FIGURE 7: Annual Age-Adjusted Death Rates from Malignant Neoplasms per 100,000 Population at Ages 15-74 Years, by Sex, in Each City, 19621964 17

FIGURE 8: Annual Age-Adjusted Death Rates from Four Groups of Cardiovascular Diseases per 100,000 Population at Ages 15-74 Years, by Sex, in Each City, 1962-1964 18

Comparing these data on death rates, mortality differs with respect to disease and to sex.

For males, neoplasms at ages 15-74 years cause death at a higher rate than for females in seven cities: La Plata, Bristol Ribeirão Prêto, Caracas, San Francisco, Santiago and São Paulo (in highest to lowest death-rate rank order). Only in Lima, Bogota, Cali, Guatemala City and Mexico City (in highest to lowest death-rate rank order) did female deaths due to malignant neoplasms exceed male deaths in each city. There is no common numerical difference between male and female deaths when all the cities are compared: each city has a specific difference in death rates between its male and its female residents.

In ten cities males have higher death rates than females when the cardiovascular disease data are compared: San Francisco, Bristol, São Paulo, La Plata, Ribeirão Prêto, Caracas, Santiago, Lima, Cali and Mexico City. Only in Bogota do females die from cardiovascular dis

disease death rate for men and women is the same. Again, each city has a specific difference in the death rates between its male and female residents.

When the death rates are compared between diseases in all cities, cardiovascular disease causes death among males more often than does malignant neoplasms. Only in Lima and Guatemala City is this trend reversed for female residents: in Lima and Guatemala City, death rates for malignant neoplasms are higher than the death rates from cardiovascular disease for female residents.

The most interesting observation which emerges from study of this data is that the comparisons hold true across the geographic zones. In no case was female or male deaths resulting from malignant neoplasms or cardiovascular diseases confined to a certain area: every zone contained cities with death rates numerically comparable to the death rates of cities in other zones.

Marked contrasts would be expected where city populations are subjected to widely differing biological and environmental conditions. and indeed such differences in the overall circumstances of life are the purpose of international epidemiological studies. There were such contrasts in specific types of cardiovascular diseases and malignant neoplasms. Indeed, the differences in mortality between some cities were often so large that neither variations due to chance, errors in the material, knowledge of the diseases, nor problems of classification could reasonably account for them. In various forms and sites of cancer, environmental differences between certain cities could not account for them either. Nevertheless, the search for environmental determi nants is incomplete and it remains the most fruitful approach. Only a continuing observation of variation in death rates and the prolonged exposure of populations to given urban environments is likely to yield clues to the real culprit.

Again, a specific type of cardiovascular disease, namely, arteriosclerotic heart disease, appears to indicate that urbanization (modernization and industrialization) plays a profound overall environmental role in increasing the risk for the disease. Arteriosclerotic heart disease is clearly a more frequent threat to males in San Francisco and Bristol than in Latin American cities (See left segment of bar graphs in Figure 8). It is not so simple an explanation as that of males in these English-speaking cities living longer than males in Latin Amer ican cities so that they reach the so-called heart disease age; the death rates were much higher in San Francisco, even for young adult males 25-44 years of age, than in the other cities.

It was stated earlier that the widely scattered cities of the world represented different degrees of industrialization, and that the death rates from cardiovascular diseases and malignant neoplasms were un expectedly high in most cities or urban areas. It was also indicated that in these two disease groups there is more often than not a numerically comparable death rate between geographic and climatic zones. In a review of the report from which the material in this section is obtained (Patterns of Urban Mortality, PAHO, 1967) ten of the twelve cities studied were in Central and South America. A table in that report on "23 causes of death in each of the twelve cities" shows that cardiovascular diseases in 1962-1964 held positions of prominence as causes of death in all of the cities, including, of course, San Fran

cisco and Bristol. It seems clear therefore, that the statistical data of the past tended to understate the true extent of cancers and diseases of the heart and blood vessels in Latin America, partly because of the pressing health needs associated with malnutrition and communicable diseases and partly because the populations had a relatively young age

The structure.

The urban populations in all parts of the Western Hemisphere are growing. This is accompanied by the increasing importance of cancers and diseases affecting the cardiovascular systems of urban populations in all countries, and especially those in Latin America. The situation appears to call for a strong, permanent, continuing program with all major nations cooperating through their city health departments. The Office for Research Development and Coordination in the Pan American Health Organization, established in 1960 under a grant from the United States Public Health Service, might serve as a base of support and operations for a more intensified effort along these lines. An alternative would be a semi-autonomous center for urban health research, affiliated with PAHO and sponsored by especially interested countries or by the Organization of American States.

VII. REGIONAL PATTERNS OF MORTALITY IN THE

AMERICAS

In a study of health conditions in the Americas,19 the five principal causes of death as a percentage of total deaths in the Northern, Middle, and South American regions leave little doubt that Northern America is a region dominated mostly by diseases of the heart, stroke, and cancer; whereas Middle and South America are regions where deaths from infectious and parasitic disease and diseases of early infancy combine to equal or exceed deaths from heart diseases or malignant neoplasms (cancers). Nevertheless, cancer is the fourth cause of death in Middle America, while in South America diseases of the heart rank No. 1 and cancer ranks No. 3.* The figures below show the pattern of age-adjusted death rates by sex from diseases of the heart and from malignant neoplasins in various countries of the Western Hemisphere. Each of the two figures has a different scale, so examining the death rates rather than the length of the bars reveals that heart disease, except for females, presents a higher death rate per 100,000 population for most countries than does cancer:

Health Conditions in the Americas, 1965-1968, Pan American Health Organization, September 1970. Wash., D.C., pages 14-24.

Ibid., page 18 (Figures 19 and 20). The apparent inconsistency of these statements with common knowledge is the ill-defined nature of the infection-malnutrition complex (see Introduction).

(81)