shin et al. (18)Modified ATP III NCEPAbdominal obesity: modified for Asian population: BMI ≥ 25 kg m−216.9%
Liu et al. (15)Modified ATP III NCEPAbdominal obesity: BMI ≥ 30 kg m−2; abnormal glucose homeostasis: type 2 diabetesNot provided
Mennen et al. (16)>2 of four factors: high fasting glucose, high serum triglycerides, high DBP, low HDL cholesterolHigh/low values defined as upper or lower sex-specific quartile of the distribution, cut-off values not providedF: 37.1%, M:27.1% Total: 32.2%
Lawlor et al. (14)Modified WHO definitionDiabetes, insulin resistance (highest quartile of HOMA score) or fasting glucose > 7.0 mmol L−1 and two of:  hypertension: ≥140/90 or taking anti-hypertensives;  dyslipidemia: triglycerides > 1.7 mmol L−1, or HDL-cholesterol < 1.0 mmol L−1;  abdominal obesity: BMI ≥ 30 kg m−2 or WHR >
Elwood et al. (21)Modified WHO definitionTwo or more of:  plasma insulin ≥ 163 mmol L−1 or plasma glucose ≥ 6.1 mmol L−1;  abdominal obesity: BMI ≥ 30 kg m−2;  dyslipidemia: triglycerides ≥ 3.25 mmol L−1, or HDL-cholesterol < 0.92 mmol L−1;  hypertension: SBP ≥ 166 mmHg or DBP ≥ 104 mmHg with self-reported hypertension.15.2% at baseline
Pereira et al. (23)>2 of four factors: obesity, abnormal glucose homeostasis, high BP, dyslipidemiaAbdominal obesity: BMI ≥ 30 kg m−2 or WHR ≥ 0.85 for women, ≥0.90 for men; Abnormal glucose homeostasis: fasting plasma insulin 20 uU mL−1, fasting glucose > 6.1 mmol L−1 or use of medications; Hypertension: ≥130/85 mmHg or use of anti-hypertensives; Dyslipidemia: HDL ≤ 35 mg dL−1 or triglycerides ≥ 200 mg dL−1.At follow-up: 17.6% BMI < 25: 9.3%, BMI ≥ 25: 37.6% (nil at baseline)
Lutsey et al. (22)American Heart Association guidelinesAny three of:  abdominal obesity: waist circumference: >102 cm in men, >88 cm in women;  dyslipidemia: triglycerides: ≥150 mg dL−1, or HDL-cholesterol < 40 mg dL−1 in men, <50 mg dL−1 in women;  hypertension: SBP ≥ 130 mmHg, DBP ≥ 85 mmHg, or use of anti-hypertensives;  abnormal glucose homeostasis: fasting glucose ≥ 100 mg dL−1, or use of medications.At follow-up: 39.8% (nil at baseline)
Snijder et al. (24)ATP III NCEPNilAt follow-up: 24.3% (nil at baseline)
Table 3.  Dairy intake measures and mean intakes of dairy in primary included studies
StudyMeasure usedIndividual dairy foods specifiedServing size definedFat content specifiedMean* intake of dairy
  • *

    Measurement of dairy intake and serving sizes varies significantly between countries. See ‘Assessment of dairy intake’ on page 5 for further description.

  • ANCOVA, analysis of covariance; ANOVA, analysis of variance; BMI, body mass index; CARDIA, Coronary Artery Risk Development in Young Adults; d, day; F, female; FFQ, food frequency questionnaire; M, male; WFR, weighed food record.

Azadbakht et al. (12)FFQMilk, yogurt, cheese Butter and ice-cream excluded1 serve milk = 1 cupDid not separate in analyses (but did control for total fat intake)Milk: 0.7 serves d−1
1 serve yogurt = 240 gYogurt: 1.06 serves d−1
1 serve cheese = 45 gCheese: 0.9 serves d−1
Ruidavets et al. (17)3-day food diaryMilk, cheeseNo: ‘grams per day’No175 g d−1
Snijder et al. (19)FFQMilk (low fat, skim, whole),1 serve liquid dairy = 150 gLow-fat dairy: ≤2% fat, high-fat dairy: >2% fatTotal dairy: 4.1 serves d−1
Yogurt (low fat, skim, whole),Milk: 0.7 serves d−1
Yogurt: 0.5 serves d−1
Dairy desserts (yogurt, curds, custard)Cheese: 1.2 serves d−1
1 serve solid dairy = 20 gDairy desserts: 0.9 serves d−1
Beydoun et al. (13)24 h recallMilk (low fat, skim, whole), yogurt, cheeseNo: ‘serves per day’Yes (milk only)Total dairy: All: 247.3 g d−1 (1.54 serves d−1)
M: 272.8 g d−1
F: 223.9 g d−1
Yoo et al. (20)Youth FFQLow-fat dairy: skim or 1% milk, non-fat or low-fat yogurt, cheeseNo: ‘serves per day’YesNot provided
High-fat dairy: whole or 2% milk, yogurt, cheese, cottage cheese, cream cheese, pudding, frozen yogurt, ice-cream, milkshake, frappe, whipped cream
Shin et al. (18)FFQNo: ‘dairy foods’No: ‘times per day, week or month’NoFrequency of dairy consumption: <2–3 times month−1: 32.2%
1–6 times week−1: 40.4%
>1 time d−1: 25.6%
Liu et al. (15)FFQLow-fat dairy: skim and low-fat milk, sherbet, yogurt, cottage/ ricotta cheeseNoYesDairy: not provided Calcium: 857 mg d−1
High-fat dairy: milk, cream, sour cream, ice-cream, cream cheese, other cheese
Mennen et al. (16)FFQMilk, cheeseNo: ‘portions per day’Not specified, assume whole fat based on discussionM: 77.7% had >1–4 portions d−1
F: 73.3% had >1–4 portions d−1
Lawlor et al. (14)FFQMilk (whole, semi-skimmed, skimmed, tinned)No: ‘does or does not drink milk’Yes‘Never drinks milk’: 2.8%
‘Drinks milk’: quantity not provided
Elwood et al. (21)FFQ 7-day WFR at baselineMilk, cheese, yogurt, butter, creamMilk: pintsNo<0.5 pint milk d−1: 34.5%
Total dairy: sum of calories from milk, cheese and yogurt0.5–1 pint milk d−1: 59.2%
Pereira et al. (23)CARDIA Diet History- average of baseline and year 7Milk and milk drinks (whole, reduced fat), cheese and sour cream (whole, reduced fat), butter, cream, yogurt, dairy dessertsNo: ‘times per week’Yes‘Times per week’ of total dairy (provided by race and weight status):  BMI < 25 kg m−2: blacks: 16.6, whites: 23.2;  BMI ≥ 25 kg m−2: blacks: 13.2, whites: 21.3
Lutsey et al. (22)FFQ at baseline and follow-upMilk, cheese, ice-cream, yogurt Low-fat dairyNo: ‘serves per day’Yes (for factor analysis)At baseline:  M: 1.70 serves d−1,  F: 1.58 serves d−1
No (for regression)
Snijder et al. (24)FFQ at baselineMilk (low fat, skim, whole) Yogurt (low fat, skim, whole) Dairy desserts (yogurt, curds, custard)1 serve liquid dairy = 150 gLow-fat dairy: ≤2% fat, High-fat dairy: >2% fatNot provided
1 serve solid dairy = 20 g
Table 4.  Results of all studies reporting associations between dairy consumption and MetS
StudyRange or mean intake of dairyMain findings with regard to dairy and MetSOther significant findingsAdditional adjustment
  1. OR/RR/HR shown are for fully adjusted models.

  2. BMI, body mass index; BP, blood pressure; CI, confidence interval; DBP, diastolic blood pressure; HDL, high density lipoprotein; HR, hazard ratio; MetS, metabolic syndrome; MI, myocardial infarction; OR, odds ratio; RR, relative risk; SBP, systolic blood pressure; T2D, type 2 diabetes; WC, waist circumference.

Azadbakht et al. (12)Dairy (milk, yogurt, cheese): <1.7 to ≥3.1 serves d−1Dairy intake inversely associated with MetS (OR 0.82, 95% CI 0.63–0.99).Dairy intake inversely associated with having: enlarged waist circumference (OR 0.80, 95% CI 0.63–0.98), hypertension (OR 0.83, 95% CI 0.69–0.99), elevated triglycerides (OR 0.90, 95% CI 0.74–1.10).Age, physical activity, smoking, BMI, energy intake, % of energy from fat, BP and oestrogen medication, calcium, protein
High intake: ≥3.1 serves d−1
Ruidavets et al. (17)Dairy (milk, cheese): above vs. below 175 g d−1 (median intake)Dairy intake inversely associated with MetS (OR 0.67, 95% CI 0.47–0.94).Likelihood of MetS decreased further with high intakes of dairy, fish and grains (intakes > median values for all foods).Age, centre, physical activity, education, smoking, dieting, alcohol, energy intake, diet quality, medications for hypertension and dyslipidemia
High intake: >175 g d−1
Snijder et al. (19)Total dairy, milk, yogurt, cheese, dairy desserts: serves d−1No association between dairy and MetS.Dairy intake associated with lower DBP, higher fasting glucose (both P < 0.05).Age, gender, energy intake, alcohol, fibre, smoking, physical activity, income, education, antihypertensive medications
High-fat dairy inversely associated with waist circumference; low-fat dairy positively associated with BMI, waist circumference, and fasting glucose (all P < 0.05).
High intake: ≥5.57 serves d−1Associations between dairy and MetS were not different between obese and non-obese subjects.Cheese intake associated with higher BMI; milk and dairy desserts associated with lower DBP and SBP, respectively (all P < 0.05).
Beydoun et al. (13)Dairy (milk, cheese, yogurt): serves d−1Yogurt intake inversely associated with MetS (OR 0.40, 95% CI 0.18–0.89).Whole fat milk and yogurt inversely associated with central obesity (P < 0.05).Age, gender, ethnicity, education, family income, energy intake, physical activity
Low-fat milk positively associated with central obesity (P < 0.05).
Cheese intake positively associated with MetS (OR 1.16, 95% CI 1.04–1.29).All milk inversely associated with SBP and DBP (P < 0.05).
Adequate dairy intake: ≥3 serves d−1Men only: total dairy positively associated with MetS (OR 1.08, 95% CI 1.00–1.17).Cheese positively associated with BMI, WC, BP, fasting glucose (all P < 0.05).
Yoo et al. (20)Low-fat dairy: serves d−1 High-fat dairy: serves d−1 High intake: not definedAdults with MetS consumed significantly less low-fat dairy (0.52 serves d−1), and more high-fat dairy (0.95 serves d−1) than those with no risk factors (0.73 serves d−1 of each) (P < 0.05).Differences in dietary intakes were greater in whites than African–Americans.Age, gender, ethnicity, energy intake, BMI, physical activity
Shin et al. (18)Dairy: <2–3 times month−1 to >1 time d−1 High intake: >1 time d−1No association between dairy intake and MetS.High intake of oily foods and seaweed, eating faster and frequent overeating all associated with increased MetS risk.Age, family history T2D, smoking, physical activity
Liu et al. (15)Total dairy: <0.91 to >3 serves d−1 High-fat dairy: <0.27 to >1.48 serves d−1 Low-fat dairy: <0.28 to >2.13 serves d−1 Total milk: <0.13 to >1.08 serves d−1Dairy intake inversely associated with MetS:  Total dairy (OR 0.66, 95% CI 0.55–0.80),  High-fat dairy (OR 0.71, 95% CI 0.58–0.87),  Low-fat dairy (OR 0.78, 95% CI 0.64–0.95),  Total milk (OR 0.85, 95% CI 0.71–1.02).Dietary calcium intake inversely associated with MetS: calcium (OR 0.73, 95% CI 0.61–0.88).Age, energy intake, smoking, physical activity, alcohol intake, multivitamin use, parental history of MI, dietary intakes of fat, cholesterol, protein, glycemic load
Mennen et al. (16)Dairy (milk, cheese): ≤1 to >4 portions d−1Dairy intake inversely associated with MetS in men (OR 0.63, 95% CI 0.40–0.99).Men: dairy intake inversely associated with serum triglycerides, fasting glucose, DBP (all P < 0.05).Age, energy intake, waist-hip ratio, alcohol intake, smoking
High intake: >4 portions d−1No relationship for women.Women: dairy intake inversely associated with DBP (P < 0.05).
Lawlor et al. (14)Milk intake: does drink milk vs. does not drink milkNon-milk drinkers had lower odds for MetS than milk drinkers (OR: 0.55, 95% CI: 0.33–0.94).Association remained significant when: full fat milk drinkers removed from analyses, stratified by body weight.Age, physical activity, smoking, diet characteristics, socioeconomic position (10 indicators)
High intake: not defined
Elwood et al. (21)Milk intake: little or none to >1 pint d−1 High intake: >1 pint d−1 (>568 mL d−1)At baseline:  Milk intake inversely associated with MetS (RR 0.38, 95% CI 0.18–0.78).  Highest quartile of dairy consumption (highest calories from dairy): inversely associated with MetS (OR: 0.40, 95% CI 0.20–0.79).No association between milk intake and incident diabetes.Age, energy intake, social class, smoking
Total dairy: sum of calories from dairy (milk, cheese, yogurt), quantities not specified
Pereira et al. (23)Total dairy: <10 to ≥35 times week−1Dairy intake inversely associated with MetS in adults overweight (BMI ≥ 25 kg m−2) at baseline (OR 0.28, 95% CI 0.14–0.58).Incidence of obesity, hypertension, abnormal glucose homeostasis significantly reduced in overweight individuals (all P < 0.01) who consumed ≥35 serves of dairy per week.Age, gender, race, energy intake, study centre, baseline BMI, education, alcohol, smoking, physical activity, dietary factors
No relationship for normal weight adults.
High intake: ≥5 times d−1Reduced odds (20–60%) of MetS for each daily serve of specified types of dairy (including total dairy) among overweight adults.
Lutsey et al. (22)Total dairy: 0.28 to 3.30 serves d−1Dairy intake inversely associated with MetS (HR 0.87, 95% CI 0.77–0.98).Western dietary pattern (high intake of refined grains, fried foods, red and processed meat) positively associated with MetS (HR 1.18, 95% CI 1.03–1.37).Age, gender, race, education, centre, energy intake, smoking, physical activity, intakes of meat, grains, fruit and vegetables
High intake: 3.30 serves d−1Individual dairy foods not associated with MetS.
Snijder et al. (24)Total dairy (milk, cheese, yogurt, dairy desserts): <2.97 to >5.75 serves d−1Dairy intake not associated with risk of developing MetS, changes in body composition or metabolic variables.High total dairy intake associated with increased BMI, weight and WC, decreased HDL-cholesterol in those with BMI < 25 kg m−2 at baseline (statistics not provided).Age, gender, energy intake, alcohol, smoking, physical activity
High intake: >5.75 serves d−1Low-fat dairy, high fat dairy, milk, cheese, yogurt, dairy dessert consumption not associated with body composition or metabolic variables.

Main findings

Results from cross-sectional studies

Of the 10 cross-sectional studies, five found inverse associations between MetS and intake of milk (21), milk and cheese (16,17), milk, cheese and yogurt (12), and total milk, low-fat dairy, high-fat dairy, and total dairy (15). One study found that American adults with MetS consumed significantly less low-fat dairy, but more high-fat dairy than those without MetS (20). A reduced risk for MetS among non-milk drinkers compared with milk drinkers was found in British women (14). This association remained significant when full fat milk drinkers were excluded from the analyses. Two studies found no association between dairy intake and MetS (18,19). The final study reported mixed findings: total dairy increased the risk for MetS in men only, but when individual dairy foods were analyzed, the syndrome was inversely associated with yogurt intake and positively associated with cheese intake (13).

Study quality did not appear to be related to study findings. The studies with the highest-quality rating (14/16, 12) and lowest-quality rating (7.5/16, 21), both found inverse associations between MetS and dairy intakes, suggesting that the findings of the studies were not influenced by the quality of their design and execution.

Results from prospective studies

A lower incidence of MetS with a higher intake of milk, cheese, yogurt and ice-cream was found in the Atherosclerosis Risk in Communities study (22). This finding supports the Coronary Artery Risk Development in Young Adults results. Pereira et al. (23) reported that dairy intake (milk, cheese, sour and regular cream, butter, yogurt and desserts) was inversely associated with MetS in young adults who were overweight at baseline, but there was no relationship in adults who were not overweight. Additionally, the odds of having MetS among overweight individuals were significantly reduced with an increasing number of daily serves of each different dairy product, including total dairy. Finally, the Hoorn Study (24) found no association between dairy consumption (milk, cheese, yogurt, dairy desserts) and risk of developing MetS and separate analyses of low-fat and high-fat dairy revealed that neither was associated with body composition or metabolic variables.

Two of the prospective studies that found inverse associations between dairy intake and MetS scored highly in the quality assessment, scoring 13.5 (23) and 13 (22) out of a possible total of 16. The lowest scoring study (10.5/16, 24) found no association between dairy intake and MetS incidence.

Methodological appraisal

The complete scoring of the quality of the included studies is presented in Appendix 2. The scores ranged from 7.5 to 14 out of a possible total of 16 points. Overall, there were no obvious differences in quality scores between those studies that found an inverse association between dairy intake and MetS, and those that found either a positive or no association.

The majority of papers provided sufficiently detailed descriptions of the baseline characteristics of the sample, the specific number of participants completing each time point, the criteria used to define MetS and the statistical analyses used. All papers addressed potential confounding variables and most reported a measure of association between dairy intake and MetS. Similarly, nearly all included studies acknowledged limitations of their designs.

The reporting of sample recruitment and specific inclusion and exclusion criteria were variable. A number of papers provided only a brief description of the study design but referred the reader to an additional paper for this information. Few papers reported any analyses conducted to determine differences between participants lost to follow-up and those that completed the study. How missing data were dealt with was also infrequently reported. All prospective studies excluded participants with MetS at baseline and provided a measure of incident MetS at follow-up (22–24). The criteria regarding the description of the predictor variable (dairy intake) were not well met, with only two papers (12,19) meeting four of the five appraisal criteria.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

This review systematically identified and reviewed the evidence for an association between dairy intake and MetS. Cross-sectionally, five out of 10 studies found an inverse association between dairy food or milk intake and MetS prevalence (12,15–17,21). Individuals with MetS were found to consume less low-fat dairy and more high-fat dairy than those without in one study (20). Prospectively, inverse associations between incidence of MetS and dairy consumption were found in two out of three studies, in both normal weight adults (22) and in adults overweight at baseline (23). Three studies suggest no association between dairy and MetS prevalence (18,19) or incidence (24). A positive relationship between dairy and MetS prevalence in men was found in one study (13), while non-milk drinkers had lower odds of having MetS in another study (14).

The methodological appraisal of the included studies highlighted differences and potential sources of bias that make comparison between studies difficult, may contribute to the heterogeneity in findings, and confound the ability to draw conclusions regarding optimal dairy intake and MetS. The widespread variation and in some cases, lack of detailed reporting with regard to dairy intake is one of the biggest limitations for most studies within the literature. The varying single and multiple combinations of different dairy foods, and therefore varying nutrient composition examined by each study may contribute to the differences in findings. Seven studies examined at least one individual dairy product, in addition to total dairy intake (13,15,19,21–24). Milk, cheese and yogurt were the most frequently included dairy foods. The remaining studies reported on various combinations of dairy foods, including: milk only (14), milk and cheese (16,17), milk, yogurt and cheese (12), or high-fat dairy and low-fat dairy (over 10 different foods) (20). One study did not define what foods were included within ‘dairy food’(18).

Few studies directly compared MetS outcomes related to low-fat dairy vs. high-fat dairy. Eight of the 13 studies mentioned the fat content of the dairy included, but only six performed separate analyses for low- and whole-fat dairy, either as individual products or the sum of a number of products (13,15,19,20,23,24). As full-fat dairy foods contribute to the overall dietary intake of saturated fat (25,26), it may be hypothesized that the fat content of dairy consumed may be an important factor in any relationship between dairy consumption and MetS status. A thorough review on the relationship between dietary fat and body weight, diabetes and MetS (27) concluded that both total and saturated fat intakes increase the risk of having MetS. Conversely, monounsaturated fatty acids and polyunsaturated fatty acids may have a risk lowering effect (27). This is consistent with the findings of Yoo et al. (20) who demonstrated a positive association between high intakes of high fat dairy and MetS prevalence in young adults. A small number of studies in the present review examined individual dairy foods, with fat content stipulated, and individual components of the syndrome (13,15,19,22–24), but findings were discrepant and did not consistently find that intakes of higher-fat products were associated with poorer metabolic outcomes. The inconsistent findings highlight the complex nature of the relationships between individual dairy foods, dairy fat, total dairy intake, related nutrients and cardiometabolic health.

The lack of distinction between low- and high-fat dairy in the literature was similarly identified in two recent reviews examining the effects of dairy or milk consumption on CVD or coronary heart disease (CHD) risk (26,28). Both concluded that there is no clear evidence to suggest that dairy consumption is associated with a higher risk of CVD or CHD, despite the saturated fat content of dairy. This is supported by a review of CHD risk and dietary fat, in which Skeaff and Miller (29) concluded that there is unlikely to be a direct relation between total fat intake and risk of CHD. Metabolic risk factors, such as those comprising MetS, may be reduced through the beneficial effects of other dairy derived nutrients, including calcium, whey protein, monounsaturated fatty acid and polyunsaturated fatty acid (28,29).

In addition to inadequacies in describing the type and fat content of dairy assessed, dairy serving sizes were poorly defined. As shown in Table 4, a high dairy intake was defined variably in terms of ‘serves per day’, in ‘pints per day’, in ‘portions per day’ and in ‘times per day’. Unless an absolute serving size has been defined, results are not comparable between populations because of this variability. Only four studies (described in five papers) provided enough information on dairy intakes and serving sizes to be able to determine the quantity of dairy associated with the observed outcome (12,17,19,21,24). Of these, intakes of greater than 175 g per day of milk or cheese (17), one pint (568 mL) of milk (21) and three serves of either one cup of milk, 240 g of yogurt, or 45 g of cheese (12) have been inversely associated with MetS prevalence. It is therefore inappropriate to compare results across study cohorts because of discrepancies in serving sizes and the number and type of dairy foods included within these serves. Therefore, definitive conclusions cannot be drawn.

One of the main limitations within the prospective dietary research was the lack of regular assessment of dairy consumption over time. Whole dietary patterns including dairy consumption are unlikely to remain stable over time, particularly with the widespread increase in the availability of low-fat dairy products over the last 25 years. In addition, more pronounced health differences in the elderly may be expected as they have had lifetime exposure to other risk factors in addition to diet (19). With the exception of Coronary Artery Risk Development in Young Adults (23), Tehran Lipid and Glucose Study (12) and the Bogalusa Heart Study (20), young adults between 18 and 30 years have largely been ignored in the diet–MetS research to date.

The different number of MetS criteria required for a diagnosis in the included studies adds to the complexity in comparing findings. Ten of the 13 studies required at least three criteria to be met; however, three studies required only two or more (16,21,23). The difficulty in directly comparing prevalence of MetS across studies, countries and ethnicities should be alleviated in the future following the recent joint scientific statement from a number of international organizations (4), providing an agreed-upon criterion for clinical diagnosis (three out of five risk factors, maintaining the Adult Treatment Panel III of the National Cholesterol Education Program cut-offs).

There are a number of other limitations in the dairy–MetS literature. To date, randomized controlled trials have been limited to dairy foods in relation to individual components of MetS. One recent study by Wennersberg et al. (30) compared changes in the individual components of MetS in a control group (low dairy intake) and a dairy group (high dairy intake) for six months. Although each participant had to have at least two of the Adult Treatment Panel III of the National Cholesterol Education Program criteria to be included in the study, the proportion of subjects with MetS at baseline and at study completion was not provided, hence the study was excluded from this review. All included studies were observational in nature, and therefore do not allow cause–effect conclusions to be drawn. Observational research does not permit the control of all diet and lifestyle factors that may influence the development of MetS. Secondly, the under-reporting of dietary intakes by overweight people and difficulties associated with the use of self-reported dietary intake are acknowledged sources of bias in any nutrition research. Only five studies excluded subjects who reported energy intakes deemed to be too high or low (12,17,20,22,23), while the remaining studies did not report on how they dealt with under- or over-reporters of dietary intake. Low-fat dairy consumption may reflect a healthy diet or lifestyle that impacts upon cardiometabolic health. The majority of studies consistently controlled for lifestyle factors, including smoking, physical activity, alcohol intake and other dietary indicators. Fewer studies controlled for socioeconomic indicators such as education or income which could introduce considerable bias to the study outcomes.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

While the majority of the existing literature suggests a benefit of dairy consumption on odds of having MetS, the data remain inconclusive. Evidence from prospective cohort studies is lacking because of the small number of studies, and no randomized controlled trials have been conducted to date. Introduced biases from the failure to suitably control for confounding variables, limited information regarding dairy intake (including type, quantity and fat content) and the use of different MetS diagnostic criteria are all limitations in the present research which restricts the usefulness of the results. These factors make it difficult to directly compare studies, to explain the inconsistent findings and to make definitive conclusions regarding the relationship between dairy intake and MetS.

Evidence from high-quality randomized controlled trials are needed to examine the effects of long-term consumption of low-fat dairy, within a healthy eating pattern based on recommended dietary guidelines, on MetS risk. Future studies should use standardized MetS diagnostic criteria and control for all known potential confounding variables. Detailed, regular assessment and recording of dairy intake is needed to distinguish high-fat dairy from low-fat dairy, and to report dairy intake in grams per day to account for global differences in serving sizes.

Conflict of interest statement

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

The authors declare that they have no conflict of interest.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices
  • 1
    Alberti KG, Zimmet P, Shaw J. Metabolic syndrome – a new world-wide definition. A consensus statement from the International Diabetes Federation. Diabet Med 2006; 23: 469480.
  • 2
    Pettman TL, Buckley JD, Coates AM, Misan GMH, Petkov J, Howe PRC. Prevalence and interrelationships between cardio-metabolic risk factors in abdominally obese individuals. Metab Syndr Relat Disord 2009; 7: 3136.
  • 3
    Galassi A, Reynolds K, He J. Metabolic syndrome and risk of cardiovascular disease: a meta-analysis. Am J Med 2006; 119: 812819.
  • 4
    Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, Fruchart JC, James WP, Loria CM, Smith SC Jr. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009; 120: 16401645.
  • 5
    Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ, Smith JSC, Spertus JA, Costa F. Diagnosis and management of the metabolic syndrome. An American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Executive summary. Cardiol Rev 2005; 13: 322327.
  • 6
    Tremblay A, Gilbert JA. Milk products, insulin resistance syndrome and type 2 diabetes. J Am Coll Nutr 2009; 28: 91S102S.
  • 7
    van Meijl LEC, Vrolix R, Mensink RP. Dairy product consumption and the metabolic syndrome. Nutr Res Rev 2008; 21: 148157.
  • 8
    Scholz-Ahrens KE, Schrezenmeir J. Milk minerals and the metabolic syndrome. Int Dairy J 2006; 16: 13991407.
  • 9
    Pfeuffer M, Schrezenmeir J. Milk and the metabolic syndrome. Obes Rev 2006; 8: 109118.
  • 10
    Tooth L, Ware R, Bain C, Purdie DM, Dobson A. Quality of reporting of observational longitudinal research. Am J Epidemiol 2005; 161: 280288.
  • 11
    Hayden JA, Cote P, Bombardier C. Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med 2006; 144: 427437.
  • 12
    Azadbakht L, Mirmiran P, Esmaillzadeh A, Azizi F. Dairy consumption is inversely associated with the prevalence of the metabolic syndrome in Tehranian adults. Am J Clin Nutr 2005; 82: 523530.
  • 13
    Beydoun MA, Gary TL, Caballero BH, Lawrence RS, Cheskin LJ, Wang Y. Ethnic differences in dairy and related nutrient consumption among US adults and their association with obesity, central obesity, and the metabolic syndrome. Am J Clin Nutr 2008; 87: 19141925.
  • 14
    Lawlor DA, Ebrahim S, Timpson N, Smith GD. Avoiding milk is associated with a reduced risk of insulin resistance and the metabolic syndrome: findings from the British Women's Heart and Health Study. Diabet Med 2005; 22: 808811.
  • 15
    Liu S, Song Y, Ford ES, Manson JE, Buring JE, Ridker PM. Dietary calcium, vitamin D, and the prevalence of metabolic syndrome in middle-aged and older U.S. women. Diabetes Care 2005; 28: 29262932.
  • 16
    Mennen LI, Lafay L, Feskens EJM, Novak M, Lepinay P, Balkau B. Possible protective effect of bread and dairy products on the risk of metabolic syndrome. Nutr Res 2000; 20: 335347.
  • 17
    Ruidavets JB, Bongard V, Dallongeville J, Arveiler D, Ducimetiere P, Perret B, Simon C, Amouyel P, Ferrieres J. High consumptions of grain, fish, dairy products and combinations of these are associated with a low prevalence of metabolic syndrome. J Epidemiol Community Health 2007; 61: 810817.
  • 18
    Shin A, Lim SY, Sung J, Shin HR, Kim J. Dietary intake, eating habits, and metabolic syndrome in Korean men. J Am Diet Assoc 2009; 109: 633640.
  • 19
    Snijder MB, van der Heijden AA, van Dam RM, Stehouwer CD, Hiddink GJ, Nijpels G, Heine RJ, Bouter LM, Dekker JM. Is higher dairy consumption associated with lower body weight and fewer metabolic disturbances? The Hoorn Study. Am J Clin Nutr 2007; 85: 989995.
  • 20
    Yoo S, Nicklas T, Baranowski T, Zakeri IF, Yang SJ, Srinivasan SR, Berenson GS. Comparison of dietary intakes associated with metabolic syndrome risk factors in young adults: the Bogalusa Heart Study. Am J Clin Nutr 2004; 80: 841848.
  • 21
    Elwood PC, Pickering JE, Fehily AM. Milk and dairy consumption, diabetes and the metabolic syndrome: the Caerphilly prospective study. J Epidemiol Community Health 2007; 61: 695698.
  • 22
    Lutsey PL, Steffen LM, Stevens J. Dietary intake and the development of the metabolic syndrome. The Atherosclerosis Risk in Communities Study. Circulation 2008; 117: 754761.
  • 23
    Pereira MA, Jacobs DR Jr, Van Horn L, Slattery ML, Kartashov AI, Ludwig DS. Dairy consumption, obesity, and the insulin resistance syndrome in young adults: the CARDIA study. JAMA 2002; 287: 20812089.
  • 24
    Snijder MB, van Dam RM, Stehouwer CD, Hiddink GJ, Heine RJ, Dekker JM. A prospective study of dairy consumption in relation to changes in metabolic risk factors: the Hoorn Study. Obesity (Silver Spring) 2008; 16: 706709.
  • 25
    Hu FB, Stampfer MJ, Manson JE, Ascherio A, Colditz GA, Speizer FE, Hennekens CH, Willett WC. Dietary saturated fats and their food sources in relation to the risk of coronary heart disease in women. Am J Clin Nutr 1999; 70: 10011008.
  • 26
    German JB, Gibson RA, Krauss RM, Nestel P, Lamarche B, van Staveren WA, Steijns JM, de Groot LC, Lock AL, Destaillats F. A reappraisal of the impact of dairy foods and milk fat on cardiovascular disease risk. Eur J Nutr 2009; 48: 191203.
  • 27
    Melanson EL, Astrup A, Donahoo WT. The relationship between dietary fat and fatty acid intake and body weight, diabetes, and the metabolic syndrome. Ann Nutr Metab 2009; 55: 229243.
  • 28
    Gibson RA, Makrides M, Smithers LG, Voevodin M, Sinclair AJ. The effect of dairy foods on CHD: a systematic review of prospective cohort studies. Br J Nutr 2009; 102: 12671275.
  • 29
    Skeaff CM, Miller J. Dietary fat and coronary heart disease: summary of evidence from prospective cohort and randomised controlled trials. Ann Nutr Metab 2009; 55: 173201.
  • 30
    Wennersberg MH, Smedman A, Turpeinen AM, Retterstol K, Tengblad S, Lipre E, Aro A, Mutanen P, Seljeflot I, Basu S, Pedersen JI, Mutanen M, Vessby B. Dairy products and metabolic effects in overweight men and women: results from a 6-mo intervention study. Am J Clin Nutr 2009; 90: 960968.


  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Conflict of interest statement
  9. References
  10. Appendices

Appendix 1

Checklist for methodological assessment of cross-sectional and prospective studies included in review, adapted from Hayden et al. (11) and Tooth et al. (10)
Study design & method
 1. Sampling, recruitment methods, time and place of recruitment adequately described.
 2. Inclusion and exclusion criteria adequately described.
 3. Baseline characteristics of study sample adequately described (number of participants, age, gender, health status).
Study attrition
Cross-sectionalProspectiveRandomised controlled trials
 4. Justification or explanation for the number of participants.4. Specification of numbers at each follow-up period and duration of follow-up.4. Specification of numbers at baseline and study completion; reasons given for withdrawals.
 5. Explanation of how missing data was dealt with.5. Analyses conducted to determine whether participants lost to follow-up do not significantly differ in key characteristics/outcomes from those who completed the study.5. Analyses conducted to determine whether participants lost to follow-up do not significantly differ in key characteristics/outcomes from those who completed the study.
Measurement of predictor variable
 6. Method used to assess dietary intake adequately described and reference provided.
 7. Reliability and validity of measurement tool(s) mentioned.
 8. Mean dairy intake adequately described in sample (including type, quantity).
 9. Dairy associated with outcome measure of interest clearly described (including type, fat content).
10. Quantity of dairy associated with outcome measure clearly described (number of servings and specified serving size, or amount in g/mL per day).
Measurement of outcome variable
11. Criteria used to define metabolic syndrome provided and any modifications to criteria adequately described.
12. Number of participants with metabolic syndrome provided at baseline (and follow-up for prospective studies).
13. Potential confounders mentioned and accounted for in analyses.
14. Specific type of analyses adequately described.
15. Measure of association provided with confidence intervals.
16. The impact of biases/ limitations to study assessed qualitatively.

Appendix 2

Methodological appraisal scores for the included studies in the primary review
  1. 1 = yes, 0 = no, 0.5 = partly.

  2. T, total.

Azadbakht et al. (12)0.511101110.5111111114.0
Ruidavets et al. (17)10111100.50.5111111113.0
Snijder et al. (19)0011110.511111110.5113.0
Beydoun et al. (13)0.501110.50.5110.511111113.0
Yoo et al. (20)0.50.511111010.511110112.5
Shin et al. (18)111110.5100011111112.5
Liu et al. (15)00.511011010.510111111.0
Mennen et al. (16)0.50.51100.50.500.50.50.5111109.5
Lawlor et al. (14)0.50.51100.500101111109.5
Elwood et al. (21)00.50.5100.5000.50.510.510.5107.5
Pereira et al. (23)0.51110110.510.511111113.5
Lutsey et al. (22)0.5111010.
Snijder et al. (24)001110.500111110.50.5110.5