NutritionMagazine, No.4, 2011, Vol.24

Conclusion drawn from the Oslo Health Study

Frequency of cheese consumption inversely associated with metabolic syndrome

In a cross-sectional study of 18,000 men and women in Norway, an inverse correlation was found of frequency of cheese consumption with the metabolic syndrome, the values of separate characteristics of the metabolic syndrome, the number of these characteristics and body mass index. The associations remained significant after correction for a large number of possible confounders.

The Oslo Health Study is a cross-section study executed in 2000 and 2001 in a sample of 18,770 inhabitants of the province of Oslo born in 1970, 1960, 1955, 1940/41 or 1924/25. The men/women ratio in the study population was 45:55. Within the study framework, variables examined included dietary patterns and anthropometric measures; blood samples were also collected. Earlier in 2011, the epidemiologists Arne Høstmark and Sissel Tomten (University of Oslo) published the outcomes of analyses of the correlation between cheese consumption and prevalence of metabolic syndrome (1).
Metabolic syndrome is a prediabetic condition characterized by a cluster of cardiovascular risk factors including central obesity, raised blood glucose levels, elevated blood pressure and dyslipidaemia. Definitions of metabolic syndrome vary in the literature. Table 1 shows the cut-off values for the risk factors used for analyses of the Oslo Health Study data.
In a prospective study in Wales published in 2007, an inverse correlation was found between consumption of milk and dairy products and prevention of metabolic syndrome (2). Høstmark and co-workers published in 2009 that cheese consumption was found in the Oslo Health Study to correlate positively with serum HL-cholesterol and inversely with serum triacylglycerol (triglycerides) (3). Høstmark and Tomten write in their 2011 paper that these observations raise the question whether the study outcomes are in keeping with the hypothesis that cheese consumption correlates inversely with the complete metabolic syndrome.

Associations with separate characteristics of the metabolic syndrome
Regression analysis in either the presence or the absence of the complete metabolic syndrome as dependent variable revealed that there is a slight but highly significant association (P<0.001) between frequency of cheese consumption and the presence of metabolic syndrome. Conversely, patients with metabolic syndrome report a significantly lower frequency of cheese consumption than people without that syndrome (P=0.005 for men, P<0.001 for women).
Development of metabolic syndrome in patients is a slow process. Obviously, on the road to the fully fledged metabolic syndrome some features are manifested earlier than other ones. An increasing number of manifest features may be predictive of development of the complete metabolic syndrome. Therefore, Høstmark and Somten have developed two scoring systems for following the development of metabolic syndrome over time: the MetSRisk score and the SumRisk score. In the MetSRisk score each of the diverse components of metabolic syndrome have the same weight while SumRisk is the score for the number of features of metabolic syndrome present.
Figure 1 shows the correlation between frequency of cheese consumption and MetSRisk score. For almost all age groups of both sexes cheese consumption had a significant inverse correlation with MetSRisk score (P<0.001) with the exception of 75-year old men (P = 0.434). Comparison of the group with the highest to that with the lowest cheese consumption showed that the risk reduction was 10.5% for young men, 9.4% for young women, 15.0% for middle-aged men, 14.4% for middle-aged women, 9.4% for older men, 20.7% for older women and 15.5% for the oldest women.

Cheese consumption correlates inversely with BMI
Figure 2 shows the correlation between cheese consumption and SumRisk score. Broadly speaking, there seems to be an inverse correlation, which was significant for middle-aged and older men and for middle-aged, older and the oldest women. Regression analysis revealed that, after correction for possible confounders, frequency of cheese consumption correlated significantly and inversely with triacylglycerol (triglyceride) level, systolic and diastolic blood pressure, waist circumference and body mass index (BMI) and positively with serum HDL-cholesterol. The possible confounders corrected for were age, sex, physical activity level, educational attainment, place of birth, time after the latest meal when the blood sample was taken, and consumption of fruit, fruit drinks, fat fish, coffee and alcohol.
Høstmark and Tomten also investigated in the Oslo Health Study whether there is a correlation between frequency of cheese consumption and body weight. They found a correlation indeed: participants who consumed more cheese had a lower body weight than those who ate less cheese. That correlation was highly significant for men and women in the three youngest age categories (P = 0.001). An inverse correlation between frequency of cheese consumption and BMI was also found in all age and sex categories (P = 0.02) except for the oldest women.

Prevalence of metabolic syndrome possibly underestimated
In the discussion at the end of their paper the authors identify as a limitation of their study the fact that no blood samples were taken from fasting participants. That did not allow to include fasting blood glucose levels (>5.6 mmol/l) as a metabolic syndrome feature. That omission may have led to underestimation of the prevalence of metabolic syndrome. No distinction was made either between different cheese types.
The researchers underline that the correlation between frequency of cheese consumption and the risk factors for metabolic syndrome was fairly robust in the various sex-age categories after correction for a large number of variables known to be associated with metabolic syndrome. Obviously, they cannot preclude the possibility of residual confounding of the association by other nutrition and lifestyle factors. However, Høstmark and Tomten state, it is hard to imagine how their results could be attributable to systematic errors.
The inverse association found between frequency of cheese consumption and prevalence of metabolic syndrome is in line with results of studies that did not find an unfavourable effect of cheese consumption on serum lipid levels (4, 5). In two Swedish studies (6, 7) an inverse correlation was found between milk fat intake and levels of cardiovascular risk factors. A meta-analysis of prospective studies (8) revealed that, generally speaking, a high intake of dairy fat is not a risk factor for coronary heart disease. In a Norwegian case-control study (9) intake of dairy fat was found to correlate inversely with risk of myocardial infarction. In the Rotterdam Study, consumption of dairy products proved to correlate inversely with blood pressure (10).
Høstmark and Tomten conclude that the results of the Oslo Health Study fit in with the hypothesis that cheese consumption can protect against the development of metabolic syndrome.

Jan Blom

1. A.T. Høstmark, S.E. Tomten (2011) The Oslo Health Study: cheese intake was negatively associated with the metabolic syndrome. J. Am. Coll. Nutr. 30: 182—190.
2. P.C. Elwood, J.E. Pickering, A.M. Fehily (2007) Milk and dairy consumption, diabetes and the metabolic syndrome: the Caerphilly prospective study. J. Epidem. Comm. Health 61: 695—698.
3. A.T. Høstmark, A. Haug, S.E. Tomten et al. (2009) Serum HDL cholesterol was positively associated with cheese intake in the Oslo Health Study. J. Food Lipids 16: 89—102.
4. D.M. Colquhoun, S. Somerset, K. Irish, L.M. Leontjew (2003) Cheese added to a low fat diet does not affect serum lipids. Asia Pac. J. Clin. Nutr. 12 (suppl.): S65.
5. D. O'Callaghan, A. Stanton, S. Rafferty et al. (1996) Are butter and cheese rich in monounsaturates beneficial in hyperlipidemic patients? J. Cardiovasc. Risk 3: 441—445.
6. A.E. Smedman, I.B. Gustafsson, L.G. Berglund, B.O. Vessby (1999) Pentadecanoic acid in serum as a marker for intake of milk fat: relations between intake of milk fat and metabolic risk factors. Am. J. Clin. Nutr. 69: 22—29.
7. E. Warensko, J.H. Jansson, L. Berglund et al. (2004) Estimated intake of milk fat is negatively associated with cardiovascular risk factors and does not increase the risk of a first acute myocardial infarction: a prospective case-control study. Br. J. Nutr. 91: 635—642.
8. P.C. Elwood, J.E. Pickering, J. Hughes et al. (2004) Milk drinking, ischaemic heart disease and ischaemic stroke II. Evidence from cohort studies. Eur. J. Clin. Nutr. 58: 718—724.
9. A.S. Biong, H.M. Rebnord, R.L. Fimreite et al. (2008) Intake of dairy fat and dairy products, and risk of myocardial infarction. Int. J. Food Sci. Nutr. 59: 155—165.
10. M.F. Engberink, M.A. Hendriksen, E.G. Schouten et al. (2009) Inverse association between dairy intake and hypertension: the Rotterdam Study. Am. J. Clin. Nutr. 89: 1877—1883.

PRINT - BACK