Health and Development in the 20th Century HEALTH AND DEVELOPMENT IN THE ZOTH CENTURY istory. The magnitude of this transformation can be illustrated by looking at the example of Chile. By the mid-1990s, Chile had a per capita income of about US$ 4000 (adjusted for purchasing power of currency), i.e. it had a high average standard of living with an income level sufficient to provide its people with more than adequate food, shelter and sanitation. Yet Chilean women today have a life expectancy of 79 years-perhaps 25 years more than women in a country with a similar income level in 1900 (and 46 years more than Chilean women had in the early 1900s). This chapter briefly describes this 20th cen tury revolution in human health, then examines both its profound consequences for hu man demography and its contribution to the worldwide diffusion of rapid economic growth. THE ZOTH CENTURY REVOLUTION IN HUMAN HEALTH The steady improvement in life expectancy that began in Europe in the late 1900s con tinued virtually without interruption throughout the 20th century. In England and wales for example, life expectancy was around 40 years in the late 19th century; but by early in the 20th century it had risen to almost 50 years. Other countries experienced similar take-off periods. In Europe these mostly occurred in the late 19th or early 20th century Economic historians and demographers debate the genesis of these increases in life expectancy, but the increases appear to be at least partially linked to the economic changes resulting from the agricultural and industrial revolutions. One aspect of economic change urbanization - actually affected health adversely by exposing an increasing proportion of the population to crowded conditions, thereby facilitating the spread of infection. Some what more than counterbalancing this effect, though, were increases in nutrient intake and provements in sanitation and water supply resulting from higher income levels (1). Bet ter health and nutritional status were both a result and a cause of income growth. Although northen Europeans had began immunizing against smallpox by early in the 19th century, this was exceptional, and other specific knowledge and tools forimproving health probably played only a limited role in the minor health improvements of the 19th century(2). In contrast, the 20th century health revolution appears to have resulted far more substantially from the generation and application of new knowledge
Health and Development in the 20th Century 1 1 HEALTH AND DEVELOPMENT IN THE 20TH CENTURY The 20th century has seen a global transformation in human health unmatched in history. The magnitude of this transformation can be illustrated by looking at the example of Chile. By the mid-1990s, Chile had a per capita income of about US$ 4000 (adjusted for purchasing power of currency), i.e. it had a high average standard of living, with an income level sufficient to provide its people with more than adequate food, shelter and sanitation. Yet Chilean women today have a life expectancy of 79 years – perhaps 25 years more than women in a country with a similar income level in 1900 (and 46 years more than Chilean women had in the early 1900s). This chapter briefly describes this 20th century revolution in human health, then examines both its profound consequences for human demography and its contribution to the worldwide diffusion of rapid economic growth. THE 20TH CENTURY REVOLUTION IN HUMAN HEALTH The steady improvement in life expectancy that began in Europe in the late 1900s continued virtually without interruption throughout the 20th century. In England and Wales for example, life expectancy was around 40 years in the late 19th century; but by early in the 20th century it had risen to almost 50 years. Other countries experienced similar take-off periods. In Europe these mostly occurred in the late 19th or early 20th century. Economic historians and demographers debate the genesis of these increases in life expectancy, but the increases appear to be at least partially linked to the economic changes resulting from the agricultural and industrial revolutions. One aspect of economic change – urbanization – actually affected health adversely by exposing an increasing proportion of the population to crowded conditions, thereby facilitating the spread of infection. Somewhat more than counterbalancing this effect, though, were increases in nutrient intake and improvements in sanitation and water supply resulting from higher income levels (1). Better health and nutritional status were both a result and a cause of income growth. Although northern Europeans had began immunizing against smallpox by early in the 19th century, this was exceptional, and other specific knowledge and tools for improving health probably played only a limited role in the minor health improvements of the 19th century (2). In contrast, the 20th century health revolution appears to have resulted far more substantially from the generation and application of new knowledge
The World Health Report 1999 mortality rates in European countries continued their decline in the 20th century, and by the second half of the century this mortality revolution had spread to the rest of the world. The 20th century global revolution in health transformed-and is transforming -not only the quality of individual lives, but also the demography of populations. These changed health and demographic circumstances have themselves contributed to wide diffusion of economic growth. This chapter overviews the health revolution and its demographic and economic consequences, as well as looking at why it occurred. Chapter 2 then turms to the epidemiological consequences of the health revolution that result from ageing populations, and describes how the incompleteness of the health revolution has left perhaps a billion people behind. Addressing this"double burden"is perhaps the central issue for health policy Table 1.1 Life expectancy at birth, selected countries, around for the 21st century. 1910 and in1998 THE PRECIPITOUS DECLINE IN MORTALITY . ountry und 1910 Whereas life expectancy in England and Wales varied Australia 75 81 round an average of 40 years during the two centuries prior to 1870, in the subsequent 125 years it almost doubled. Other England and Wales 75 countries shared this pattern of improvement in the 20th 43 4 83 Chile, to continue with the example, provides an inter- ng where the take-off in life Nor 59 75 81 expectancy occurred well within this century. The life ex pectancy at birth for a Chilean female in 1910 was 33 years United Statesb Today her life expectancy would be 78 years, an increase of a remarkable 45 years. How has that made a difference in Registration states only, includes District of Columbia. the lives of Chilean women? Figure 1.1 quantifies one ob- Sources: 1910 data. Preston SH, Keyfitz N, Schoen R. Causes of death: Life tables i vious dimension of change: the probability that a Chilean dations. New Yorkand London, Seminar Press, 1972. For Australia: Cumpston M.) Heath and disease in Australia: A history. Department of Community female would die before her fifth birthday has declined from Health, Canberra, AGPS. 1989. 36% to 1.9%. Less obvious, perhaps, is that throughout 1998 data: United Nations Population Division. Word population prospects The 1998 revi. middle life death rates are far lower: she is now much less on New York United Nations, 1998. likely to die as a young adult from tuberculosis or in child- Figure 1.1 Age distribution of deaths in Chile, females, 1909 and 1999 cohorts Age groups 25-29 Percentage of all deaths in the cohort Percentage of all deaths in the cohort 口 Infectious diseases口 Noncommunicable diseases口 Injuries
2 The World Health Report 1999 Mortality rates in European countries continued their decline in the 20th century, and by the second half of the century this mortality revolution had spread to the rest of the world. The 20th century global revolution in health transformed – and is transforming – not only the quality of individual lives, but also the demography of populations. These changed health and demographic circumstances have themselves contributed to wide diffusion of economic growth. This chapter overviews the health revolution and its demographic and economic consequences, as well as looking at why it occurred. Chapter 2 then turns to the epidemiological consequences of the health revolution that result from ageing populations, and describes how the incompleteness of the health revolution has left perhaps a billion people behind. Addressing this “double burden” is perhaps the central issue for health policy for the 21st century. THE PRECIPITOUS DECLINE IN MORTALITY Whereas life expectancy in England and Wales varied around an average of 40 years during the two centuries prior to 1870, in the subsequent 125 years it almost doubled. Other countries shared this pattern of improvement in the 20th century, as Table 1.1 shows. Chile, to continue with the example, provides an interesting and well-documented case where the take-off in life expectancy occurred well within this century. The life expectancy at birth for a Chilean female in 1910 was 33 years. Today her life expectancy would be 78 years, an increase of a remarkable 45 years. How has that made a difference in the lives of Chilean women? Figure 1.1 quantifies one obvious dimension of change: the probability that a Chilean female would die before her fifth birthday has declined from 36% to 1.9%. Less obvious, perhaps, is that throughout middle life death rates are far lower; she is now much less likely to die as a young adult from tuberculosis or in childTable 1.1 Life expectancy at birth, selected countries, around 1910 and in 1998 Country Around 1910 1998 Males Females Males Females Australia 56 60 75 81 Chile 29 33 72 78 England and Wales 49 53 75 80 Italy 46 47 75 81 Japan 43 43 77 83 New Zealanda 60 63 74 80 Norway 56 59 75 81 Sweden 57 59 76 81 United Statesb 49 53 73 80 a Excluding Maoris. b Registration states only; includes District of Columbia. Sources: 1910 data: Preston SH, Keyfitz N, Schoen R. Causes of death: Life tables for national populations. New York. and London, Seminar Press, 1972. For Australia: Cumpston JHL (Lewis MJ ed.) Health and disease in Australia: A history. Department of Community Services and Health, Canberra, AGPS, 1989. 1998 data: United Nations Population Division. World population prospects: The 1998 revision. New York, United Nations, 1998. Figure 1.1 Age distribution of deaths in Chile, females, 1909 and 1999 cohorts 0–4 5–9 10–14 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 80–84 85+ WHO99087 Infectious diseases Noncommunicable diseases Injuries Percentage of all deaths in the cohort 1909 50 40 30 20 10 0 1999 Percentage of all deaths in the cohort 0 10 20 30 40 50 Age groups
Health and Development in the 20th Century birth, or in middle age from cancer. Mirroring this mortality reduction- but less easily quantified -are marked changes in her health-related quality of life. She will spend less time in pregnancy and child-rearing. From an average of about 5.3 children at mid-centur Chilean womens fertility has dropped to its current level of 2.3-barely above replacement level. She will have fewerinfections, less anaemia, greater strength and stature, and a quicker mind. Her life is not only much longer, it is much healthier as well. While Chile s progress in this century has been exceptional, most low and middle in come countries have undergone(or are undergoing) a similar transformation of health and mortality levels. Recent exceptions to these favourable trends occur in AIDS-ravaged parts of Africa and, for a variety of reasons, among adult males in central and eastern Europe DEMOGRAPHIC TRANSITION lthough mortality declines have typically led to increases in population growth rates, these increases prove temporary. Fertility decline accompanies or soon follows mortality decline, bringing growth rates back to low levels. Figure 1. 2 shows a half century of decline, for each WHO Region, in total fertility rate (TFr)-the expected number of children a woman would bear at the prevailing age-specific fertility rates. A TFR of a little over 2 rep resents a replacement level of fertility, ie a level that if maintained in the long run would result neither in population growth nor decline. At mid-century, fertility rates were extremely high in most countries of the world(with the exception of the high income countries). TFRs of 6 were not uncommon. Figure 1.2 shows that every region except Africa has experienced sharp declines in fertility And evi dence is mounting that the decline in Africa has now commenced Declining birth rates lead to stabilization in the size of the youngest age cohorts. Over me these youths become middle-aged, while the younger cohorts remain about the same ze Figure 1.3 provides a further example of change, with data illustrating the population age distributions(pyramids) in WHOs South-East Asia Region for 1950, 2000 and 2050 After the rapid decline in fertility, age distributions change, but only slowly. If the South East Asia Region s TFR remains at 2.9, its average population age will continue increasing for decades to come The world today is perhaps somewhat past the halfway point of a two-century period during which the demographic characteristics of the human population will have been totally transformed. This trans- Figure 1. 2 Declines in fertility by WHO Region, 1950 and 1998 formation (or demographic transition)entails a move from very high birth and death rates to low ones; a move from Africa initially low population growth rates through a period of The americas high rates and a vast increase in total population then back to low or zero growth rates; and a move from an age distri- bution with numerous young and few elderly to one with nearly equal numbers in most age groups. Enormous social, Europe economic and epidemiological changes follow the demo- South-East asia graphic transition, which is itself a consequence of the still ongoing revolution in mortality. Chapter 2 will point to the Western Pacific tion and the concluding section of this chapter will outline possible economic consequences. This report simply notes 19501998 the great importance of these changes, rather than discuss Source: United Nations Population Division World population prospects: The 1998 revision. New ng them in any detail
Health and Development in the 20th Century 3 birth, or in middle age from cancer. Mirroring this mortality reduction – but less easily quantified – are marked changes in her health-related quality of life. She will spend less time in pregnancy and child-rearing. From an average of about 5.3 children at mid-century, Chilean women’s fertility has dropped to its current level of 2.3 – barely above replacement level. She will have fewer infections, less anaemia, greater strength and stature, and a quicker mind. Her life is not only much longer, it is much healthier as well. While Chile’s progress in this century has been exceptional, most low and middle income countries have undergone (or are undergoing) a similar transformation of health and mortality levels. Recent exceptions to these favourable trends occur in AIDS-ravaged parts of Africa and, for a variety of reasons, among adult males in central and eastern Europe. DEMOGRAPHIC TRANSITION Although mortality declines have typically led to increases in population growth rates, these increases prove temporary. Fertility decline accompanies or soon follows mortality decline, bringing growth rates back to low levels. Figure 1.2 shows a half century of decline, for each WHO Region, in total fertility rate (TFR) – the expected number of children a woman would bear at the prevailing age-specific fertility rates. A TFR of a little over 2 represents a replacement level of fertility, i.e. a level that if maintained in the long run would result neither in population growth nor decline. At mid-century, fertility rates were extremely high in most countries of the world (with the exception of the high income countries). TFRs of 6 were not uncommon. Figure 1.2 shows that every region except Africa has experienced sharp declines in fertility. And evidence is mounting that the decline in Africa has now commenced. Declining birth rates lead to stabilization in the size of the youngest age cohorts. Over time these youths become middle-aged, while the younger cohorts remain about the same size. Figure 1.3 provides a further example of change, with data illustrating the population age distributions (pyramids) in WHO’s South-East Asia Region for 1950, 2000 and 2050. After the rapid decline in fertility, age distributions change, but only slowly. If the SouthEast Asia Region’s TFR remains at 2.9, its average population age will continue increasing for decades to come. The world today is perhaps somewhat past the halfway point of a two-century period during which the demographic characteristics of the human population will have been totally transformed. This transformation (or demographic transition) entails a move from very high birth and death rates to low ones; a move from initially low population growth rates through a period of high rates and a vast increase in total population then back to low or zero growth rates; and a move from an age distribution with numerous young and few elderly to one with nearly equal numbers in most age groups. Enormous social, economic and epidemiological changes follow the demographic transition, which is itself a consequence of the stillongoing revolution in mortality. Chapter 2 will point to the epidemiological consequences of the demographic transition and the concluding section of this chapter will outline possible economic consequences. This report simply notes the great importance of these changes, rather than discussing them in any detail. Figure 1.2 Declines in fertility by WHO Region, 1950 and 1998 Western Pacific South-East Asia Europe Eastern Mediterranean The Americas Africa 0123 4 567 1950 1998 WHO99091 Source: United Nations Population Division. World population prospects: The 1998 revision. New York, United Nations, 1998. Total fertility rate
The World Health Report 1999 Figure 1.3 Distribution of the population of the South-East Asia Region, by age and sex, 1950, 2000 and 2050 Male Age groups Female 75-79 65-69 50-54 45-49 40-44 302010 50-54 45-49 35-39 30-34 25-29 20-24 15-19 45-49 40-44 25-29 9494 2050 Source: United Nations Population Division. Ward population prospects: The 1998 revision. New York, United Nations, 1998
4 The World Health Report 1999 Figure 1.3 Distribution of the population of the South-East Asia Region, by age and sex, 1950, 2000 and 2050 Males Age groups Females WHO99092 Percentage of total population (both sexes) Source: United Nations Population Division. World population prospects: The 1998 revision. New York, United Nations, 1998. 012345678 0–4 5–9 10–14 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 80+ 876543210 1950 0–4 5–9 10–14 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 80+ 876543210 012345678 2000 0–4 5–9 10–14 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 80+ 876543210 012345678 2050
Health and Development in the 20th Century SOURCES OF MORTALITY DECLINE ncome improvements can lead to mortality reductions, and numerous studies have attempted to quantify this effect. Analyses undertaken as background to this report, for example, assessed for all countries the effects of national income on health outcomes dur- I1952-1992. Figure 1.4 shows results from this analysis in curves relating the infant mortality rate(IMR) to gross domestic product(GDP) per capita(adjusted for pur- chasing power). Income increases do indeed correlate with mortality declines and there are ood reasons to believe that the relation is causal in both directions How much of the remarkable decline in infant mortality rates has resulted from income growth during that period? The upper curve shows the income-mortality relation in 1952 and the lower one, for 1992, shows how much lower mortality rates had become by then for any given level of income. Figure 1. 4 suggests that however important income growth may be, the changing relation beteen mortality and other factors (e.g. access to health technol- ogy) is likely to be more important. Between 1952 and 1992, for example, per capita income increased by about two thirds, on average, across the coun- tries included in the analysis-from about $1530 to $2560 Table 1. 2 Sources of mortality reduction, 1960-1990 The upper curve in Figure 1.4 shows that had the income- Reduction Income Educational level Generation and mortality relation remained as it was in 1952, the IMRwould of adult females have declined from 144 to 116. In fact it declined to only 55 knowledge Table 1.2 reports the results of an attempt to quantify the Under-5mortality rate relative importance of key determinants of mortality reduc- Female adult mortal tion. It draws on a statistical assessment of how the relation Male adult mortality as changed over time between various health indicators Female life expectancy at birth 19 and both income levels and average educational levels(of Male life expectancy at birth 30 adult females ) The table reflects a decomposition of the Total fertility rate improvement in health into three components: Note: The results are based on analysis of data from 115 low and middle increases in average income levels; improvements in aver- Source: Wang Jet al. Measuring countryperformanceon heath: Selected indicators for 115coun. age educational levels; and a favourable shift in the under- tries Washington DC, The World Bank, 1999 (Human Develo etwork, Health, Nutrition and populated lying curve. This favourable shift is ascribed to the generation Figure 1.4 The role of improvements in income in reducing infant mortality rates 1952 (36 countries) mean income in 1952=$1,534: Predicted IMR= 144 Predicted IMR (at 1992 mean income)=116 1992 (101 countries Predicted IMR (at 1952 mean income)=62 50 Mean income in 1992=52 561: Predicted IMR=55 52000 54000 GDP per capita, adjusted for purchasing power (in 1985 international dollar rithm of income, the square of the natural logarithm of income and indicator variables for time. Data sources are the same as for Annex Tab Note: Results al ar intervals, the natural logarithm of IM
Health and Development in the 20th Century 5 SOURCES OF MORTALITY DECLINE Income improvements can lead to mortality reductions, and numerous studies have attempted to quantify this effect. Analyses undertaken as background to this report, for example, assessed for all countries the effects of national income on health outcomes during the period 1952–1992. Figure 1.4 shows results from this analysis in curves relating the infant mortality rate (IMR) to gross domestic product (GDP) per capita (adjusted for purchasing power). Income increases do indeed correlate with mortality declines and there are good reasons to believe that the relation is causal in both directions. How much of the remarkable decline in infant mortality rates has resulted from income growth during that period? The upper curve shows the income–mortality relation in 1952 and the lower one, for 1992, shows how much lower mortality rates had become by then for any given level of income. Figure 1.4 suggests that however important income growth may be, the changing relation between mortality and other factors (e.g. access to health technology) is likely to be more important. Between 1952 and 1992, for example, per capita income increased by about two thirds, on average, across the countries included in the analysis – from about $1530 to $2560. The upper curve in Figure 1.4 shows that had the income– mortality relation remained as it was in 1952, the IMR would have declined from 144 to 116. In fact it declined to only 55. Table 1.2 reports the results of an attempt to quantify the relative importance of key determinants of mortality reduction. It draws on a statistical assessment of how the relation has changed over time between various health indicators and both income levels and average educational levels (of adult females). The table reflects a decomposition of the causes of improvement in health into three components: increases in average income levels; improvements in average educational levels; and a favourable shift in the underlying curve. This favourable shift is ascribed to the generation Table 1.2 Sources of mortality reduction, 1960–1990 Reduction Percentage contribution of gains in Income Educational level Generation and of adult females utilization of new knowledge Under-5 mortality rate 17 38 45 Female adult mortality rate 20 41 39 Male adult mortality rate 25 27 49 Female life expectancy at birth 19 32 49 Male life expectancy at birth 20 30 50 Total fertility rate 12 58 29 Note: The results are based on analysis of data from 115 low and middle income countries. Source: Wang J et al. Measuring country performance on health: Selected indicators for 115 countries. Washington DC, The World Bank, 1999 (Human Development Network, Health, Nutrition and Population Series). Note: Results are based on a cross-sectional time-series regression that relates, at 5-year intervals, the natural logarithm of IMR to the natural logarithm of income, the square of the natural logarithm of income and indicator variables for time. Data sources are the same as for Annex Table 6. Predicted IMR (at 1992 mean income) = 116 Figure 1.4 The role of improvements in income in reducing infant mortality rates WHO 9983 Infant mortality rate (IMR) (per thousand) GDP per capita, adjusted for purchasing power (in 1985 international dollars) $0 $1 000 $3 000 $4 000 $2 000 mean income in 1952 = $1,534; Predicted IMR = 144 Predicted IMR (at 1952 mean income) = 62 Mean income in 1992 = $2,561; Predicted IMR = 55 1952 (36 countries) 1992 (101 countries) 150 200 100 50 0