《卫生经济学》课程参考文献（WHO年度报告）：World Health Report 2000_Health Systems：Improving performance_1 Why do Health Systems Matter?

CHAPTER ONE y go Health ystems (atter Health systems consist ofall the people and actions whose primary purpose is to improve health. They may be integrated and centrally directed, but often they are not. After centuries as small-scale, largely private or charitable, mostly inef- factual entities, they have grown explosively in this century as knowledge has been gained and applied. They have contributed enormously to better health, but their contribution could be greater still, especially for the poor. Failure to achieve that potential is due more to systemic failings than to technical limita- tions. It is therefore urgent to assess current performance and to judge how health systems can reach their potential

Why do Health Systems Matter? 1 CHAPTER ONE
hy do ealth ystems atter? Health systems consist of all the people and actions whose primary purpose is to improve health. They may be integrated and centrally directed, but often they are not. After centuries as small-scale, largely private or charitable, mostly inef￾fectual entities, they have grown explosively in this century as knowledge has been gained and applied. They have contributed enormously to better health, but their contribution could be greater still, especially for the poor. Failure to achieve that potential is due more to systemic failings than to technical limita￾tions. It is therefore urgent to assess current performance and to judge how health systems can reach their potential. 1

WHY DO HEALTH SYSTEMS MATTER? THE CHANGING LANDSCAPE O n 13 October 1999, in a maternity clinic in Sarajevo, Helac Fatima gave birth to a on. This was a special occasion, because United Nations demographers had cal culated the global population would reach six billion on that day. The little Sarajevo boy was designated as the sixth billionth person on the planet Today there are four times as many people in the world as there were 100 years ago there are now about 4000 babies born every minute of every day -and among the count less, bewildering changes that have occurred since then, some of the most profound have occurred in human health. For example, few if any of Helac Fatimas ancestors around 1899 were likely to have seen a hospital, far less been born in one. The same was true for the great majority of the 1.5 billion people then alive. Throughout he world, childbirth invariably occurred at home, rarely with a physician present. Most people relied on traditional remedies and treatments, some of them thousands of years old. Most babies were born into large families and faced an infancy and childhood threatened by a host of potentially fatal diseases-measles, smallpox, malaria and poliomyelitis among them. Infant and child mortality rates were very high, as were maternal mortality rates. Life expectancy for adults was short-even half a century ago it was a mere 48 years at birth Last year the son of Helac Fatima entered the world with a life expectancy at birth of 73 rears-the current Bosnian average. The global average is 66 years. He was born in a big city hospital staffed by well-trained midwives, nurses, doctors and technicians-who were sup- ported by modern equipment, drugs and medicines. The hospital is part of a sophisticated Ith service. It is connected in turn to a wide network of people and actions that in( way or another are concerned with maintaining and improving his health for the rest of his fe-as for the rest of the population. Together, all these interested parties, whether they provide services, finance them or set policies to administer them, make up a health system. c Health systems have played a part in the dramatic rise in life expectancy that occurred aring the 20th century. They have contributed enormously to better health and influenced the lives and well-being of billions of men, women and children around the world. Their role has become increasingly important. Enormous gaps remain, however, between the potential of health systems and their actual performance, and there is far too much variation in outcomes among countries which seem to have the same resources and possibilities. Why should this be so? Health systems would seem no different from other social systems in facing demands and incentives

Why do Health Systems Matter? 3 1 WHY DO HEALTH SYSTEMS MATTER? THE CHANGING LANDSCAPE  n 13 October 1999, in a maternity clinic in Sarajevo, Helac Fatima gave birth to a son. This was a special occasion, because United Nations demographers had cal￾culated the global population would reach six billion on that day. The little Sarajevo boy was designated as the sixth billionth person on the planet. Today there are four times as many people in the world as there were 100 years ago – there are now about 4000 babies born every minute of every day – and among the count￾less, bewildering changes that have occurred since then, some of the most profound have occurred in human health. For example, few if any of Helac Fatima’s ancestors around 1899 were likely to have seen a hospital, far less been born in one. The same was true for the great majority of the 1.5 billion people then alive. Throughout the world, childbirth invariably occurred at home, rarely with a physician present. Most people relied on traditional remedies and treatments, some of them thousands of years old. Most babies were born into large families and faced an infancy and childhood threatened by a host of potentially fatal diseases – measles, smallpox, malaria and poliomyelitis among them. Infant and child mortality rates were very high, as were maternal mortality rates. Life expectancy for adults was short – even half a century ago it was a mere 48 years at birth. Last year the son of Helac Fatima entered the world with a life expectancy at birth of 73 years – the current Bosnian average. The global average is 66 years. He was born in a big city hospital staffed by well-trained midwives, nurses, doctors and technicians – who were sup￾ported by modern equipment, drugs and medicines. The hospital is part of a sophisticated health service. It is connected in turn to a wide network of people and actions that in one way or another are concerned with maintaining and improving his health for the rest of his life – as for the rest of the population. Together, all these interested parties, whether they provide services, finance them or set policies to administer them, make up a health system. Health systems have played a part in the dramatic rise in life expectancy that occurred during the 20th century. They have contributed enormously to better health and influenced the lives and well-being of billions of men, women and children around the world. Their role has become increasingly important. Enormous gaps remain, however, between the potential of health systems and their actual performance, and there is far too much variation in outcomes among countries which seem to have the same resources and possibilities. Why should this be so? Health systems would seem no different from other social systems in facing demands and incentives to

The World Health Report 2000 perform as well as possible, and it might be expected that-with some degree of regulation by the state- their performance could be largely left to markets, just as with the provision of most other goods and services. But health is fundamentally different from other things that people want, and the dif ference is rooted in biology. As eloquently expressed by Jonathan Miller, "Of all the objects in the world, the human body has a peculiar status: it is not only possessed by the person who has it, it also possesses and constitutes him. Our body is quite different from all the other things we claim as our own. We can lose money, books and even houses and still remain recognisably ourselves, but it is hard to give any intelligible sense to the idea of disembodied person. Although we speak of our bodies as premises that we live in, it is a special form of tenancy: our body is where we can always be contacted"(1).The person who seeks health care is of course a consumer- as with all other products and services-and may also be a co-producer of his or her health, in following good habits of diet, hygiene and exercise, and complying with medication or other recommendations of providers. But he or she is also the physical object to which all such care is directed Health, then, is a characteristic of an inalienable asset, and in this respect resembles other forms of human capital, such as education, professional knowledge o athletic skills. But it still differs from them in crucial respects. It is subject to large and un predictable risks, which are mostly independent of one another. And it cannot be accumu- lated as knowledge and skills can. These features are enough to make health radically unlike all other assets which people insure against loss or damage, and are the reason why health insurance is more complex than any other kind of insurance. If a car worth US$ 10 000 would cost $15 000 to repair after an accident, an insurer would only pay $10 000. The apossibility of replacing the body, and the consequent absence of a market value for it, precludes any such ceiling on health costs Since the poor are condemned to live in their bodies just as the rich are, they need protection against health risks fully as much. In contrast, where other assets such as hous ing are concerned, the need for such protection either does not arise, or arises only in proportion to income. This basic biological difference between health and other assets even exaggerates forms of market failure, such as moral hazard and imperfect and asymmetric information, that occur for other goods and services Directly or indirectly, it explains much of the reason why markets work less well for health than for other things, why there is need for a more active and also more complicated role for the state, and in general why good performance cannot be taken for granted. The physical integrity and dignity of the individual are recognized in international law, yet there have been shameful instances of the perversion of medical knowledge and skills, such as involuntary or uninformed participation in experiments, forced sterilization, or vio- lent expropriation of organs. Health systems therefore have an additional responsibility to ensure that people are treated with respect, in accordance with human rights This report sets out to analyse the role of health systems and suggest how to make them more efficient and, most importantly, more accessible and responsive to the hundreds of millions of people presently excluded from benefiting fully from them. The denial of access to basic health care is fundamentally linked to poverty -the greatest blight on humanitys landscape. For all their achievements and good intentions, health systems have failed glo- bally to narrow the health divide between rich and poor in the last 100 years. In fact, the gap is actually widening. Some such worsening often accompanies economic progress, as the already better-off are the first to benefit from it. But the means exist to accelerate the sharing by the poor in these benefits, and often at relatively low cost(see Box 1.1). Finding

4 The World Health Report 2000 perform as well as possible, and it might be expected that – with some degree of regulation by the state – their performance could be largely left to markets, just as with the provision of most other goods and services. But health is fundamentally different from other things that people want, and the dif￾ference is rooted in biology. As eloquently expressed by Jonathan Miller, “Of all the objects in the world, the human body has a peculiar status: it is not only possessed by the person who has it, it also possesses and constitutes him. Our body is quite different from all the other things we claim as our own. We can lose money, books and even houses and still remain recognisably ourselves, but it is hard to give any intelligible sense to the idea of a disembodied person. Although we speak of our bodies as premises that we live in, it is a special form of tenancy: our body is where we can always be contacted”(1). The person who seeks health care is of course a consumer – as with all other products and services – and may also be a co-producer of his or her health, in following good habits of diet, hygiene and exercise, and complying with medication or other recommendations of providers. But he or she is also the physical object to which all such care is directed. Health, then, is a characteristic of an inalienable asset, and in this respect it somewhat resembles other forms of human capital, such as education, professional knowledge or athletic skills. But it still differs from them in crucial respects. It is subject to large and un￾predictable risks, which are mostly independent of one another. And it cannot be accumu￾lated as knowledge and skills can. These features are enough to make health radically unlike all other assets which people insure against loss or damage, and are the reason why health insurance is more complex than any other kind of insurance. If a car worth US$ 10 000 would cost $15 000 to repair after an accident, an insurer would only pay $10 000. The impossibility of replacing the body, and the consequent absence of a market value for it, precludes any such ceiling on health costs. Since the poor are condemned to live in their bodies just as the rich are, they need protection against health risks fully as much. In contrast, where other assets such as hous￾ing are concerned, the need for such protection either does not arise, or arises only in proportion to income. This basic biological difference between health and other assets even exaggerates forms of market failure, such as moral hazard and imperfect and asymmetric information, that occur for other goods and services. Directly or indirectly, it explains much of the reason why markets work less well for health than for other things, why there is need for a more active and also more complicated role for the state, and in general why good performance cannot be taken for granted. The physical integrity and dignity of the individual are recognized in international law, yet there have been shameful instances of the perversion of medical knowledge and skills, such as involuntary or uninformed participation in experiments, forced sterilization, or vio￾lent expropriation of organs. Health systems therefore have an additional responsibility to ensure that people are treated with respect, in accordance with human rights. This report sets out to analyse the role of health systems and suggest how to make them more efficient and, most importantly, more accessible and responsive to the hundreds of millions of people presently excluded from benefiting fully from them. The denial of access to basic health care is fundamentally linked to poverty – the greatest blight on humanity’s landscape. For all their achievements and good intentions, health systems have failed glo￾bally to narrow the health divide between rich and poor in the last 100 years. In fact, the gap is actually widening. Some such worsening often accompanies economic progress, as the already better-off are the first to benefit from it. But the means exist to accelerate the sharing by the poor in these benefits, and often at relatively low cost (see Box 1.1). Finding

Why do Health Systems Matter? a successful new direction for health systems is therefore a powerful weapon in the fight against poverty to which WHO is dedicated. Not least for the children of the new century countries need systems that protect all their citizens against both the health risks and the financial risks of illnes WHAT IS A HEALTH SYSTEM? In today's complex world, it can be difficult to say exactly what a health system is, what it consists of, and where it begins and ends. This report defines a health system to inchude all the activities whose primary purpose is to promote, restore or maintain health Formal health services, including the professional delivery of personal medical atten tion, are clearly within these boundaries. So are actions by traditional healers, and all use of medication, whether prescribed by a provider or not. So is home care of the sick, which is how somewhere between 70% and 90% of all sickness is managed (2). Such traditional public health activities as health promotion and disease prevention, and other health nhancing interventions like road and environmental safety improvement, are also part of the system. Beyond the boundaries of this definition are those activities whose primary purpose is something other than health-education, for example -even if these activities have a secondary, health-enhancing benefit. Hence, the general education system is out side the boundaries, but specifically health-related education is included. So are actions intended chiefly to improve health indirectly by influencing how non-health systems fur tion-for example, actions to increase girls school enrolment or change the curriculum to make students better future caregivers and consumers of health care Box 1.1 Poverty, ill-health and cost-effectiveness The series of global estimates of years(DALYs) were similar, with a country, where deaths at early ages communicable and noncom- the burden of disease do not dis- larger contribution from noncom- have almost been eliminated municable diseases among the tinguish between rich and poor, municable diseases. The large differ- among the wealthy oor. If the projected rate of de- but an approximate breakdown ence between the effects of There are relatively cost-effective cline of communicable disease can be derived by ranking coun- communicable and noncommunic. interventions available against the damage could be doubled, the t incomes to form groups communicable diseases among the particularly to combat deaths and yea f s ould gain only 04 ies by per capita income, aggi ple diseases reflects the concentra- diseases that account for most of global rich gating from the lowest and tion of deaths and DALYs lost to these rich-poor differences, and obal poor would gain an addi constituting 20% of the bal poor: about 60% of all ill- health losses among young chil- tional 4. 1 years, narrowing the dif population, and studying health for the poor versus 8-11% dren. Interventions costing an es- ference between the two groups stribution of deaths in each among the richest quintile. This is timated S100 or less per DALY saved from 18. 4 to 13. 7 years. Doubling group, by age, cause and sex 2 strongly associated with differences could deal with 8 or 9 of the 10 lead- the pace of reduction of non- These estimates show that in in the age distribution of deaths: just ing causes of ill-health under the communicable disease damage, 1990, 70% of all deaths and fully over half of all deaths among the age of 5 years, and 6 to 8 of the 10 in contrast, would preferentially 92%of deaths from communica- poor occur before 15 years of age, main causes between the ages of 5 benefit the well-off as well as ble diseases in the poorest quintile compared to only 49% among the and 14 years. All of these are either costing considerably more. The as- were"excess"compared to the rich. The difference between the communicable diseases or forms of sociation betweer mortality that would have oc- poor and the rich is large even in a malnutrition. Death and disability cost-effectiveness is only part curred at the death rates of the typical high-mortality African coun- from these causes is projected to and probably transi tinto- ichest quintile. The figures for to- try, and much greater in a typical decline rapidly by 2020, roughly day epidemiological andeconomic tal losses of disability-adjusted life lower-mortality Latin American equalizing the health damage from conditions it is quite strong DR Thee of knowledge about how well govemment health services reach the poor implications for sector-wide approaches. Washington, DC, The World Bank, Guillot M The burden of disease among the world's poor: current situation, future trends, and implications for policy. Washington, DC, Human Development Network of The world Bank. 2000 3 World development report 1993-lnvesting in health. New York, Oxford University Press for The World Bank, 1993: Tables B6 and B7

Why do Health Systems Matter? 5 a successful new direction for health systems is therefore a powerful weapon in the fight against poverty to which WHO is dedicated. Not least for the children of the new century, countries need systems that protect all their citizens against both the health risks and the financial risks of illness. WHAT IS A HEALTH SYSTEM? In today’s complex world, it can be difficult to say exactly what a health system is, what it consists of, and where it begins and ends. This report defines a health system to include all the activities whose primary purpose is to promote, restore or maintain health. Formal health services, including the professional delivery of personal medical atten￾tion, are clearly within these boundaries. So are actions by traditional healers, and all use of medication, whether prescribed by a provider or not. So is home care of the sick, which is how somewhere between 70% and 90% of all sickness is managed (2). Such traditional public health activities as health promotion and disease prevention, and other health￾enhancing interventions like road and environmental safety improvement, are also part of the system. Beyond the boundaries of this definition are those activities whose primary purpose is something other than health – education, for example – even if these activities have a secondary, health-enhancing benefit. Hence, the general education system is out￾side the boundaries, but specifically health-related education is included. So are actions intended chiefly to improve health indirectly by influencing how non-health systems func￾tion – for example, actions to increase girls’ school enrolment or change the curriculum to make students better future caregivers and consumers of health care. Box 1.1 Poverty, ill-health and cost-effectiveness The series of global estimates of the burden of disease do not dis￾tinguish between rich and poor, but an approximate breakdown can be derived by ranking coun￾tries by per capita income, aggre￾gating from the lowest and highest incomes to form groups each constituting 20% of the world’s population, and studying the distribution of deaths in each group, by age,1 cause and sex.2 These estimates show that in 1990, 70% of all deaths and fully 92% of deaths from communica￾ble diseases in the poorest quintile were “excess” compared to the mortality that would have oc￾curred at the death rates of the richest quintile. The figures for to￾tal losses of disability-adjusted life years (DALYs) were similar, with a larger contribution from noncom￾municable diseases. The large differ￾ence between the effects of communicable and noncommunic￾able diseases reflects the concentra￾tion of deaths and DALYs lost to communicable diseases among the global poor: about 60% of all ill￾health for the poor versus 8–11 % among the richest quintile. This is strongly associated with differences in the age distribution of deaths: just over half of all deaths among the poor occur before 15 years of age, compared to only 4% among the rich. The difference between the poor and the rich is large even in a typical high-mortality African coun￾try, and much greater in a typical lower-mortality Latin American country, where deaths at early ages have almost been eliminated among the wealthy. There are relatively cost-effective interventions available against the diseases that account for most of these rich–poor differences, and particularly to combat deaths and health losses among young chil￾dren.3 Interventions costing an es￾timated $100 or less per DALY saved could deal with 8 or 9 of the 10 lead￾ing causes of ill-health under the age of 5 years, and 6 to 8 of the 10 main causes between the ages of 5 and 14 years. All of these are either communicable diseases or forms of malnutrition. Death and disability from these causes is projected to decline rapidly by 2020, roughly equalizing the health damage from communicable and noncom￾municable diseases among the poor. If the projected rate of de￾cline of communicable disease damage could be doubled, the global rich would gain only 0.4 years of life expectancy, but the global poor would gain an addi￾tional 4.1 years, narrowing the dif￾ference between the two groups from 18.4 to 13.7 years. Doubling the pace of reduction of non￾communicable disease damage, in contrast, would preferentially benefit the well-off as well as costing considerably more. The as￾sociation between poverty and cost-effectiveness is only partial, and probably transitory, but in to￾day’s epidemiological and economic conditions it is quite strong. 1 Gwatkin DR. The current state of knowledge about how well government health services reach the poor: implications for sector-wide approaches. Washington, DC, The World Bank, 5 February 1998 (discussion draft). 2 Gwatkin DR, Guillot M. The burden of disease among the world’s poor: current situation, future trends, and implications for policy. Washington, DC, Human Development Network of The World Bank, 2000. 3 World development report 1993 – Investing in health. New York, Oxford University Press for The World Bank, 1993: Tables B.6 and B.7

The World Health Report 2000 This way of defining a system does not imply any particular degree of integration, nor that anyone is in overall charge of the activities that compose it. In this sense, every country has a health system, however fragmented it may be among different organizations or how ever unsystematically it may seem to operate. Integration and oversight do not determine the system, but they may greatly influence how well it performs Unfortunately, nearly all the information available about health systems refers only to the provision of, and investment in, health services: that is, the health care system, includ ing preventive, curative and palliative interventions, whether directed to individuals or to populations. In most countries, these services account for the great bulk of employment, expenditure and activity that would be included in a broader notion of the health system, so it might seem that little is lost in concentrating on a narrower definition that fits the existing data. Those data have required great efforts to collect-and this report furtheroffers several kinds of information and analysis, such as estimates of life expectancy adjusted for time lived with disability, assessments of how well health systems treat patients, national health accounts, and estimates of household contribution to financing Nonetheless, efforts are needed to quantify and assess those activities implied by the wider definition, so as to begin to gauge their relative cost and effectiveness in contributing the goals of the system To take one example, in the United States between 1966 and 1979 the introduction of a variety of safety features in automobile design (laminated windshields, collapsible steering columns, interior padding, lap and shoulder belts, side marker lights, head restraints, leak resistant fuel systems, stronger bumpers, increased side door strength and better brakes) helped reduce the vehicle accident fatality rate per mile travelled by 40% Only three of these innovations added more than $10 to the price of a car, and in total they accounted for only 2% of the average price increase during 1975-1979(3) From 1975 to 1998. seat belts saved an estimated 112 000 lives in the United States, and tal traffic fatalities continued to fall. The potential health gains were even greater: in 1998 alone, 9000 people died because they did not use their belts(4) The potential savings in other countries are very large. Road traffic accidents are increas ing rapidly in poor countries and are projected to move from the ninth to third place in the worldwide ranking of burden of ill-health by the year 2020. Even in many middle income countries, the fatality rates per head or per vehicle mile are much higher than in the United States (5). Sub-Saharan Africa has the world's highest rate of fatalities per vehicle. The cost of improving vehicles may be high, relative to expenditure on health care, in low and dle income countries, so the effect of including such activities in the definition of the he system may be greater. Unsafe roads also contribute greatly to the vehicular toll in po countries, and the cost of improving roads could be much larger than the cost of making cars safer. But behavioural changes such as using seat belts once installed, and respecting d limits, are nearly costless and could save many lives; they are very likely to be more cost-effective than treatment of crash victims Where information corresponding to a broader definition of health systems is not avail able, this report necessarily uses the available data that match the notion of the health care system. Even by this more limited definition, health systems today represent one of the largest sectors in the world economy. Global spending on health care was about $2985 billion(thousand million)in 1997, or almost 8% of world gross domestic product (GDP), and the International Labour Organisation estimates that there were about 35 million health orkers worldwide a decade ago, while employment in health services now is likely to be substantially higher. These figures reflect how what was for thousands of years a basic private relationship-in which one person with an illness was looked after by family mem-

6 The World Health Report 2000 This way of defining a system does not imply any particular degree of integration, nor that anyone is in overall charge of the activities that compose it. In this sense, every country has a health system, however fragmented it may be among different organizations or how￾ever unsystematically it may seem to operate. Integration and oversight do not determine the system, but they may greatly influence how well it performs. Unfortunately, nearly all the information available about health systems refers only to the provision of, and investment in, health services: that is, the health care system, includ￾ing preventive, curative and palliative interventions, whether directed to individuals or to populations. In most countries, these services account for the great bulk of employment, expenditure and activity that would be included in a broader notion of the health system, so it might seem that little is lost in concentrating on a narrower definition that fits the existing data. Those data have required great efforts to collect – and this report further offers several kinds of information and analysis, such as estimates of life expectancy adjusted for time lived with disability, assessments of how well health systems treat patients, national health accounts, and estimates of household contribution to financing. Nonetheless, efforts are needed to quantify and assess those activities implied by the wider definition, so as to begin to gauge their relative cost and effectiveness in contributing to the goals of the system. To take one example, in the United States between 1966 and 1979 the introduction of a variety of safety features in automobile design (laminated windshields, collapsible steering columns, interior padding, lap and shoulder belts, side marker lights, head restraints, leak resistant fuel systems, stronger bumpers, increased side door strength and better brakes) helped reduce the vehicle accident fatality rate per mile travelled by 40%. Only three of these innovations added more than $10 to the price of a car, and in total they accounted for only 2% of the average price increase during 1975–1979 (3). From 1975 to 1998, seat belts saved an estimated 112 000 lives in the United States, and total traffic fatalities continued to fall. The potential health gains were even greater: in 1998 alone, 9000 people died because they did not use their belts (4). The potential savings in other countries are very large. Road traffic accidents are increas￾ing rapidly in poor countries and are projected to move from the ninth to third place in the worldwide ranking of burden of ill-health by the year 2020. Even in many middle income countries, the fatality rates per head or per vehicle mile are much higher than in the United States (5). Sub-Saharan Africa has the world’s highest rate of fatalities per vehicle. The cost of improving vehicles may be high, relative to expenditure on health care, in low and mid￾dle income countries, so the effect of including such activities in the definition of the health system may be greater. Unsafe roads also contribute greatly to the vehicular toll in poorer countries, and the cost of improving roads could be much larger than the cost of making cars safer. But behavioural changes such as using seat belts once installed, and respecting speed limits, are nearly costless and could save many lives; they are very likely to be more cost-effective than treatment of crash victims. Where information corresponding to a broader definition of health systems is not avail￾able, this report necessarily uses the available data that match the notion of the health care system. Even by this more limited definition, health systems today represent one of the largest sectors in the world economy. Global spending on health care was about $2985 billion (thousand million) in 1997, or almost 8% of world gross domestic product (GDP), and the International Labour Organisation estimates that there were about 35 million health workers worldwide a decade ago, while employment in health services now is likely to be substantially higher. These figures reflect how what was for thousands of years a basic, private relationship – in which one person with an illness was looked after by family mem-

Why do Health Systems Matter? bers or religious caregivers, or sometimes paid a professional healer to treat him or her has expanded over the past two centuries into the complex network of activities that now comprise a health system More than simple growth, the creation of modern health systems has involved increas- ing differentiation and specialization of skills and activities. It has also involved an im mense shift in the economic burden of ill-health. Until recently, most of that burden took the form of lost productivity, as people died young or became and remained too sick to work at full strength. The cost of health care accounted for only a small part of the economic loss, because such care was relatively cheap and largely ineffective. Productivity losses are still substantial, especially in the poorest countries, but success in prolonging life and re- ducing disability has meant that more and more of the burden is borne by health systems This includes the cost of drugs- for controlling diabetes, hypertension, and heart disease, for example that allow people to stay active and productive. Part of the growth in re- sources used by health systems is a transfer from other ways of paying for the economic damage due to illness and early death. The resources devoted to health systems are very unequally distributed, and not at all in proportion to the distribution of health problems. Low and middle income countries ac count for only 18% of world income and 11% of global health spending($250 billion or 4% of GDP in those countries). Yet 84% of the worlds population live in these countries, and they bear 93% of the worlds disease burden. These countries face many difficult challenges in meeting the health needs of their populations, mobilizing sufficient financing in an equitable and affordable manner, and securing value for scarce resources Today in most developed countries- and many middle income countries- govern ments have become central to social policy and health care. Their involvement is justified on the grounds of both equity and efficiency. However, in low income countries-where total public revenues for all uses are scarce(often less than 20% of GDP)and institutional capacity in the public sector is weak-the financing and delivery of health services is largely he hands of the private sector. In many of these countries, large segments of the poor still have no access to basic and effective care WHAT DO HEALTH SYSTEMS DO? For rich and poor alike, health needs today are very different from those of 100 or even 50 years ago. There are growing expectations of access to health care in some form, and growing demands for measures to protect the sick, and their families, against the financial costs of ill-health. The circle in which health systems are required to function has been pushed yet wider by raised awareness of the impact on health of developments such as industrialization, road transport, environmental damage and the globalization of trade People also now turn to health systems for help with a much wider variety of problems than before -not just for the relief of pain and treatment of physical limitations and emo- tional disorders but for advice on diet, child-rearing and sexual behaviour that they used to seek from other sources People typically come into direct contact with a health system as patients, attended by providers, only once or twice a year. More often their contact is as consumers of non- prescription medications and as recipients of health-related information and advice.They meet the system as contributors to paying for it, knowingly every time they buy care out of pocket or pay insurance premiums or social security contributions, and unknowingly when ever they pay taxes that are used in part to finance health. It matters very much how the

Why do Health Systems Matter? 7 bers or religious caregivers, or sometimes paid a professional healer to treat him or her – has expanded over the past two centuries into the complex network of activities that now comprise a health system. More than simple growth, the creation of modern health systems has involved increas￾ing differentiation and specialization of skills and activities. It has also involved an im￾mense shift in the economic burden of ill-health. Until recently, most of that burden took the form of lost productivity, as people died young or became and remained too sick to work at full strength. The cost of health care accounted for only a small part of the economic loss, because such care was relatively cheap and largely ineffective. Productivity losses are still substantial, especially in the poorest countries, but success in prolonging life and re￾ducing disability has meant that more and more of the burden is borne by health systems. This includes the cost of drugs – for controlling diabetes, hypertension, and heart disease, for example – that allow people to stay active and productive. Part of the growth in re￾sources used by health systems is a transfer from other ways of paying for the economic damage due to illness and early death. The resources devoted to health systems are very unequally distributed, and not at all in proportion to the distribution of health problems. Low and middle income countries ac￾count for only 18% of world income and 11% of global health spending ($250 billion or 4% of GDP in those countries). Yet 84% of the world’s population live in these countries, and they bear 93% of the world’s disease burden. These countries face many difficult challenges in meeting the health needs of their populations, mobilizing sufficient financing in an equitable and affordable manner, and securing value for scarce resources. Today in most developed countries – and many middle income countries – govern￾ments have become central to social policy and health care. Their involvement is justified on the grounds of both equity and efficiency. However, in low income countries – where total public revenues for all uses are scarce (often less than 20% of GDP) and institutional capacity in the public sector is weak – the financing and delivery of health services is largely in the hands of the private sector. In many of these countries, large segments of the poor still have no access to basic and effective care. WHAT DO HEALTH SYSTEMS DO? For rich and poor alike, health needs today are very different from those of 100 or even 50 years ago. There are growing expectations of access to health care in some form, and growing demands for measures to protect the sick, and their families, against the financial costs of ill-health. The circle in which health systems are required to function has been pushed yet wider by raised awareness of the impact on health of developments such as industrialization, road transport, environmental damage and the globalization of trade. People also now turn to health systems for help with a much wider variety of problems than before – not just for the relief of pain and treatment of physical limitations and emo￾tional disorders but for advice on diet, child-rearing and sexual behaviour that they used to seek from other sources. People typically come into direct contact with a health system as patients, attended by providers, only once or twice a year. More often their contact is as consumers of non￾prescription medications and as recipients of health-related information and advice. They meet the system as contributors to paying for it, knowingly every time they buy care out of pocket or pay insurance premiums or social security contributions, and unknowingly when￾ever they pay taxes that are used in part to finance health. It matters very much how the

system treats people's health needs and how it raises revenues from them, including how much protection it offers them from financial risk. But it also matters how it responds to eir expectations. In particular, people have a right to expect that the health system will treat them with individual dignity. So far as possible, their needs should be promptly at long delays in waiting for diagnosis and treatme health outcomes but also to respect the value of peoples time and to reduce their anxiety Patients also often expect confidentiality, and to be involved in choices about their own alth, including where and from whom they receive care. They should not always be ex- pected passively to receive services determined by the provider alone In summary, health systems have a responsibility not just to improve peoples health but rotect them against the financial cost of illness- and to treat them with dignity. As is discussed in more detail in Chapter 2, health systems thus have three fundamental objec- improving the health of the population they serve; responding to peoples expectations the costs of ill-health Because these objectives are not always met, public dissatisfaction with the way health services are run or financed is widespread, with accounts of errors, delays, rudeness, hostil tous financial risks by insurers and governments, on a grand scale orexposure to calam ty and indifference on the part of health workers, and denial of care or exposure to calam Because better health is the most important objective of a health system, and becaus health status is worse in poor populations, one might assume that for a low income coun- try, improving health is all that matters. Concern for the non-health outcomes of the sys- tem, for fairly sharing the burden of paying for health so that no one is exposed to great financial risk, and attending to people's wishes and expectations about how they are to be nted, would then be considered luxuries, gaining in importance only as income rises and health improves. But this view is mistaken, for several reasons. Poor people, as indicated earlier, need financial protection as much as or more than the well-off, since even small absolute risks may have catastrophic consequences for them. And the poor are just as enti tled to respectful treatment as the rich, even if less can be done for them materially. More ver, pursuing the objectives of responsiveness and financial protection does not necessarily take substantial resources away from activities to improve health Much improvement in how a health system performs with respect to these responsibilities may often be had little or no cost. So all three objectives matter in every country, independently of how rich poor it is or how its health system is organized. Better ways of achieving these objectives, treated in later chapters, are similarly relevant for all countries and health systems, although the specific implications for policy will vary according to income level and the cultural and organizational features of the system. WHY HEALTH SYSTEMS MATTER The contribution that health systems make to improving health has been examined much more closely than how well they satisfy the other two objectives mentioned above, which there is little comparable information and analysis. This report therefore develops measures corresponding to all three objectives, for assessing how systems perform. Even the contribution that health systems make to improved health is difficult to judge, because different kinds of evidence seem to give conflicting answers. At the level of interventions

8 The World Health Report 2000 system treats people’s health needs and how it raises revenues from them, including how much protection it offers them from financial risk. But it also matters how it responds to their expectations. In particular, people have a right to expect that the health system will treat them with individual dignity. So far as possible, their needs should be promptly at￾tended to, without long delays in waiting for diagnosis and treatment – not only for better health outcomes but also to respect the value of people’s time and to reduce their anxiety. Patients also often expect confidentiality, and to be involved in choices about their own health, including where and from whom they receive care. They should not always be ex￾pected passively to receive services determined by the provider alone. In summary, health systems have a responsibility not just to improve people’s health but to protect them against the financial cost of illness – and to treat them with dignity. As is discussed in more detail in Chapter 2, health systems thus have three fundamental objec￾tives. These are: • improving the health of the population they serve; • responding to people’s expectations; • providing financial protection against the costs of ill-health. Because these objectives are not always met, public dissatisfaction with the way health services are run or financed is widespread, with accounts of errors, delays, rudeness, hostil￾ity and indifference on the part of health workers, and denial of care or exposure to calami￾tous financial risks by insurers and governments, on a grand scale. Because better health is the most important objective of a health system, and because health status is worse in poor populations, one might assume that for a low income coun￾try, improving health is all that matters. Concern for the non-health outcomes of the sys￾tem, for fairly sharing the burden of paying for health so that no one is exposed to great financial risk, and attending to people’s wishes and expectations about how they are to be treated, would then be considered luxuries, gaining in importance only as income rises and health improves. But this view is mistaken, for several reasons. Poor people, as indicated earlier, need financial protection as much as or more than the well-off, since even small absolute risks may have catastrophic consequences for them. And the poor are just as enti￾tled to respectful treatment as the rich, even if less can be done for them materially. More￾over, pursuing the objectives of responsiveness and financial protection does not necessarily take substantial resources away from activities to improve health. Much improvement in how a health system performs with respect to these responsibilities may often be had at little or no cost. So all three objectives matter in every country, independently of how rich or poor it is or how its health system is organized. Better ways of achieving these objectives, treated in later chapters, are similarly relevant for all countries and health systems, although the specific implications for policy will vary according to income level and the cultural and organizational features of the system. WHY HEALTH SYSTEMS MATTER The contribution that health systems make to improving health has been examined much more closely than how well they satisfy the other two objectives mentioned above, for which there is little comparable information and analysis. This report therefore develops measures corresponding to all three objectives, for assessing how systems perform. Even the contribution that health systems make to improved health is difficult to judge, because different kinds of evidence seem to give conflicting answers. At the level of interventions

Why do Health Systems Matter? against particular diseases or conditions, there is now substantial and growing evidence that large improvements in health can be achieved at reasonable cost, for individuals and or large populations(6). Such data are the basis for estimates that in poor countries, roughly one-third of the disease burden in 1990 might be averted at a total cost per person of only $12(7 Even without progress in fundamental science, changes in the way currently available interventions are organized and delivered can reverse the spread of an epidemic and dra matically reduce the cost of saving a life. For example, in the Brazilian Amazon, greater emphasis on early malaria case detection and treatment, together with more focused ef- forts on mosquito control, turned around an epidemic and cut the cost of saving a life by case prevention from nearly $13000 to only about $2000 (8) At the level of overall progress in health, as reported in The world health report 1999, the generation and utilization of knowledge -that is, scientific and technical progress -ex plained almost half of the reduction in mortality between 1960 and 1990 in a sample of 115 low and middle income countries, while income growth explained less than 20% and in creases in the educational level of adult females less than 40%. Such estimates summarize progress in developing and applying interventions of many kinds against a large number of diseases. Prominent among these are antimalaria and immunization programmes, and the increasing use of antibiotics for the treatment of respiratory and other infectious diseases Since it is the health system that develops and applies those interventions, two kinds of evidence, one detailed and the other aggregated, indicate clearly that health systems not only can but do make a large difference to health. Taking a narrower focus on diseases for which there are effective treatments, numerous studies beginning in the 1970s(9, 10) have consistently found that preventable deaths, that is"deaths due to causes amenable to medical care "have fallen at a faster rate than other deaths. Similarly, a comparison of death rate differences between western Europe and for- erly communist countries of eastern Europe attributed 24% of the difference in male life expectancy and 39% of that in female life expectancy to the availability of modern medical care. Such care is not guaranteed simply by the existence of medical facilities (11) At the same time, other evidence seems to show that health systems make little or no difference. This emerges from some other comparisons across countries rather than through time. Often these show that while per capita income is strongly related to some measure of health status-as are other factors such as female education, income inequality or cultural characteristics-there is little independent connection with inputs such as doctors or hos- pital beds(12), with total health expenditure(13), with expenditure only on conditions menable to medical care (14), or with public spending on health(15). It is not surprising to find that these relations are weak in rich countries, since many causes of death and disabil ity are already controlled and there are many different ways to spend health system re- sources, with quite varying effects on health status. But health system expenditure often seems to make little difference even in poor countries with high infant and child mortality hich it should be a priority to reduce Furthermore, health systems make costly, even fatal mistakes far too frequently. In the United States alone, medical errors in hospitals cause at least 44 000 needless deaths a year, with another 7000 occurring as a result of mistakes in prescribing or using medication, making these errors more deadly than such killers as motor vehicle accidents, breast cancer and AIDS (16). The economic cost of these mistakes is at least $17 billion of which health care costs are more than half. And even when no one makes errors, patients often acquire

Why do Health Systems Matter? 9 against particular diseases or conditions, there is now substantial and growing evidence that large improvements in health can be achieved at reasonable cost, for individuals and for large populations (6). Such data are the basis for estimates that in poor countries, roughly one-third of the disease burden in 1990 might be averted at a total cost per person of only $12 (7). Even without progress in fundamental science, changes in the way currently available interventions are organized and delivered can reverse the spread of an epidemic and dra￾matically reduce the cost of saving a life. For example, in the Brazilian Amazon, greater emphasis on early malaria case detection and treatment, together with more focused ef￾forts on mosquito control, turned around an epidemic and cut the cost of saving a life by case prevention from nearly $13 000 to only about $2000 (8). At the level of overall progress in health, as reported in The world health report 1999, the generation and utilization of knowledge – that is, scientific and technical progress – ex￾plained almost half of the reduction in mortality between 1960 and 1990 in a sample of 115 low and middle income countries, while income growth explained less than 20% and in￾creases in the educational level of adult females less than 40%. Such estimates summarize progress in developing and applying interventions of many kinds against a large number of diseases. Prominent among these are antimalaria and immunization programmes, and the increasing use of antibiotics for the treatment of respiratory and other infectious diseases. Since it is the health system that develops and applies those interventions, two kinds of evidence, one detailed and the other aggregated, indicate clearly that health systems not only can but do make a large difference to health. Taking a narrower focus on diseases for which there are effective treatments, numerous studies beginning in the 1970s (9, 10) have consistently found that preventable deaths, that is “deaths due to causes amenable to medical care” have fallen at a faster rate than other deaths. Similarly, a comparison of death rate differences between western Europe and for￾merly communist countries of eastern Europe attributed 24% of the difference in male life expectancy and 39% of that in female life expectancy to the availability of modern medical care. Such care is not guaranteed simply by the existence of medical facilities (11). At the same time, other evidence seems to show that health systems make little or no difference. This emerges from some other comparisons across countries rather than through time. Often these show that while per capita income is strongly related to some measure of health status – as are other factors such as female education, income inequality or cultural characteristics – there is little independent connection with inputs such as doctors or hos￾pital beds (12), with total health expenditure (13), with expenditure only on conditions amenable to medical care (14), or with public spending on health (15). It is not surprising to find that these relations are weak in rich countries, since many causes of death and disabil￾ity are already controlled and there are many different ways to spend health system re￾sources, with quite varying effects on health status. But health system expenditure often seems to make little difference even in poor countries with high infant and child mortality, which it should be a priority to reduce. Furthermore, health systems make costly, even fatal mistakes far too frequently. In the United States alone, medical errors in hospitals cause at least 44 000 needless deaths a year, with another 7000 occurring as a result of mistakes in prescribing or using medication, making these errors more deadly than such killers as motor vehicle accidents, breast cancer and AIDS (16). The economic cost of these mistakes is at least $17 billion, of which health care costs are more than half. And even when no one makes errors, patients often acquire

The World Health Report 2000 new infections in hospital, and the massive use of antibiotics promotes pathogen resis ance to them, so that some part of ill-health is caused by the very efforts to treat it. These conflicting kinds of evidence can be reconciled in two ways: first, by noting that while health systems account for much health progress through time, that progress is far from uniform among countries at any one time, even among countries with similar levels of income and health expenditure; and second, by recognizing that the errors of the system diminish but do not offset the good it accomplishes. Nonetheless, there is an enormous gap between the apparent potential of h status and the actual performance"(15), and the same is doubtless true of resource use in general. One measure of that gap is that many deaths of children under 5 years of age could be averted for $10 or less, as estimated from cost-effectiveness studies of particularly valuable inter- ventions, but the average actual expenditure in poor countries per death prevented, as estimated from the overall relation between spending and mortality, is $50 000 or more The overall relation between child mortality and income implies that in a poor country of wo million population, total income would have to rise by roughly $l million in order to avert a single death. This is several orders of magnitude higher than the average health expenditure needed to save a life. Per capita, these numbers imply health expenditure of only $0.025 versus an income increase of $0.50. Income differences may explain more of health variation among countries than do differences in health expenditure But raising income is not on that account a cheaper or easier way to improve health. oncerning the more distant past, historians debate whether declines in mortality rates in some European and Latin American countries in the 19th and 20th centuries owe more to such factors as an improving diet and other socioeconomic progress than to personal medical care. But health systems, defined broadly, include all of the non-personal, popula tion-based or public health interventions such as the promotion of healthy lifestyles, insec ticide spraying against vector-bone diseases, anti-tobacco campaigns and the protection of food and water. So even if personal services accounted for very little health gain until recently, the health system as defined in this report began to make a large difference more Box 1. 2 Health knowledge, not income, explains historical change in urban-rural health differences In the first half of the 19th cen- ened by burning soft coal and by tary waste disposal, and much gar- residents in Lima, for example, are tury, life expectancy was much discharges from factories bage is simply dumped or bumed in as likely to immunize their chil- spectively, than in the rural areas many of the same problems in the consequences are not so severe as cal care when sick as residents of of England and Wales or of France; large cities of poor countries today, in European cities 150 years ago On better-off neighbourhoods, and a similar difference prevailed be- which typically have more polluted one hand, increased knowledge of much more likely to do so than een the urban and rural areas of air and water than urban areas in how diseases are caused and trans- people living in Perus mountain- weden in the first decades of the richer countries. Vehicular exhaust, mitted has led to valiant efforts to ous interior 2 Both the publi 20th century. Large cities were unknown a century ago, is already a reduce contamination, control dis- health and the personal care in- unhealthy because unclean per- major health threat in such areas as ease vectors and educate the popu- terventions have contributed to nal habits did more to spread Delhi and Mexico City. Rapid growth lation to take better care of their reversing the urban-rural differ- disease when people has made it hard to expand such health. On the other hand, even very ences in health status; better crowded together and because services as piped water, sewerage poor urban dwellers now have bet- health among urban populations garbage and even excrement were facilities and garbage collection fast ter access to effective personal is due more to the application of accumulated, drawing flies and ro- enough to keep pace. In slum areas, health care than much of the rural improved knowledge than to dents and contaminating the air even if safe water is available, many population, adding to the induce- higher incomes in cities and water. Pollution was wors- households have no access to sani- ments to migrate to the city Slum rtality. Los Angeles, University of Southern Califomia, t musgrove P. Measurement of equity in health. World Health Statistics Quarterly, 1986, 39(4)

10 The World Health Report 2000 new infections in hospital, and the massive use of antibiotics promotes pathogen resist￾ance to them, so that some part of ill-health is caused by the very efforts to treat it. These conflicting kinds of evidence can be reconciled in two ways: first, by noting that while health systems account for much health progress through time, that progress is far from uniform among countries at any one time, even among countries with similar levels of income and health expenditure; and second, by recognizing that the errors of the system diminish but do not offset the good it accomplishes. Nonetheless, “there is an enormous gap between the apparent potential of public spending to improve health status and the actual performance” (15), and the same is doubtless true of resource use in general. One measure of that gap is that many deaths of children under 5 years of age could be averted for $10 or less, as estimated from cost-effectiveness studies of particularly valuable inter￾ventions, but the average actual expenditure in poor countries per death prevented, as estimated from the overall relation between spending and mortality, is $50 000 or more. The overall relation between child mortality and income implies that in a poor country of two million population, total income would have to rise by roughly $1 million in order to avert a single death. This is several orders of magnitude higher than the average health expenditure needed to save a life. Per capita, these numbers imply health expenditure of only $0.025 versus an income increase of $0.50. Income differences may explain more of health variation among countries than do differences in health expenditure. But raising income is not on that account a cheaper or easier way to improve health. Concerning the more distant past, historians debate whether declines in mortality rates in some European and Latin American countries in the 19th and 20th centuries owe more to such factors as an improving diet and other socioeconomic progress than to personal medical care. But health systems, defined broadly, include all of the non-personal, popula￾tion-based or public health interventions such as the promotion of healthy lifestyles, insec￾ticide spraying against vector-borne diseases, anti-tobacco campaigns and the protection of food and water. So even if personal services accounted for very little health gain until recently, the health system as defined in this report began to make a large difference more Box 1.2 Health knowledge, not income, explains historical change in urban–rural health differences In the first half of the 19th cen￾tury, life expectancy was much shorter in London and Paris, re￾spectively, than in the rural areas of England and Wales or of France; a similar difference prevailed be￾tween the urban and rural areas of Sweden in the first decades of the 20th century. Large cities were unhealthy because unclean per￾sonal habits did more to spread disease when people were crowded together and because garbage and even excrement were accumulated, drawing flies and ro￾dents and contaminating the air and water.1 Pollution was wors￾ened by burning soft coal and by discharges from factories. Crowding and poverty produce many of the same problems in the large cities of poor countries today, which typically have more polluted air and water than urban areas in richer countries. Vehicular exhaust, unknown a century ago, is already a major health threat in such areas as Delhi and Mexico City. Rapid growth has made it hard to expand such services as piped water, sewerage facilities and garbage collection fast enough to keep pace. In slum areas, even if safe water is available, many households have no access to sani￾tary waste disposal, and much gar￾bage is simply dumped or burned in the open. Nonetheless the health consequences are not so severe as in European cities 150 years ago. On one hand, increased knowledge of how diseases are caused and trans￾mitted has led to valiant efforts to reduce contamination, control dis￾ease vectors and educate the popu￾lation to take better care of their health. On the other hand, even very poor urban dwellers now have bet￾ter access to effective personal health care than much of the rural population, adding to the induce￾ments to migrate to the city. Slum residents in Lima, for example, are as likely to immunize their chil￾dren and to take them for medi￾cal care when sick as residents of better-off neighbourhoods, and much more likely to do so than people living in Peru’s mountain￾ous interior.2 Both the public health and the personal care in￾terventions have contributed to reversing the urban–rural differ￾ences in health status; better health among urban populations is due more to the application of improved knowledge than to higher incomes in cities. 1 Easterlin RA. How beneficient is the market? A look at the modern history of mortality. Los Angeles, University of Southern California, 1998 (unpublished paper). 2 Musgrove P. Measurement of equity in health. World Health Statistics Quarterly, 1986, 39(4)

Why do Health Systems Matter? than a century ago, chiefly through improvements in urban sanitation and personal hy giene. These changes - removing excrement and garbage, protecting water supplies, and rashing ones hands- happened because of more understanding of how diseases are spread, even before there was any useful knowledge of how they are caused. Some im proved individual hygienic practices are centuries old, while collective measures are gener- ally more recent Growth in income alone would not have improved health under the conditions of the time, and may even have worsened it because of urban filth and crowd ng similar conditions often prevail in the cities of poor countries today, but the threat to health is better controlled (see Box 1.2) So health systems are valuable and important, but they could accomplish much more with the available understanding of how to improve health. The failings which limit pe formance do not result primarily from lack of knowledge but from not fully applying what is already known: that is, from systemic rather than technical failures. This is true even of most medical errors, because"the problem is not bad people; the problem is that the sys- tem needs to be made safer"(16). How to measure current performance and how to achieve the potential improvements in it are the subject of this report. Research to expand knowl- edge is crucial in the long run, as progress over the last two centuries shows; in the short run,much could be accomplished by wider and better application of existing knowledge This can improve health more quickly than continued and more equally distributed socio- economic progress, important as that is. The next sections discuss how modern health ystems arose, and how they have been repeatedly subjected to reforms intended to make them work better in one way or another HOW MODERN HEALTH SYSTEMS EVOLVED Health systems of some sort have existed for as long as people have tried deliberately to protect their health and treat diseases. Throughout the world, traditional practices based on herbal cures, often integrated with spiritual counselling, and providing both preventive and curative care, have existed for thousands of years, and often coexist today with modern medicine. Many of them are still the treatment of choice for some health conditions, or are resorted to because modern alternatives are not understood or trusted or fail, or are expensive. Traditional Chinese medicine can be traced back more than 3000 years, and still plays a huge role in the Chinese health system, as do its equally ancient equivalents in the Indian sub-continent and similar systems of belief and practice among indigenous African and American peoples. But until the modern growth of knowledge about disease, there vere few cures for ailments and little effective prevention of disease With rare exceptions, even in industrialized countries, organized health systems in the modern sense, intended to benefit the population at large, barely existed a century ago Although hospitals have a much longer history than complete systems in many countries, few people living 100 years ago would ever visit one -and that remains true for many millions of the poor today. Until well into the 19th century they were for the most part run by charitable organizations, and often were little more than refuges for the orphaned, the crippled, the destitute or the insane. And there was nothing like the modern practice of referrals from one level of the system to another, and little protection from financial risk apart from that offered by charity or by small-scale pooling of contributions among work ers in the same occupation. Towards the close of the 19th century, the industrial revolution was transforming the res of people worldwide. At the same time societies began to recognize the huge toll of

Why do Health Systems Matter? 11 than a century ago, chiefly through improvements in urban sanitation and personal hy￾giene. These changes – removing excrement and garbage, protecting water supplies, and washing one’s hands – happened because of more understanding of how diseases are spread, even before there was any useful knowledge of how they are caused. Some im￾proved individual hygienic practices are centuries old, while collective measures are gener￾ally more recent. Growth in income alone would not have improved health under the conditions of the time, and may even have worsened it because of urban filth and crowd￾ing; similar conditions often prevail in the cities of poor countries today, but the threat to health is better controlled (see Box 1.2). So health systems are valuable and important, but they could accomplish much more with the available understanding of how to improve health. The failings which limit per￾formance do not result primarily from lack of knowledge but from not fully applying what is already known: that is, from systemic rather than technical failures. This is true even of most medical errors, because “the problem is not bad people; the problem is that the sys￾tem needs to be made safer”(16). How to measure current performance and how to achieve the potential improvements in it are the subject of this report. Research to expand knowl￾edge is crucial in the long run, as progress over the last two centuries shows; in the short run, much could be accomplished by wider and better application of existing knowledge. This can improve health more quickly than continued and more equally distributed socio￾economic progress, important as that is. The next sections discuss how modern health systems arose, and how they have been repeatedly subjected to reforms intended to make them work better in one way or another. HOW MODERN HEALTH SYSTEMS EVOLVED Health systems of some sort have existed for as long as people have tried deliberately to protect their health and treat diseases. Throughout the world, traditional practices based on herbal cures, often integrated with spiritual counselling, and providing both preventive and curative care, have existed for thousands of years, and often coexist today with modern medicine. Many of them are still the treatment of choice for some health conditions, or are resorted to because modern alternatives are not understood or trusted, or fail, or are too expensive. Traditional Chinese medicine can be traced back more than 3000 years, and still plays a huge role in the Chinese health system, as do its equally ancient equivalents in the Indian sub-continent and similar systems of belief and practice among indigenous African and American peoples. But until the modern growth of knowledge about disease, there were few cures for ailments and little effective prevention of disease. With rare exceptions, even in industrialized countries, organized health systems in the modern sense, intended to benefit the population at large, barely existed a century ago. Although hospitals have a much longer history than complete systems in many countries, few people living 100 years ago would ever visit one – and that remains true for many millions of the poor today. Until well into the 19th century they were for the most part run by charitable organizations, and often were little more than refuges for the orphaned, the crippled, the destitute or the insane. And there was nothing like the modern practice of referrals from one level of the system to another, and little protection from financial risk apart from that offered by charity or by small-scale pooling of contributions among work￾ers in the same occupation. Towards the close of the 19th century, the industrial revolution was transforming the lives of people worldwide. At the same time societies began to recognize the huge toll of