Search

Manpower has often been said to be the key problem in the expansion of the health services, despite the fact that health manpower resources have expanded extremely rapidly (about four per cent per year in the 1950's, and 2.5 per cent per year so far in the 1960's). Yet, as in all services, manpower remains the crucial resource in health services. How rapidly and how well manpower is developed essentially determines how rapidly health services rise. This paper provides an appraisal of the recent research into manpower for the health services, both to indicate its accomplishments and to suggest the main lines which should be pursued in the future. The key question has always been: How can research on manpower contribute to the expansion and improvement of health services to the nation? Since the number of those who are engaged in health manpower research is quite limited, this paper has been written particularly with the potential, but inexperienced, researcher in this field in mind. The intent has been to make the key issues explicit, to indicate what has been emphasized to date and to point out open research areas where specific efforts are likely to yield significant contributions. At this stage, more rapid progress will be gained from solidly constructed, finely focused efforts rather than from grand designs.

D IABETES mellitus as a public health problem has become more important because of the aging of the population. Its importance in public health is clearly illustrated by a survey conducted in Oxford, Massachusetts, where 71 per cent of the total population, 3,516 persons, had urine tests and blood sugar examinations. This study was initiated with the aid of the State and district medical societies, the State and local boards of health, and the local physicians. The study was conducted by personnel assigned by the United States Public Health Service (1). A total of seventy cases of diabetes was discovered in this study. Forty were previously known. The fact that thirty cases were newly discovered indicates the need for public health endeavor in this field. This particular paper presents a description of diabetic persons reported in a sample population observed from two to five years in the Eastern District of Baltimore. These persons are described with respect to position in the household, age and year of diagnosis, and first symptoms or complaints of illness. Severity of illness, medical care, and amount of disability during observation are also described.

As the health services in developed countries continue to expand, their demands increase on skilled and expensively trained manpower. Already they employ approximately six per cent of the total manpower in some countries, ranking as the second or third largest employers. As a matter of fact, at the present rate of growth the health industry could be the biggest employer of labor well before the end of the century. On the one hand, the consumer wishes to have the latest and the best, in the way of treatment and specialties, of the technologic revolution in medicine. On the other, there has been little measurable payoff during the past three decades in terms of reduced death rates in any age group and little evidence of benefit in reduced disability; indeed, often the very advances in biomedical science simply mean that more people survive into old age, frequently on maintenance therapy and at further risk of an even more critical episode. Thus, most developed countries now proclaim a shortage of skilled manpower, some admit to having to ration resources, a few are considering a ceiling on training specialist physicians, and at least the United Kingdom is reducing the supply of acute hospital beds by one-third. In the Liverpool area, the present and related studies have contributed to such a radical redeployment of resources. This paper presents a preliminary budget of investment in the twelve study areas of the World Health Organization/International Collaborative Study of Medical Care Utilization. It gives

Knowledge of the causes of death in a given population depends on the ways in which the death certificates are filled out, collected, and processed. The first two steps are generally responsible for the errors which afflict mortality statistics in Latin America. Factors involved in such errors are probably due more to questions of quantity than of quality; this should be remembered in order to interpret properly the vital statistics of the region. Therefore, death rates in general, and those for certain diseases in particular, tend to be below rather than above the actual figures in Latin America. This, however, does not generally affect the relative importance of each disease in the same country, nor does it make it difficult to judge the weight that each one should have in a public health program. By comparing the available figures of one country with those of another and with those of the more developed countries 30 or 50 years ago, it will be possible to determine the leading diseases that affect health conditions in a given area and the stage of their evolution. To carry out a study of the type presented in this paper, it is necessary to have the figures easily available. Unfortunately, only 10 Latin American countries have tried to supply the data regularly for the yearbook on the subject published by the World Health Organization.1 Chile, Colombia, Costa Rica, El Salvador, Mexico, Guatemala, Nicaragua, Panama, Puerto Rico, and Venezuela publish data by sex and age groups for the different causes of death. Data for other countries have been taken from publications of the Pan American Sanitary Bu

During recent years the high mortality rates once existing in many Latin American countries have undergone extremely rapid declines. At the same time, moderate increases in fertility have occurred in most of these countries.l The question therefore arises as to whether the drastic fall in mortality, quite apart from the associated improvement in health, is tending to increase the fertility of Latin American populations.2 This paper will attempt to answer the question affirmatively by demonstrating that a drop in mortality automatically causes the gross reproduction rate3 to rise. By definition, however, changes in mortality cannot directly affect the gross reproduction rate. The latter is based on the total female population in the reproductive ages and on the number of births within this population. It assumes that a woman lives through the reproductive period. Therefore, when mortality changes, but the age-specific fertility rates remain constant, the gross reproduction rate cannot vary. Yet this theory does not consider the indirect effect of mortality decline on the mating system. If only those women living in unions4 are having children,5 the gross reproduction rate will vary with mortality even though the age-specific fertility within unions remains the same. A decline in mortality has the effect of expanding the lifespan of unions and thus of increasing the proportion of women living in unions. The longevity of the union itself increases and

In two of his more recent papers, Spiegelman stressed that it was necessary to examine the experience of generations rather than the cross section of period experience to obtain a better understanding of the underlying changes in mortality with advance in age.l 2 As an example of how current life tables can sometimes present a misleading picture, Spiegelman cited the case of tuberculosis: current life tables pointed to a rise in death rates with advancing age whereas generation experience indicates a decline in tuberculosis mortality at the older ages. The age pattern of generation mortality from specific causes of death can differ significantly from that deduced on the basis of current life tables, depending on the trends in the age-specific death rates. In a broader context Spiegelman observed that the mortality experience of white males in the United States has

The effects of changes in mortality rates on a population's growth and age composition depend on a number of factors: the initial level of mortality, the magnitude and age pattern of the mortality changes and the levels of fertility prior to and subsequent to the changes. The variety of effects associated with variations in these factors has been studied analytically as well as empirically.l5 In this paper the empirical approach will be stressed, the focus being on the role of mortality in the development of the United States population in the twentieth century.6 Three key generalizations have emerged from viewing the problem analytically. These will prove useful in studying the United States data and the data in turn will serve to illustrate and amplify these points.

IN THE United States, growing interest in disability and medical care insurance, as well as group plans for provision of medical service presents an increasing need for statistics on the incidence of illness among persons of opposite sex and of different ages. Since 1920 data of this type have been made available for certain population groups by special morbidity studies of families. One of the studies gives illness rates by sex and age for the population of sample areas in the United States: the Committee on the Costs of Medical Care study of 19281931 (1). Two other studies provide data for local areas in Maryland: one made in Hagerstown from 1921-1924 (2) and the other conducted in the Eastern Health District of Baltimore from 1938-1943. A number of preliminary reports of the study in the Eastern Health District of Baltimore have been published (3, 4). The purpose of the present paper is to make available data from this study on illness by age and sex in more detail than has been given in the past.

NUMBER of longitudinal studies of families have been made in order to observe the incidence of illness among their numbers. It is recognized by those who have conducted such studies that the interval between visits to the family has a marked effect upon the accuracy of reporting past illnesses, especially those of a minor nature. Thus, the memory of the informant constitutes an important factor which may influence the level of illness obtained in any morbidity study. The purpose of this paper is to consider some of the effects of the frequency of family visiting upon the accuracy of reporting minor illnesses. The data are limited to acute respiratory diseases because most illnesses in this class are of a minor

5T HE morbidity study, conducted from June, 1938 to May, 1943, in the original Eastern Health District of Baltimore, is the first investigation of illness where, due in large part to the method of sampling, families with relatively short periods of observation constitute an important part of the total observed population of families. Preliminary analysis of the morbidity experience of the first year of the study indicated that about one-third of the total families either moved out of or into the study area during that period (1). Movement of families continued throughout the five-year period. This paper presents a study of some of the socio-economic characteristics of the moving and non-moving families-characteristics which form a background for forthcoming analyses of illness. Reed, et al. have presented some general characteristics of the population in the Eastern Health District (Wards 6 and 7) from which the morbidity study population was drawn (2). They found that in 1939, 56 per cent of the white families in the district were home owners. They concluded also that "The population is essentially in the lower middle economic class with a greater proportion of skilled and semi-skilled workers 'relatively' than in the rest of the city."

T HE formal structure and function of Soviet health services have been described abundantly over the last 25 years. As the first fully socialist nation, its approach to problems of health and the organization of medical care has interested hundreds of western observers, and we have available as a result of this interest many excellent and thorough accounts, from the Sigerist volumes of the 1930's to the recent study-tour reports of the World Health Organization and the United States Public Health Service. The excuse for another exposition would not be to add further details to the picture of Soviet medicine. A two-week observation could hardly do that anyway. It is, rather, to attempt to analyze the Soviet health service system in terms of certain key concepts that are of interest in the United States and perhaps in all countries. Under each concept, one may explore the Soviet approach in several of the formal subdivisions of the health services. One need hardly be reminded that the Soviet Union is a huge country-the largest land area of any nation in the world, with a population of 220,000,000. Because it is built from fifteen republics and an enormous diversity of local national and ethnic groups, one should hardly expect a simple, clear picture of uniformity, in the health services or any other aspect of life. This would be true even had the past 40 years since the 1917 revolution been times of quiet social development. But since the overthrow of Czarism, almost half the time has been spent in periods of war and reconstruction. After the First World War the country experienced a period of hostility from the * Based largely on a two-week study-tour made by a party of ten American and British physicians and social scientists in August, 1961. At the time of this study and preparation of this paper, the author was Research Professor of Administrative Medicine at Cornell University. ** Professor of Public Health, University of California, Los Angeles.

LTHOUGH the problem of age misreporting in the census (and other social surveys) has long been recognized,l few techniques have been developed for estimating and correcting the errors involved. Those techniques which are available for demographic analysis are, for the most part, concerned with digital preferences or five year age distributions;2 as such they are inadequate for determining preferences or avoidances for individual years of age. The method employed in this paper is an attempt to establish the size of error for each year of age, relative to its neighbors, for the native white male and native white female populations enumerated in the United States censuses from 1880 to