Causality In Medical Science with Particular Reference to Heart Disease and Atherosclerosis

CAUSALITY IN MEDICAL SCIENCE WITH PARTICULAR REFERENCE TO HEART DISEASE AND ATHEROSCLEROSIS WILLIAM E. STEHBENS* The correlation of fat consumption with coronary heart disease (CHD) mortality in six countries specifically selected by Keys [1] and the introduction of the risk concept in the Framingham Study [2] triggered an upsurge in CHD epidemiology. These studies established the prevailing view that a high dietary intake of saturated fats and cholesterol results in an elevation of blood cholesterol or in low density lipoprotein levels which, aided at times by other risk factors, are primarily responsible for an enhanced tendency to severe atherosclerosis and CHD. Despite enormous expenditure of time, finance, and animals, controversy over the etiology, pathogenesis, and preventive management persists unabated. Serious criticisms of the consensus concept of atherosclerosis [3, 4] and possible deleterious effects associated with dietary changes recommended for whole populations reinforce the need for review of the evidence. At the crux of the issue is the use and meaning of cause in medicine. The aim of epidemiology has been to identify factors that may facilitate prevention and control of disease. Of particular value in infectious and occupational diseases, epidemiology has limited applicability to chronic degenerative diseases, especially when, like atherosclerosis, they are ubiquitous. As a scientific discipline, epidemiology must reflect scrupulous adherence to the principles of logic, clarity, precision of thought, and quality of data. It follows that the meaning of cause must be clear and precise in all scientific communications. The author acknowledges financial support from the National Heart Foundation of New Zealand, the New Zealand Dairy Board, and the New Zealand Meat Producers' Board. Appreciation is expressed to Mrs. T. Todd for her assistance. *Department of Pathology, Wellington School of Medicine, P.O. Box 7343, Wellington South, New Zealand.© 1992 by The University of Chicago. All rights reserved. 0031-5982/93/3601-0803$01.00 Perspectives in Biology and Medicine, 36, 1 ¦ Autumn 1992 | 97 Basic Scientific Precepts Causal knowledge is indispensable to science [5], and in considering disease causation several basic precepts are essential to facilitate logical discussion: 1. Basic to disease prevention and management is the concept ofspecific disease entities. The word disease must be restricted in usage to indicate a specific malady [6] and not used carelessly or synonymously with (1) symptoms, signs, or laboratory findings, e.g., headache, hypertension, pyrexia, hypercholesterolemia; (2) nonspecific complications, e.g., embolism , hemorrhage, ischemia, necrosis; and (3) a group or class of pathological states, e.g., stroke, subarachnoid hemorrhage, myocardial ischemia, CHD. Each is a manifestation of several diseases and not a final diagnosis in itself, even though often regarded as such clinically. The diseases responsible for each condition must be sought for diagnostic accuracy and successful clinical management, a precept mandatory for epidemiology and clinical science. Moreover it does not follow that because symptoms, signs, laboratory data, and nonspecific complications and pathological states have multiple causes that all specific diseases have multiple causes. CHD is not a specific disease, though it has been erroneously defined as such [7] and is frequently treated as such in epidemiology. It is in reality an imprecise clinical diagnosis comprising several clinical syndromes of myocardial ischemia attributable to insufficient coronary blood flow associated with occlusion or severe stenosis of coronary arteries complicating many diseases [8]. It is synonymous with myocardial ischemia but neither synonymous with nor pathognomonic of coronary atherosclerosis. Although CHD was incorrectly defined by a panel of experts [9] of the National Institutes of Health (NIH) as coronary atherosclerosis (usually implying considerable severity), it does not indicate any specific degree of severity of atherosclerosis [8]. In epidemiology, CHD and atherosclerosis are used synonymously with invalid extrapolations commonly made from CHD risk factors to the etiology of atherosclerosis , even though the latter commences in infancy if not in utero, whilst CHD occurs primarily in the sixth decade and beyond (Figure 1). It is the severity of atherosclerosis that should be sought [8], not the prevalence of CHD, which is a nonspecific manifestation of end-stage atherosclerosis of variable severity. Dawber and Kannel [10] admitted CHD was not a definable epidemiological disease entity and that the use of CHD vital statistics was unwarranted . Their inaccuracy and unreliability are well recognized [11—13...