In the epidemiologic study of chronic diseases, especially that of cancer and heart disease, it is often necessary to resort not only to the work of epidemiologists but also to that of mathematicians and statisticians. The present study is an application of two main methods of multivariate analysis -- factorial analysis and canonical correlation analysis -- to the determination of the evolution of the risk to develop coronary heart disease in an urban population.
Transversal epidemiologic investigations carried out in different populations from several regions of Romania : Gurghiu Valley (lumberjacks from a mountain region), Danube Delta and Razelm lagoon complex (fishermen), and Bucharest have shown that, in spite of the high caloric value of food and even of a high intake of saturated fats, mean serum cholesterol is lower in the rural areas than in Bucharest, probably owing to the strenuous physical work. However, except myocardial infarction, more frequent in the urban than in the rural regions, the other forms of coronary heart disease have a relatively higher frequency in villages, particularly atrial fibrillation and ECG signs of ischemia. These findings might be explained by a greater prevalence of hypertension in these populations. It is concluded that the risk factors, which act synergically, depend on the complex structure of the "ecologic niche".
By means of the factor analysis and the discriminant analysis we studied the structure of the interrelations between the risk factor variables in cardiovascular diseases. The reduction of the dimensionality of data by extracting a small number of common factors has allowed us the identification of the three main factors whose structural change reflects the successive passage through the states of health, risk and illness. To estimate the effectiveness of the prevention action we calculated the D2 distance between the groups of subjects at the two extreme moments. In the last part of the study we determined the predictive value of the risk factor variables considering in turn the main risk factors as criterion variables in terms of the other risk factors taken as predictors.
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)