6 The World Health Report 1999 and application of new knowledge. Other indicators of social welfare show a similarly modest relation to income growth along with favourable time trends at any given level of income (3). Much research remains to be done, however, in order to achieve a complete under standing of why the income-health relation has improved so much. Typically, half the gains in health between 1952 and 1992 result from access to better technology. The remaining gains result from movement along the curve (income improvements and, more impe tantly, better education). Figure 1. 4(which does not control for education changes) illt trates the magnitude of the effect of moving along the curve relative to shifts in the curve (4, 5). Higher levels of income and education affect health through a variety of mechanisms often involving many sectors of the economy. Box 1.1 outlines the main multisectoral de terminants of health The historical evidence points in the same direction. In some countries, many decades nge in health status(for example, Engla Sweden the take-off in health occurred at about the same time as in britain but economi growth began three-quarters of a century later; and in India the take-off in life expectancy preceded that for economic growth(2) Shifting the curve and moving along the curve are, then, both sources of improvements in health. Some countries lie far above the curve, i.e. their mortality rates are much highe than would be predicted by their income. For these countries, joining the curve may be the quickest way to improve health. Chapter 2 will illustrate the extent to which some of the world s most populous countries could make great health gains by joining the curve What conclusions can be drawn from this analysis? First and foremost, it is clear that Box 1.1 The multisectoral determinants of health Hungry children easily acquiredis- terminants of health are truly ple, poorer societies may forego ex- sense the answer is surely yes: the ses, and easily die from the dis- multisectoral health community should measure eases they do acquire Dwellings An assessment commissioned for luted air or water from factories, and the effects on health of actions out- thoutsanitation provide fertileen- WHOs 1997 Ad Hoc Committee on poorer households lack the resources side the health sector. It should en- ironmentsfortransmission of intes- Health Research estimated the per- to purchase indoor sanitation or piped sure that these findings are inal infections. Air dense with centage of deaths, by region and glo- water. Poorly educated individuals may communicated, and are considered particulates or adids destroys lungs, bally, associated with each of ten risk fail toobserve basic hygiene orneglect in making policy choices. The mag- and lives. Hopeless lifecircumstances factors It concluded that the follow- appropriateweaning practices for their nitude of the demonstrated effecto thrust young girls(and boys)into ing risks contributed to global mortal- children; and they are increasingly the girls'education on health and fertil- prostitution with its attendant risks ity in 1990 population that smokes Theeffects of ity outcomes, for example, provides of violence and sexually transmitted tobacco use-6.0% education and income are indeed real a powerful argument for investing iseases, induding HIV/AIDS. Manu- hypertension-5.8% and quantitatively important, even in extension of educational accessto facturers of tobacco and alcohol.inadequate water and sanitation- though only about half of health im- girls. 53%. provements indeveloping countries in But the health community has nd promotion that spreads addic- .risky sexual activity-2.2%; the period 1960-1990 result from limited capacity directactionout- tion. Rapid growth in vehicular traf- .alcohol use.5% hese factors side the health sector -and limited fic-often with untrained drivers on If an important fraction of ill-health credibility. It will make more of adi unsaferoads-generates a rising toll Underlying most specific risks are results from poverty and low educa- ference if it focuses its energy,exper- of injury. Poorly designed imigation more general deteminants of health tional levels-or from their conse- tise and resources on ensuring that rojects create breeding grounds for -income and education levels. The quences in inadequate food or healthsystems efficiently deliverthe ectors of disease. The list could be effects of income and education op- sanitation or other specificrisks-then powerful interventions provided by much extended, and it could be re- erate for the most part through influ- ought the task of the health profes. modern scence. erased in terms of factors favorable encing risk (and being able to utilize sional lie principally in addressing health, but the point is clear: de- health services effectively). For exam- these underlying problems? In on Source: vesting in health research and development Report of the Ad Hoc Committee an Health Research Relating to Future Interention Options Geneva, World Health Organization, 1996 document WHO/TDR/Gen/96.1)
6 The World Health Report 1999 and application of new knowledge. Other indicators of social welfare show a similarly modest relation to income growth along with favourable time trends at any given level of income (3). Much research remains to be done, however, in order to achieve a complete understanding of why the income–health relation has improved so much. Typically, half the gains in health between 1952 and 1992 result from access to better technology. The remaining gains result from movement along the curve (income improvements and, more importantly, better education). Figure 1.4 (which does not control for education changes) illustrates the magnitude of the effect of moving along the curve relative to shifts in the curve (4,5). Higher levels of income and education affect health through a variety of mechanisms, often involving many sectors of the economy. Box 1.1 outlines the main multisectoral determinants of health. The historical evidence points in the same direction. In some countries, many decades of economic growth saw no change in health status (for example, England and Wales); in Sweden the take-off in health occurred at about the same time as in Britain but economic growth began three-quarters of a century later; and in India the take-off in life expectancy preceded that for economic growth (2). Shifting the curve and moving along the curve are, then, both sources of improvements in health. Some countries lie far above the curve, i.e. their mortality rates are much higher than would be predicted by their income. For these countries, joining the curve may be the quickest way to improve health. Chapter 2 will illustrate the extent to which some of the world’s most populous countries could make great health gains by joining the curve. What conclusions can be drawn from this analysis? First and foremost, it is clear that Box 1.1 The multisectoral determinants of health Hungry children easily acquire diseases, and easily die from the diseases they do acquire. Dwellings without sanitation provide fertile environments for transmission of intestinal infections. Air dense with particulates or acids destroys lungs, and lives. Hopeless life circumstances thrust young girls (and boys) into prostitution with its attendant risks of violence and sexually transmitted diseases, including HIV/AIDS. Manufacturers of tobacco and alcohol profit enormously from advertising and promotion that spreads addiction. Rapid growth in vehicular traffic – often with untrained drivers on unsafe roads – generates a rising toll of injury. Poorly designed irrigation projects create breeding grounds for vectors of disease. The list could be much extended, and it could be rephrased in terms of factors favorable to health, but the point is clear: determinants of health are truly multisectoral. An assessment commissioned for WHO’s 1997 Ad Hoc Committee on Health Research estimated the percentage of deaths, by region and globally, associated with each of ten risk factors. It concluded that the following risks contributed to global mortality in 1990: • tobacco use – 6.0%; • hypertension – 5.8%; • inadequate water and sanitation – 5.3%. • risky sexual activity – 2.2%; • alcohol use – 1.5%; Underlying most specific risks are more general determinants of health – income and education levels. The effects of income and education operate for the most part through influencing risk (and being able to utilize health services effectively). For example, poorer societies may forego expensive mechanisms for cleaning polluted air or water from factories, and poorer households lack the resources to purchase indoor sanitation or piped water. Poorly educated individuals may fail to observe basic hygiene or neglect appropriate weaning practices for their children; and they are increasingly the population that smokes. The effects of education and income are indeed real and quantitatively important, even though only about half of health improvements in developing countries in the period 1960–1990 result from these factors. If an important fraction of ill-health results from poverty and low educational levels – or from their consequences in inadequate food or sanitation or other specific risks – then ought the task of the health professional lie principally in addressing these underlying problems? In one sense the answer is surely yes: the health community should measure the effects on health of actions outside the health sector. It should ensure that these findings are communicated, and are considered in making policy choices. The magnitude of the demonstrated effect of girls’ education on health and fertility outcomes, for example, provides a powerful argument for investing in extension of educational access to girls. But the health community has limited capacity for direct action outside the health sector – and limited credibility. It will make more of a difference if it focuses its energy, expertise and resources on ensuring that health systems efficiently deliver the powerful interventions provided by modern science. Source: Investing in health research and development. Report of the Ad Hoc Committee on Health Research Relating to Future Intervention Options. Geneva, World Health Organization, 1996 (document WHO/TDR/Gen/96.1)
Health and Development in the 20th Century health system development is a key priority. The effects of economic growth on health, while real, are relatively weak and likely to be slow in coming. Rather than waiting for movement along the curve, countries should focus health system development on the task of g the curve or going beyond it to the point of best practice Second, in the medium to long term, shifting the curve will underpin health improve ents. The high income countries now commit vast sums(over US$ 55 billion per year)to he research and development (r&D) efforts that will shift the curve favourably. But only a fraction of that amount is directed to solving the particular problems of poor and disadvan taged people Greater R&D commitments to such problems would be likely to pay off enormously in improving health. Ensuring an adequate commitment to R&d is surely an integral element of health system development There is every reason to expect, then, that focused investments by health systems on specific problems of the poor can generate major short to medium term gains in health, and that investment in R&D can sustain medium to long term gains. Such gains are of immense intrinsic value. The association between income and health moreover suggests that health investments may have an economic payoff as well. Supporting evidence for this assertion is presented below. Indeed, rather than continuing to point to poverty as the root cause of ill-health, decision-makers may come to focus on the two-way relationship be- tween poverty and ill-health, identifying the latter as one of the root causes of poverty and one that is particularly amenable to public intervention HEALTH AND ECONOMIC PRODUCTIVITY o The global gains in health documented above constitute, arguably, humankinds most matic achievement In our era it is possible for every individual to expect to live a long and substantially disease-free life. This accomplishment transcends the need for economic valuation. Health gains have intrinsic value. That said, two particular reasons exist for as- essing the economic consequences of better health nderstanding health,'s economic role may help to understand the sources of another of humankinds great accomplishments of the 20th century widespread rapid eco- nomic growth To the extent that better health has contributed to increased growth rates, investing in health can become a tool of macroeconomic policy Conquering poverty constitutes the central task for development policy at the begin ng of the 21st century. Despite rapid economic growth, over a billion humans still exist in absolute, degrading poverty. Because ill-health traps people in poverty, sustained investment in the health of the poor could provide a policy lever for alleviating persist ent poverty Research has begun to provide clearer evidence of the economic benefits of improving health. But data sets underpinning the research -on characteristics of countries over time or on large numbers of households within a country at a given time -rarely permit conchu sive determination of cause and effect. Conclusions drawn from the literature remain, there- fore, suggestive rather than definitive. Those conclusions do, though, accord with common sense: healthier people are more productive. Health differences have played a significar role in determining why some countries have grown more rapidly than others, although technological advances and physical capital accumulation may have been more important still. What is the evidence? This section summarizes the literature by, first, reviewing cross- ountry macroeconomic analyses, then by turning to microeconomic comparisons across
Health and Development in the 20th Century 7 health system development is a key priority. The effects of economic growth on health, while real, are relatively weak and likely to be slow in coming. Rather than waiting for movement along the curve, countries should focus health system development on the task of joining the curve or going beyond it to the point of best practice. Second, in the medium to long term, shifting the curve will underpin health improvements. The high income countries now commit vast sums (over US$ 55 billion per year) to the research and development (R&D) efforts that will shift the curve favourably. But only a fraction of that amount is directed to solving the particular problems of poor and disadvantaged people. Greater R&D commitments to such problems would be likely to pay off enormously in improving health. Ensuring an adequate commitment to R&D is surely an integral element of health system development. There is every reason to expect, then, that focused investments by health systems on specific problems of the poor can generate major short to medium term gains in health, and that investment in R&D can sustain medium to long term gains. Such gains are of immense intrinsic value. The association between income and health moreover suggests that health investments may have an economic payoff as well. Supporting evidence for this assertion is presented below. Indeed, rather than continuing to point to poverty as the root cause of ill-health, decision-makers may come to focus on the two-way relationship between poverty and ill-health, identifying the latter as one of the root causes of poverty – and one that is particularly amenable to public intervention. HEALTH AND ECONOMIC PRODUCTIVITY The global gains in health documented above constitute, arguably, humankind’s most dramatic achievement. In our era it is possible for every individual to expect to live a long and substantially disease-free life. This accomplishment transcends the need for economic valuation. Health gains have intrinsic value. That said, two particular reasons exist for assessing the economic consequences of better health: • Understanding health’s economic role may help to understand the sources of another of humankind’s great accomplishments of the 20th century – widespread rapid economic growth. To the extent that better health has contributed to increased growth rates, investing in health can become a tool of macroeconomic policy. • Conquering poverty constitutes the central task for development policy at the beginning of the 21st century. Despite rapid economic growth, over a billion humans still exist in absolute, degrading poverty. Because ill-health traps people in poverty, sustained investment in the health of the poor could provide a policy lever for alleviating persistent poverty. Research has begun to provide clearer evidence of the economic benefits of improving health. But data sets underpinning the research – on characteristics of countries over time or on large numbers of households within a country at a given time – rarely permit conclusive determination of cause and effect. Conclusions drawn from the literature remain, therefore, suggestive rather than definitive. Those conclusions do, though, accord with common sense: healthier people are more productive. Health differences have played a significant role in determining why some countries have grown more rapidly than others, although technological advances and physical capital accumulation may have been more important still. What is the evidence? This section summarizes the literature by, first, reviewing crosscountry macroeconomic analyses, then by turning to microeconomic comparisons across
The World Health Report 1999 households. It closes with a brief discussion of the multiple pathways through which bette health influences economic outcomes MACROECONOMIC EVIDENCE Since publication of Adam Smith's The wealth of nations over two centuries ago, econo- mists have sought answers to the question of why some countries are wealthy and others or Why have economic growth rates differed? The main empirical tool now used to study economic growth is cross-country analysis of the relationship between economic growth(typically measured in terms of the growth rate of per capita GDP)and a range of variables believed to account for why growth rates differ(6, 7). Among the factors being explored are: levels and patterns of educational attainment(schooling); population growth, density and age structure; natural resource abundance; personal and government saving nvestment rates); physical capital stock; economic policy, for example, the degree of trade openness; the quality of public institutions; and geography, for example, the location and climate of a country. Recent research has added several specific health indicators to these factors, and looked at the links between them and economic growth. There are direct links between economic performance and health indicators such as life expectancy. Some variables, such as geogra phy and demography, indirectly link health with economic growth. Geography, particularly tropical location, is highly correlated with disease burden, which in turn affects economic performance(8). Demography, on the other hand, is determined in part by health status, and has a direct effect on economic growth through the age structure of the population, in particular the ratio of the working age to the total population A major result to emerge from recent research is that survival rates or life expectancy powerful predictors of income levels or of subsequent economic growth. The studies con sistently find a strong effect of health on economic levels or growth rates. Interestingly economic historians have concluded that perhaps 30% of the estimated per capita growth rate in Britain between 1780 and 1979 was a result of improvement in health and nutri ional status(9).That figure lies within the range of estimates produced by cross-country studies using data from the last 30 or 40 years(10) Health improvements also influence economic growth through their impact on demog aphy. For example, in the 1940s, rapid improvements in health in East Asia provided catalyst for a demographic transition there. An initial decline in infant and child mortality swelled the youth population, and somewhat later prompted a fall in fertility rates.These asynchronous changes in mortality and fertility, which comprise the first phase of the de- mographic transition, substantially altered East Asia's age distribution. After a time lag, the working-age population began growing much faster than the young dependent popula tion, temporarily creating a disproportionately high percentage of working-age adults. This bulge in the age structure of the population created an opportunity for increased rates of economic growth. By introducing these demographic considerations into an empirical model of economic growth, analyses undertaken for the Asian Development Bank(ADB)were ble to show that East Asia's changing demography can explain perhaps a third to half the economic"miracle"experienced between 1965 and 1990(11, 12). The ADB study cautions that although a"demographic gift provides an opportunity fo increasing prosperity, it by no means guarantees such results. East Asia's growth rates were achieved because government and the private sector were able to mobilize this burgeoning work force by successfully managing other economic opportunities. Adopting new indus- trial technologies, investing in basic education and exploiting global markets allowed East
8 The World Health Report 1999 households. It closes with a brief discussion of the multiple pathways through which better health influences economic outcomes. MACROECONOMIC EVIDENCE Since publication of Adam Smith’s The wealth of nations over two centuries ago, economists have sought answers to the question of why some countries are wealthy and others poor. Why have economic growth rates differed? The main empirical tool now used to study economic growth is cross-country analysis of the relationship between economic growth (typically measured in terms of the growth rate of per capita GDP) and a range of variables believed to account for why growth rates differ (6,7). Among the factors being explored are: levels and patterns of educational attainment (schooling); population growth, density and age structure; natural resource abundance; personal and government saving (investment rates); physical capital stock; economic policy, for example, the degree of trade openness; the quality of public institutions; and geography, for example, the location and climate of a country. Recent research has added several specific health indicators to these factors, and looked at the links between them and economic growth. There are direct links between economic performance and health indicators such as life expectancy. Some variables, such as geography and demography, indirectly link health with economic growth. Geography, particularly tropical location, is highly correlated with disease burden, which in turn affects economic performance (8). Demography, on the other hand, is determined in part by health status, and has a direct effect on economic growth through the age structure of the population, in particular the ratio of the working age to the total population. A major result to emerge from recent research is that survival rates or life expectancy are powerful predictors of income levels or of subsequent economic growth. The studies consistently find a strong effect of health on economic levels or growth rates. Interestingly, economic historians have concluded that perhaps 30% of the estimated per capita growth rate in Britain between 1780 and 1979 was a result of improvement in health and nutritional status (9). That figure lies within the range of estimates produced by cross-country studies using data from the last 30 or 40 years (10). Health improvements also influence economic growth through their impact on demography. For example, in the 1940s, rapid improvements in health in East Asia provided a catalyst for a demographic transition there. An initial decline in infant and child mortality swelled the youth population, and somewhat later prompted a fall in fertility rates. These asynchronous changes in mortality and fertility, which comprise the first phase of the demographic transition, substantially altered East Asia’s age distribution. After a time lag, the working-age population began growing much faster than the young dependent population, temporarily creating a disproportionately high percentage of working-age adults. This bulge in the age structure of the population created an opportunity for increased rates of economic growth. By introducing these demographic considerations into an empirical model of economic growth, analyses undertaken for the Asian Development Bank (ADB) were able to show that East Asia’s changing demography can explain perhaps a third to half the economic “miracle” experienced between 1965 and 1990 (11,12). The ADB study cautions that although a “demographic gift” provides an opportunity for increasing prosperity, it by no means guarantees such results. East Asia’s growth rates were achieved because government and the private sector were able to mobilize this burgeoning work force by successfully managing other economic opportunities. Adopting new industrial technologies, investing in basic education and exploiting global markets allowed East
Health and Development in the 20th Century Asia to realize the economic growth potential created by the demographic transition.The next phase for East Asia will involve less favourable dependency ratios consequent to popu lation ageing. In contrast, both South Asia and Africa are now entering the period when demographic factors can enhance growth prospects. Box 1.2 describes ongoing work as nd income in the americas Analysts are extending this research in several ways. One line of work, analysing the effects of climate on income, concludes that countries in tropical regions suffer important disadvantages relative to those in temperate zones. In addition to the effects of climate and geography on soil quality, this work suggests that an important causal mechanism through which this effect operates is the interaction of tropical climates and tropical diseases, par- larly malaria which can have a significant cost in terms of economic performance(8) Another line of analysis suggests that the interaction of exogenous demographic changes with human ysical capital development can lead to a virtuous cycle of growth, ena bling a country to break free of a poverty trap(13) MICROECONOMIC ANALYSIS Unlike macroeconomic studies that compare the performance of countries over time, microeconomic analyses study the link between health and the income of households and individuals. Until recently, much of the microeconomic literature has dealt with the impact of education and training on labour outcomes. Recent individual and household level studies have, however, paid more attention to health(particularly nutritional aspects of health) and are reaching increasingly consistent findings (14) Several examples provide an indication of the results of this research. In Indonesia, men with anaemia were found to be 20% less productive than men without it. In one of the few experimental studies in the literature, the anaemic men were randomly assigned to one of two groups in a clinical trial -they received either an iron supplement or a placebo. Those who were initially anaemic and received the iron treatment increased their productivity nearly to the levels of non-anaemic workers, and the productivity gains were large when neighed against the costs of treatment. Thus the effects of improved health were found be greatest for the most vulnerable, that is, the poorest and those with the least education Box 1.3 provides more detail on another study, also from Indonesia and also involving an Box 1. 2 Assessment of the links between health and productivity: a PAHO initiative In recent years, WHO Member tries show that growth in GDP is sta- economy is greater than that of female economic policy that the relation- States in the Region of the Americas tistically associated with life expect- life expectancy, probably because of ship between health improvement have expressed interest in improv- ancy, as has been found in other the higher level of economic activity variables and economic growth is ing the understanding of linkages studies for awidersample of countries. among males. The results suggest that sufficiently significant in the long etween investments in health, eco- Life expectancy at birth alone is one for any additional year of life expect- term to justify sustained national nomic growth and poverty reduc- of the strongest explanatory variables ancy there will be an additional 19 commitment to investing in health. ion. In response, a joint PAHO/ of growth in GDP. increase in GDP 15 years later. Similar Continued work by PAHO-and its Inter-American Development Bank/ Estimates based on data from findings were observed for schooling. collaborators- should further elu- UNECLAC study has been initiated Mexico throw some light on the In this case, the correlation betwee een cidate these linkages at both the aiming at elucidating relations be- timeframe in which health affects eco- female life expectancy and schooling household level and the national tween investments in health, eco- nomic indicators. High life expectancy is greater than that for male lifeexpect- level nomic growth and household at birth for males and females has an ancy, probably because of the larger productivity. Preliminary data from economic impact 0-5 years later. The role that women play in child-rearing. Latin American and Caribbean coun- impact of male life expectancyon the This work drew the implication for Contributed by the WHO Regional Office for the Americas/ Pan American Health Organization
Health and Development in the 20th Century 9 Asia to realize the economic growth potential created by the demographic transition. The next phase for East Asia will involve less favourable dependency ratios consequent to population ageing. In contrast, both South Asia and Africa are now entering the period when demographic factors can enhance growth prospects. Box 1.2 describes ongoing work assessing linkages between health and income in the Americas. Analysts are extending this research in several ways. One line of work, analysing the effects of climate on income, concludes that countries in tropical regions suffer important disadvantages relative to those in temperate zones. In addition to the effects of climate and geography on soil quality, this work suggests that an important causal mechanism through which this effect operates is the interaction of tropical climates and tropical diseases, particularly malaria which can have a significant cost in terms of economic performance (8). Another line of analysis suggests that the interaction of exogenous demographic changes with human and physical capital development can lead to a virtuous cycle of growth, enabling a country to break free of a poverty trap (13). MICROECONOMIC ANALYSIS Unlike macroeconomic studies that compare the performance of countries over time, microeconomic analyses study the link between health and the income of households and individuals. Until recently, much of the microeconomic literature has dealt with the impact of education and training on labour outcomes. Recent individual and household level studies have, however, paid more attention to health (particularly nutritional aspects of health) and are reaching increasingly consistent findings (14). Several examples provide an indication of the results of this research. In Indonesia, men with anaemia were found to be 20% less productive than men without it. In one of the few experimental studies in the literature, the anaemic men were randomly assigned to one of two groups in a clinical trial – they received either an iron supplement or a placebo. Those who were initially anaemic and received the iron treatment increased their productivity nearly to the levels of non-anaemic workers, and the productivity gains were large when weighed against the costs of treatment. Thus the effects of improved health were found to be greatest for the most vulnerable, that is, the poorest and those with the least education. Box 1.3 provides more detail on another study, also from Indonesia and also involving an Box 1.2 Assessment of the links between health and productivity: a PAHO initiative In recent years, WHO Member States in the Region of the Americas have expressed interest in improving the understanding of linkages between investments in health, economic growth and poverty reduction. In response, a joint PAHO/ Inter-American Development Bank/ UNECLAC study has been initiated aiming at elucidating relations between investments in health, economic growth and household productivity. Preliminary data from Latin American and Caribbean countries show that growth in GDP is statistically associated with life expectancy, as has been found in other studies for a wider sample of countries. Life expectancy at birth alone is one of the strongest explanatory variables of growth in GDP. Estimates based on data from Mexico throw some light on the timeframe in which health affects economic indicators. High life expectancy at birth for males and females has an economic impact 0–5 years later. The impact of male life expectancy on the economy is greater than that of female life expectancy, probably because of the higher level of economic activity among males. The results suggest that for any additional year of life expectancy there will be an additional 1% increase in GDP 15 years later. Similar findings were observed for schooling. In this case, the correlation between female life expectancy and schooling is greater than that for male life expectancy, probably because of the larger role that women play in child-rearing. This work drew the implication for economic policy that the relationship between health improvement variables and economic growth is sufficiently significant in the long term to justify sustained national commitment to investing in health. Continued work by PAHO – and its collaborators – should further elucidate these linkages at both the household level and the national level. Contributed by the WHO Regional Office for the Americas/Pan American Health Organization
The World Health Report 1999 tervention. Here the intervention(introduction of user fees)resulted in lower levels of nutritional status and productivity among those initially poor(15) A careful statistical analysis of the effects of illness on wages and labour supply in Cote d'lvoire and Ghana found that wages were significantly lower, in both countries, for each day of disability. Ill-health in the form of disability, in these poor communities, contributed their At the household level, it is also possible to measure directly the economic burden cre ated by particular diseases. Tuberculosis provides a relevant example. The economic costs of tuberculosis are made up of two main elements. First, there are the direct costs of preven tion and treatment(drugs, health care provider fees, transport, and costs of subsistence at a health centre). Second, there are the indirect costs of labour time lost because of illness Given these two components of cost, there are several ways in which tuberculosis affects economic outcomes. Tuberculosis-related morbidity directly increases household and pub- lic sector expenditures. It reduces labour inputs and can reduce human capital as a result of declines in school attendance. In a case study of costs of improving tuberculosis control in Thailand in 1995, the cost of treatment was estimated to be US$ 343 per case. The research- ers also estimated the total indirect cost of lost productivity in Thailand as a result of mor- bidity associated with treated and untreated cases of tuberculosis, amounting to $57 million PATHWAYS OF INFLUENCE Delineating potential pathways of influence sheds light on healths role within the larger web of determinants of income levels and growth rates. A paper presented to the World Health Assembly in 1952 foreshadowed much of the current work on understanding thes pathways(17) Box 1.3 User fees, health outcomes and labour force participation in Indonesia: a two-year study In an intervention study in Indo- was poor. Equal numbers of control than the control group because of ill- ference between those in the test nesia-the Indonesian Resource and test households were selected ness. But theeffects much greater and control groups, tothe disadvan- Mobilization Study(IRMS)-the ef- from each province. mong the poor, among men over 40, tage of those paying higher health fect of changes in prices of publidy Use of health care declined in test and among women in households fees. In the test districts, labour force provided health services on labour areas, relative to controls, as did some with low economic and educational force participation was examined In health status indicators Using self-re- status. Men in the bottom quartile of education fell 14%. Women over 40 the experiment, user fees at public ports about limitations in their ability per capita income in the test areas re- were also likely to have high drop- health centres were raised in ran- ed losing almost a full day more out rates from the labour market in mly selected test districts, while such as walking 5 kilometres, carrying of activity compared to the control the areas where health costs had fees were held constant (in real a heavy load 20 metres, or having group. terms)in neighbouring control dis- spent a day in bed in the previous Moreover, the follow-up study Wage rates for men were also af- tricts. a baseline household survey month-the follow-up study in 1993 showed significant declines in labour fected. While on average both test vas conducted at the end of 1991, showed that the great majority of force participation in the test area and control groups increased their prior to the intervention, and the those where prices had been raised among the more vulnerable groups. nominal wages by 30% in the tw same households, evenly divided showed at least some ill effects. In Men in the over 40 age group had a years, the increase came 15%sooner between those that were subjected IRMS, higher prices are associated with slight tendency to drop out of the la. in the control areas. The compara- to the fee increase and those that greater difficulty walking 5 kms, more bour market in the test area. Among tive slippage in the test areas was ere not, were surveyed again two limitations on daily activities and more all women in the survey, both in the particularly great for older workers, rs later. The experiment involved days spent in bed For example both control and test groups, labour force whose health is presumably a 6000 households in several districts men and women in the test districts participation dropped from 50% to greater factorintheirworkperform- in each of two provinces. One of the reported having had to spend an av- 46% between 1991 and 1993, but ance. provinces was well-to-do and one erage of a third of a day in bed more there was a 7.3 percentage point dif- Source Dow WH et al. Health care prices, health and laboroutcomes: Experimental evidence. Santa Monica CA, RAND, 1997 (unpublished paper)
10 The World Health Report 1999 intervention. Here the intervention (introduction of user fees) resulted in lower levels of nutritional status and productivity among those initially poor (15). A careful statistical analysis of the effects of illness on wages and labour supply in Côte d’Ivoire and Ghana found that wages were significantly lower, in both countries, for each day of disability. Ill-health in the form of disability, in these poor communities, contributed to their continuing poverty (16). At the household level, it is also possible to measure directly the economic burden created by particular diseases. Tuberculosis provides a relevant example. The economic costs of tuberculosis are made up of two main elements. First, there are the direct costs of prevention and treatment (drugs, health care provider fees, transport, and costs of subsistence at a health centre). Second, there are the indirect costs of labour time lost because of illness. Given these two components of cost, there are several ways in which tuberculosis affects economic outcomes. Tuberculosis-related morbidity directly increases household and public sector expenditures. It reduces labour inputs and can reduce human capital as a result of declines in school attendance. In a case study of costs of improving tuberculosis control in Thailand in 1995, the cost of treatment was estimated to be US$ 343 per case. The researchers also estimated the total indirect cost of lost productivity in Thailand as a result of morbidity associated with treated and untreated cases of tuberculosis, amounting to $57 million. PATHWAYS OF INFLUENCE Delineating potential pathways of influence sheds light on health’s role within the larger web of determinants of income levels and growth rates. A paper presented to the World Health Assembly in 1952 foreshadowed much of the current work on understanding these pathways (17). Box 1.3 User fees, health outcomes and labour force participation in Indonesia: a two-year study In an intervention study in Indonesia – the Indonesian Resource Mobilization Study (IRMS) – the effect of changes in prices of publicly provided health services on labour force participation was examined. In the experiment, user fees at public health centres were raised in randomly selected test districts, while fees were held constant (in real terms) in neighbouring control districts. A baseline household survey was conducted at the end of 1991, prior to the intervention, and the same households, evenly divided between those that were subjected to the fee increase and those that were not, were surveyed again two years later. The experiment involved 6000 households in several districts in each of two provinces. One of the provinces was well-to-do and one was poor. Equal numbers of control and test households were selected from each province. Use of health care declined in test areas, relative to controls, as did some health status indicators. Using self-reports about limitations in their ability to perform activities of daily living – such as walking 5 kilometres, carrying a heavy load 20 metres, or having spent a day in bed in the previous month – the follow-up study in 1993 showed that the great majority of those where prices had been raised showed at least some ill effects. In IRMS, higher prices are associated with greater difficulty walking 5 kms, more limitations on daily activities and more days spent in bed. For example, both men and women in the test districts reported having had to spend an average of a third of a day in bed more than the control group because of illness. But the effects were much greater among the poor, among men over 40, and among women in households with low economic and educational status. Men in the bottom quartile of per capita income in the test areas reported losing almost a full day more of activity compared to the control group. Moreover, the follow-up study showed significant declines in labour force participation in the test area among the more vulnerable groups. Men in the over 40 age group had a slight tendency to drop out of the labour market in the test area. Among all women in the survey, both in the control and test groups, labour force participation dropped from 50% to 46% between 1991 and 1993, but there was a 7.3 percentage point difference between those in the test and control groups, to the disadvantage of those paying higher health fees. In the test districts, labour force participation for women with no education fell 14%. Women over 40 were also likely to have high dropout rates from the labour market in the areas where health costs had gone up. Wage rates for men were also affected. While on average both test and control groups increased their nominal wages by 30% in the two years, the increase came 15% sooner in the control areas. The comparative slippage in the test areas was particularly great for older workers, whose health is presumably a greater factor in their work performance. Source: Dow WH et al. Health care prices, health and labor outcomes: Experimental evidence. Santa Monica CA, RAND, 1997 (unpublished paper)