Easy Diagnosis  
Health Controversies

Letter to the editor of Science, published 22 Febr. 2002

IN HIS LETTER ABOUT THE ARTICLE “THE soft science of dietary fat” (News Focus, G. Taubes, 30 Mar. 2001, p. 2536), Scott M. Grundy says that saturated fatty acids (SFA) are the main dietary cause of coronary heart disease (CHD) (“Dietary fat: at the heart of the matter,” 3 Aug., p. 801), and he cites two reviews in support (1, 2).

In one of the reviews, there are no references (1); in the other, of which Grundy is a co-author, most of the references do not appear to be supportive of his statement (2). For instance, the authors say that “populations consuming diets high in saturated fats have relatively high levels of serum cholesterol and carry a high prevalence of coronary heart disease” (2, p. 34), referring to 12 studies (3–14). In the eight cohort studies (3–10), only one had examined the association between SFA and serum cholesterol (10), five found no increased SFA consumption among CHD patients (3, 4, 6, 9, 10), and one found a smaller consumption (7). In addition, three of the 12 studies were reports from a project comparing the incidence of CHD in native Japanese living in Japan with Japanese-Americans living in the United States (12–14). Although it is correct that the Japanese-Americans, on average, had higher cholesterol, ate more saturated fat, and had a higher incidence of CHD, the determining factor for heart disease was not their cholesterol levels or their diets, but how acculturated they were to Western culture (13).

Grundy also writes in his letter that lowering serum LDL cholesterol by dietary means reduces CHD risk. But the study he cites did not specifically address this question (15), and more to the point, meta-analyses of all controlled and randomized trials that have used modification of dietary fat as the only type of intervention have shown that neither the incidence of nonfatal CHD, nor coronary or total mortality, was lowered significantly (16, 17).

Grundy’s way of presenting scientific data is not unique. An analysis of three influential reviews in this field showed that insignificant findings in favor of the diet-heart connection were systematically inflated, and unsupportive studies were either not included or they were quoted as if they were supportive (18).

By Uffe Ravnskov et al., uffe.ravnskov@swipnet.se

References and Notes

1.Expert Panel on Detection,Evaluation,and Treatment of High Blood Cholesterol in Adults,J.Am.Med.Assoc.2 28 85 5,2486 (2001).
2 Report of the Dietary Guidelines Committee on the Dietary Guidelines for Americans,2000(U.S.Department of Agriculture,Agricultural Research Service, Washington,DC,2000).
3.R.B.Shekelle et al.,N.Engl.J.Med.3 30 04 4,65 (1981).
4.D.Kromhout,C.D.L.Coulander,Am.J.Epidemiol. 1 11 19 9,733 (1984).
5.D.L.McGee et al.,Am.J.Epidemiol.1 11 19 9,667 (1984).
6.L.H.Kushi et al.,N.Engl.J.Med.3 31 12 2,811 (1985).
7.P.Pietinen et al.,Circulation9 94 4,2720 (1996).
8.K.L.Esrey,L.Joseph,S.A.Grover,J.Clin.Epidemiol. 4 49 9,211 (1996).
9.F.B.Hu et al.,N.Engl.J.Med.3 33 37 7,1491 (1997).
10.R.B.Singh et al.,J.Am.Coll.Nutr..1 17 7,342 (1998).
11.A.W.Caggiula,V.A.Mustad,Am.J.Clin.Nutr.6 65 5 (suppl.),1597S (1997).
12.A.Kagan et al.,J.Chronic Dis.2 27 7,345 (1974).
13.M.G.Marmot et al.,Am.J.Epidemiol.1 10 02 2,514 (1975).
14.R.M.Worth et al.,Am.J.Epidemiol.1 10 02 2,481 (1975).
15.M.R.Law et al.,Br.Med.J.3 30 08 8,363 (1994).
16.U.Ravnskov,J.Clin.Epidemiol.5 51 1,443 (1998).
17.L.Hooper et al.,Br.Med.J.3 32 22 2,757 (2001).
18.U.Ravnskov,J.Clin.Epidemiol.4 48 8,713 (1995)

 |  |  |  |  |