Correcting Past Health Policy Mistakes

William C. Hsiao

Abstract: China’s health policy in the 1980s followed its economic policy of marketization. China shifted health ½nancing from public to private and commercialized the country’s public health services. Unwit- tingly, the Chinese government did not grasp the serious market failures in health care, which resulted in a pro½t-driven public health service in which patients pay directly for services. China’s health policy created three major unintended consequences: disparity between rural and urban residents, poor quality of health care, and rapid inflation in health expenditures. Since 2003, China has tried to correct its policy mistakes through public ½nancing and by establishing social health insurance. However, strong pro½t motives have become embedded within the culture of medical professionals and have eroded the professional ethics that prioritize medical practices for patients’ bene½ts. Restoring medical ethics is a formidable challenge. This paper analyzes the transformation of the Chinese health system and its ongoing challenges.

Alongside the shiny marble-walled shopping cen - ters, lined with Gucci, Louis Vuitton, and Prada stores, and nestled among soaring glass skyscrapers, glittering modern hospitals now dot all major Chi- nese cities. These state-of-the-art facilities, full of expensive high-technology chromatic machines and white-coated staff, stand in stark contrast to the mud-brick, single-room health stations found in the Chinese countryside. Although more and more of China’s 1.3 billion people are moving to metrop- WILLIAM C. HSIAO is the K. T. Li olises, the majority of its rural population, which Professor of Economics at the Har - still numbers some seven hundred million, remains vard School of Public Health. He in desperate need of well-trained village doctors currently serves as an advisor to the Chinese State Council on health and basic essential drugs. The disparity in access to sector reform. Hsiao has published quality health care between rural and urban areas more than 180 papers on health has in essence created a two-tiered system. Although systems and health care ½nancing. the top level is similar to health care available in His books include Getting Health ½rst-world nations, the lower tier of the Chinese Reform Right (with Marc Roberts, health care system is more typical of that found in Peter Berman, and Michael Reich, the third world. 2003), Social Insurance for Developing Nations (with R. Paul Shaw, 2007), After a century of being derided as the “sick man and What Macroeconomists Should of Asia,” China has now become the second largest Know about Health Care Policy (with economy in the world. As explained in Barry Peter S. Heller, 2007). Naughton’s essay in this volume, the economic

© 2014 by the American Academy of Arts & Sciences doi:10.1162/DAED_a_00272 53

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting reforms that began in 1978 have trans- has been the transformation of its health Past Health formed the economic and social land- care system. The country’s previous so - Policy 1 gdp Mistakes scape. As a result, per capita average cialistic health care system largely im- rose from $300 to $5,450, and six hundred ploded, in part because China adopted a million Chinese were lifted out of extreme market strategy, relying on private sources poverty.2 Compared to the historical eco- to ½nance health care and commercialize nomic development of England, China col - the provision of health services. Ironically, lapsed about two hundred years of devel- bastions of free-market capitalism such as opment into a thirty-year period.3 the United States and Singapore long ago Until a decade ago, China’s leaders fa- learned that the marketization of health vored the cities in terms of policy and care–with its severe market failures– investment, leaving the rural areas to lan- would create havoc and yield profound guish. Meanwhile, the Chinese hukou sys- inequities in health. Nonetheless, in 1980, tem (household registration) has restricted China adopted a market approach to the migration of people from one area to health care, following the general policy another, particularly rural residents want - of the economic sector. ing to move to urban areas on a permanent The reliance on private sources to ½- basis. The economic boom did transform nance health care and commercialize some rural areas to cities, converting about health care delivery has left behind three four hundred million rural residents to deep and enduring wounds for current and urban; but that number includes three future Chinese leaders to address. First, hundred million so-called migrant work- private ½nancing resulted in disparities ers who work in big industrial centers in in access to quality health care, leading to Shanghai, Shenzhen, and Guangzhou.4 health status differences between urban And when rural residents become migrant and rural residents. China has worked to workers in cities, they are often prohibited address this issue during the past decade from bringing their families, and they have by increasing public ½nancing for the poor, no rights to urban social services such as establishing universal health insurance, education, housing, health insurance, and and investing in health facilities in poor social welfare support. The result of these areas. However, China has not been able changes has been huge and growing dis- to close the gap to any signi½cant degree. parities in the well-being of urban and More important, effective health care rural residents, including migrant workers. relies on well-trained, quali½ed practi- The average disposable income per person tioners, and China has not been able to for urban residents is 3.5 times that of rural close the gap in human resources be - residents,5 and the Gini coef½cient, which tween cities and the countryside, nor be - measures the inequality of income distri- tween poor and rich provinces. bution, has doubled from 0.21 in the 1980s Second, the free market strategy en- to 0.47 in 2012.6 As Martin Whyte dis- couraged hospitals and physicians to pur- cusses in his essay in this issue, China now sue pro½ts with distorted prices. Pro½t- seems to consist of two separate societies: driven medicine has become the norm, urban and rural.7 And the disparity be - resulting in poor quality of health care, tween the two has caused increasing social incorrect diagnoses, inappropriate treat- strife.8 ments, and perhaps even the delivery of Beyond economic growth, one of the harmful health care. As a result, patients most startling aspects of China’s rapid have lost respect for and trust in physicians development over the past thirty years and other health professionals; there

54 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 have even been cases of patients and their production and provided social services, William C. families resorting to violence, physically including health care, which was provided Hsiao assaulting physicians when they did not through the Cooperative Medical System recover from their illnesses.9 (cms). The cms operated health posts in Lastly, pro½t-driven medicine has cre- villages and township health centers that ated deeply embedded waste and corrup- were staffed mostly by so-called barefoot tion in Chinese medical practices, leading doctors. to high inflation in health expenditures. From 1952 to 1982, the Chinese health While China’s economy grew at a phe- care system made enormous improve- nomenal rate, health expenditures grew ments in health and health care by empha- even faster, due in part to the pro½t-seeking sizing prevention and primary care pro- behavior of hospitals and physicians. This vided by modestly trained practitioners. rapid increase raises the critical question Infant mortality fell from 200 to 34 per of whether China’s current health care 1,000 live births, and life expectancy in - system is sustainable when a rapidly grow- creased from about 35 to 68 years. The Chi - ing share of government and household nese public health apparatus also achieved budgets is being diverted to health with- major gains in controlling infectious dis- out commensurate bene½ts to patients. ease through immunization, improved The Chinese government has conducted sanitation, and the control of disease vec- in-depth investigations (including by en - tors, such as mosquitoes for malaria and gaging international organizations and snails for schistosomiasis.10 scholars), gathered extensive evidence, and When China embarked on its economic analyzed the major health problems out- reform in 1978, it overhauled four major lined above. Top of½cials, as well as the health policies: it shifted public ½nancing general public, have a comprehensive to private sources; it turned public hospi- and accurate understanding of the issues, tals and clinics into commercial enterpris- and the causes have been largely identi½ed. es; it decentralized its health system; and it However, some of these causes are rooted altered the price structure for public facili- in the unethical behavior of health pro- ties, thereby enabling them to earn pro½ts. fessionals, as well as in government As China’s centrally planned socialist structure, both of which are dif½cult to economy was transformed to a market correct. econ omy centered on private enterprise, the Chinese government experienced a After the Chinese Communist Party drastic reduction in revenue and had to (ccp) took control of China in 1949, it completely rebuild its ½scal system. Con- created a health care system typical of sequently, the government lacked the Communist countries. The government revenue to fund social programs such as owned, funded, and ran all hospitals: from health care. Like other transition econ - large specialized hospitals in urban areas omies, the Chinese government’s revenue to small township hospitals in the coun- as a percentage of gdp fell from 30 per- tryside. The private practice of medicine cent to 10 percent, and subsidies for pub- and the private ownership of health facil- lic health facilities fell from 50 to 60 per- ities disappeared; physicians became cent to a mere 10 percent of the facilities’ employees of the state. In rural areas, the total revenues by the early 1990s.11 The cornerstone of the health care system government therefore decided to replace was the commune, the critical institution public ½nancing with private sources: pub- in rural life. Communes oversaw economic lic health facilities would charge patients

143 (2) Spring 2014 55

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting directly for services and patients would local public health agencies the authority Past Health pay out of pocket. On the one hand, the to charge for certain services, such as Policy Mistakes existing health insurance program for inspections of hotels and restaurants for employed workers was reformed and sanitary conditions and industrial enter- maintained, largely protecting health prises for environmental compliance. care for urban workers in the formal Health agencies could also establish fee- economy. On the other hand, in its efforts for-service health centers and hospitals to privatize the agricultural economy, the for delivering curative services. Predict - government completely dismantled the ably, local public health authorities con- communes, thereby destroying the health centrated on these revenue-generating care safety net for most rural Chinese. activities, neglecting preventive programs Without the cms, Chinese peasants had such as health education, maternal and no way to pool risks for health care ex- child health, and epidemic control. penses, and nine hundred million rural, The last major change was the intro- mostly poor citizens became, in effect, un- duction of a new pricing policy for health insured overnight. Consequently, the once- services, stemming from the Chinese gov- vaunted barefoot doctors were forced to ernment’s desire for prices to be afford- become private health practitioners. able to patients, yet suf½ciently high for The second policy change was for public public facilities to survive and flourish. health clinics and hospitals to rely on But the newly established pricing policy payments from patients as their main was at its core irrational and reflected a source of income. Virtually unregulated, lack of adequate understanding of eco- this policy basically turned public facilities nomic theory and international best prac- into free commercial enterprises. As public tices. That unsound pricing policy set in funding declined, public facilities relied motion signi½cant changes in the organi- more and more on the sale of services in zational culture, motivation, and behavior private markets to cover their expenses. of hospital leaders and practitioners. The De facto public hospitals and clinics came government set prices below costs for to function much like for-pro½t commer- personal services such as a physician visit cial entities, focusing primarily on their or hospital daily bed charges, but set bottom lines, with Chinese policy infor- prices above costs for new and high-tech mally sanctioning their actions.12 Clinics diagnostic services. It also allowed a 15 per - and hospitals quickly found that selling cent pro½t margin on drugs. This system drugs and conducting tests were the most created perverse incentives for providers, lucrative ways to stay afloat, pay bonuses who had to cover 90 percent of their to staff, and generate funds for expan- budgets with revenue-generating activi- sion. Thus, the number of drug sales and ties.13 Over time, public hospitals, clinics, test orders exploded. and village doctors gradually became The third policy change decentralized pro½t-seeking entities, while the healing the public health system in order to re- of ill patients took a back seat. Equally duce central governmental expenditures. important, the government’s pricing pol- While rich provinces had adequate ½nan- icy created a leveraging effect whereby a cial resources to cover the costs of the provider had to dispense $7 in drugs to public health services, the poor provinces earn $1 in pro½t. Subsequently, providers did not, resulting in signi½cant dispari- overprescribed drugs and tests, and hos- ties between provinces and counties. Fur- pitals raced to introduce high-tech ser - thermore, the central government granted vices and expensive imported drugs that

56 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 returned higher pro½t margins. These sector: hospitals and physicians receive William C. med ical practices not only caused rapid kickbacks from drug companies for pre- Hsiao health expenditure inflation, but also scribing their products, and doctors’ harmed patients with unnecessary sur- bonuses are often tied to these kickbacks. geries and hospitalizations, adverse reac- In rural areas, village doctors buy expired tions from the overuse of drugs, drug tox- and counterfeit drugs at low cost and sell icity from the use of multiple drugs, and them as valid products at higher prices. false-positives from poorly executed tests. Some investigators estimate that until two The unfortunate consequences of this years ago, one-third of drugs dispensed combination of policy changes are best in rural areas were actually counterfeit, understood from three perspectives: dis- earning huge pro½ts for their vendors. parities between rural and urban areas, China’s rapidly rising health expenditure de½cient quality of health care, and bur- inflation rate is one broad consequence geoning health expenditure inflation. of the country’s health policies. Chinese Urban/rural disparities in ½nancial and health expenditures have burgeoned over physical access to medical care, and pub- the past thirty years, albeit from a low lic health expenditures are reflected in base. From 1978 to 2011, personal health the available health statistics. spending per capita in China increased by In China’s privatized, market-based a multiple of 164, from 11 rmb to 1,801 health care system, the wealth of con- rmb (roughly $6 to $280). At the same sumers is a critical determinant of both time, the Consumer Price Index increased their access to services and the quality of nearly sixfold. National health care spend- services they receive. With urban in - ing rose from 3 percent to 5.15 percent of comes triple those in rural areas, urban gdp. A huge portion of this expenditure residents have fared far better than rural was for unnecessary drugs and high-tech residents in terms of health services. In tests; half of Chinese health care spend- 1999, infant mortality was 38 per 1,000 ing is devoted to drugs, compared to only live births in rural areas, compared to 12 10 percent in the United States.16 per 1,000 in urban areas. In 2002, mortality among children under age 5 was 39.6 per By the mid-1990s, the Chinese central 1,000 in rural areas, but only 14.6 per government was well aware of the wide- 1,000 in urban locales. Maternal mortality spread problems created by its health ½gures in 2002 were 58.2 and 22.3 per policies, although top of½cials were 100,000, respectively.14 reluctant to address them. In some rural One indicator of the reduced quality of districts, de½ciencies in health care be - health care in China is the inappropriate came a cause of growing anger toward use of prescription drugs, a problem the Chinese government and contributed closely linked with medical practitioners to local riots and disturbances.17 In a and hospitals having become pro½t-driven country where threats to established po - entities. In China, 75 percent of patients litical authority have historically sprung suffering from a common cold are pre- from the grievances of an impoverished scribed antibiotics–more than twice the peasantry, the consequences of rural/ international average of 30 percent.15 urban health care differentials carry pro- Unfortunately, no authority has been held found political signi½cance for the Chi- accountable for these practices. Com- nese leadership. pounding the problem is collusion be - China ½nally took action in 2003, when tween providers and the pharmaceutical its rapidly growing economy yielded large

143 (2) Spring 2014 57

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting tax revenues that gave the capacity for investing and upgrading primary care Past Health the central and some local governments facilities and human resources, and pilot- Policy Mistakes to make substantial health care invest- ing public hospital reforms. ments. The central government created As of the end of 2012, the Chinese gov- the New Cooperative Medical Scheme ernment had invested an impressive ad- (ncms), a health insurance program for ditional $125 billion in public health. Cov- rural citizens. All provinces were com- erage is now remarkably broad, albeit still pelled to establish ncms by pooling risk far from egalitarian. China currently has at the county level. The provinces and more than 95 percent of its population counties must co½nance the majority of covered under its three social insurance ncms, though the central government programs. Although employed workers does provide ½scal transfers to poor and currently have a much richer bene½ts middle-income provinces to fund a sig- package, the goal is to equalize bene½ts ni½cant portion of the costs. ncms began within a decade.19 Preventive and primary in 2003 with a total public subsidy of 20 care are better funded, supplied, and rmb for each resident’s health insurance used, and many new public facilities have premium; the subsidy has since grown to been built or renovated. Public health facil- 300 rmb per resident. In 2007, the gov- ities are equipped with reasonable medical ernment created a similar insurance pro- equipment, and essential drugs are avail- gram for the 250 million urban residents able almost everywhere at a reasonable who were not insured under the Employee price. Insurance Plan.18 These new insurance Despite these signi½cant advances, three programs lowered ½nancial barriers to major problems remain unsolved. The ½rst medical care, and rural residents have is the need for reform of public hospitals, since increased their use of health services. particularly high-level urban hospitals that But it was not until 2008 that the Chi- absorb the majority of ½nancial and hu- nese government publicly acknowledged man resources. These are the cradle of and began to address the other huge overtesting and overprescribing drugs, of health care problems it had created. In waste, and of corruptive practices, such designing its reform, China carefully as demanding red envelopes (bribes) from examined solutions pioneered in other patients and receiving kickbacks from countries and organized a wide consulta- pharmaceutical and medical device com- tion with domestic and international panies. The shortage of quali½ed human experts, as well as with the public. A sys- resources to serve the rural population and temic reform was launched in 2009 with the resultant poor quality of health care the goal of providing affordable and equi- in these regions constitute the remaining table basic health care for all citizens by unsolved challenges. 2020. With this ambitious program, the Chinese government is af½rming its role as China’s recent health policy changes the primary ½nancing source for health are intended to address the deep wounds care, while also giving priority to preven- that resulted from the country’s earlier tive and primary care. The reforms were interrelated policies of privatization and anchored by ½ve related measures: pro- commercialization of health care, as well viding insurance coverage for more than as its irrational pricing policy. As noted, 90 percent of the population, establishing a these earlier policies contributed to stark national essential drug system, establish- urban/rural health disparities, poor qual- ing and funding effective prevention, ity of health care, and health expenditure

58 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 inflation. One of the most critical issues gate data on health differences between William C. surrounding Chinese health policy is its urban and rural residents. Hsiao differential effect on the health status of The previous section discussed how the various social and economic groups. China decentralization of China’s ½scal system collects voluminous data, but much of it occurred in conjunction with its eco- is not accessible for public use. Essentially, nomic liberalization, resulting in provin- China only publishes national data relat- cial and county governments being made ed to distributional impact between urban responsible for funding health care. and rural residents, as well as some infor- However, most of the poor provinces in mation about variation among provinces. China’s rural western region did not have Data by income, education, and ethnicity the ½nancial resources to fund even basic are scarce. There are individual studies health care for their residents. Meanwhile, conducted by researchers for a given the people–mostly farmers–lacked the community or a city, but they are not rep- income to pay for health care services resentative of the nation as a whole. themselves. China does not publish reli- The most common measures of health able data on health spending over time by are infant mortality rate (imr), maternal province. It does, however, publish data mortality rate (mmr), and life expectancy. on total public and private per capita Table 1 shows that imr for rural residents spending on health for urban and rural is 260 percent that of urban residents; residents. This crude information does mmr for rural residents is 38 percent not adjust for differences in age/sex and greater than that of urban residents; varying disease patterns. Nevertheless, while urban life expectancy is 7 percent this is the only reliable and up-to-date greater than that of rural residents. Chil- national information we have to shed dren born into rural households suffer a some light on the urban-rural disparity. much higher death rate than urban chil- Figure 1 shows the difference in per capi- dren, indicating the lack of basic health ta spending on health for urban and rural care in rural areas. The lower mmr in residents from 1990 to 2011. The latest urban areas seems to indicate that urban data available show that in 2011, spending China has a higher quality child delivery on health care for the average urban resi- system. The smaller difference in life ex- dent was more than three times the aver- pectancy may reflect the less contaminated age amount spent on care for rural resi- food supply and lower air pollution in rural dents. Despite the 2008 ncms initiative, areas, and the possibility that rural people the difference in health expenditure has lead a healthier lifestyle with more exer- hardly narrowed. cise than city dwellers. More substantial health insurance cov- Differences in health outcomes are erage for workers employed in the formal caused by many factors: varied ½nancial sector is a major reason why spending is and human resources invested in preven- so different between urban and rural res- tive and health services; different in- idents. Table 2 shows the estimated aver- come, education, environmental living age premium paid per covered person in conditions; varied occupational hazards 2011. The premium was estimated at 1,960 and lifestyles; and different ethnic back- rmb under employee insurance, which grounds and social classes. While we can- covers civil servants and formal sector not ½nd reliable national data on differ- workers. More than 90 percent of these ences in health status by socioeconomic workers are urban residents. Meanwhile, group, China does publish reliable aggre- the health insurance plan for rural resi-

143 (2) Spring 2014 59

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting Table 1 Past Health Comparison of Health Status between Rural and Urban Residents, 2006–2010 Policy Mistakes

National Urban Rural Ratio (urban to rural)

imr (per 1,000 live births) 14.86 6.84 17.96 1:2.62 mmr (per 10,000 births) 34.76 27.1 37.4 1:1.38 Life Expectancy 74.83 77.33 72.29 1.07:1

Source: imr and mmr are calculated by averaging the ½ve-year period; data are from National Bureau of Sta- tistics of People’s Republic of China, China Statistical Yearbook 2012 (Beijing: China Statistics Press, 2012). The national life expectancy is from the 2010 Census, and the rural and urban life expectancy are extrapolated using the existing data before the 2010 Census. See Ying Hu, “The Average Life Expectancy Analysis on China’s Current Population in Urban and Rural,” Population & Development 16 (2) (2010): 41–47.

Figure 1 Health Care Expenditure Per Capita, 1990–2011

Source: National Bureau of Statistics of People’s Republic of China, China Statistical Yearbook 2012 (Beijing: China Statistics Press, 2012).

60 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Table 2 William C. Estimated Premium Per Capita in 2011 Hsiao

New Cooperative Employee Medical Insurance, Medical Schemes (ncms) including civil servants

Average Premium (rmb) 246 1,960

Source: National Bureau of Statistics of People’s Republic of China, China Statistical Yearbook 2012 (Beijing: China Statistics Press, 2012); average premium estimated by author.

dents, the ncms, had an estimated aver- ating from junior high school followed by age premium of 246 rmb per person, an additional three years of education in only one-eighth of the average premium a health school. At best, they have the for the employed workers. This vast dif- knowledge, training, and competency of ference reflects several factors, including a nurse’s assistant in a modern hospital. the more generous bene½ts of employee The older village doctors do not even have insurance (most migrant workers are not this level of formal training; they learned eligible), the higher cost of health care in on the job. cities, and the higher average age of people Providing the next level of care are the covered under the employee insurance as physicians, mostly three-year medical compared to the ncms. And when more school graduates, who staff the township generous bene½ts are available to em - health centers that serve between ten ployees, they demand more services and thousand and twenty-½ve thousand resi- more expensive facilities, drugs, and tests. dents. The physicians who would be con- Along with unequal ½nancial inputs for sidered reasonably quali½ed by Western urban and rural residents, human re - standards, with training comparable to source inputs also differ. Most compar- ½ve or more years of college-level medical isons using aggregate numbers of practi- education, practice mostly in cities. China tioners and hospital beds for each urban certi½es its hospitals into levels 1, 2, 2A, 3, and rural resident lead to erroneous con- and 3A. The beds in township health cen- clusions. The aggregate numbers show ters are classi½ed as level 1, which means that urban and rural areas have similar that they are mostly for observation or numbers of beds and practitioners per convalescence. Rural county hospitals resident. However, the competency of the serving populations of approximately practitioners varies wildly, as does the three hundred thousand are designated as quality of the hospitals, both in terms of level 2. They typically offer ½ve or more their medical capabilities and sophistica- basic medical specialty services such as tion. In rural areas, basic preventive and internal medicine, general surgery, ob - routine health care is provided by village stetrics, pediatrics, and ear/nose/throat. doctors, who become quali½ed by gradu- County hospitals offer very few subspe-

143 (2) Spring 2014 61

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting cialties. In contrast, urban residents are were medically unnecessary, and many of Past Health mostly served by level 3 and 3A hospitals, these prescriptions were implicated in Policy Mistakes which offer every known subspecialty more than one million children becoming and the highest level of care. deaf or suffering neurological disorders.20 The clearest way to illustrate the differ- Such patterns also contribute to the global ence in the capability and competency of problem of antimicrobial resistance. health services in urban and rural regions In a large study of urban health centers is to compare the number of registered and stations, the authors randomly ex - physicians (three or more years of medical amined 203,080 outpatients’ records ret- school education) serving rural and urban rospectively from the years 2007 to 2009.21 residents. Figure 2 shows that urban areas, They found that prescriptions of antibi- on average, have 2.2 times the number of otics, simultaneous use of two or more registered physicians per one thousand antibiotics, administration of iv injections, residents as rural areas. This gap has and prescription of steroid hormones far widened over the past two decades. exceeded the World Health Organization’s It is perhaps understandable that highly (who) reference standards during this educated physicians would not want to period. Prescriptions of multiple antibi- work in rural areas, due to less desirable otics and use of iv injections were at least socioeconomic conditions and reduced twice the who standards. Overuse of educational opportunities for their chil- steroid hormones can harm patients, and dren. Many nations try to rely on trained patients can suffer from side effects of nurses to substitute for physicians in antibiotics as well as toxicity resulting rural areas, but nurses are also unequally from the use of multiple drugs. Overuse of distributed in China. Figure 3 compares the drugs can also cause patients long-term number of nurses per one thousand resi- harm in the form of drug resistance. These dents in urban and rural areas. In 2011, the ½ndings are summarized in Table 3. most recent year of data, there were 3.4 How does the quality of health care in times the number of nurses per one thou- rural communities compare to that found sand residents in urban than in rural areas, in cities? One study in the rural counties also a gap that has widened over the past of Ningxia, a low-income province, found two decades. As a result, prevention and that 60 to 70 percent of drug prescrip- health services in rural China have dete- tions for the common cold and upper res- riorated, triggering the central govern- piratory infections were inappropriate.22 ment’s effort to mount a comprehensive Another study in the rural areas of Shan- reform ten years ago. dong province found similar results.23 The quality of health care itself is closely These studies suggest that rural areas suffer related to health care accessibility and from more widespread inappropriate drug expenditure, and is a critical outcome of prescriptions than urban areas. the policy reforms. Numerous studies Another harmful effect of government prior to the 2008 reform documented the policies on health care is illustrated by the overprescription of drugs and overuse of discretional surgery rate. Child delivery intravenous (iv) therapy, both effects of by Caesarean section (cs) is a pro½table the introduction of distorted prices and procedure in many countries, and thus the commercialization of public hospitals can be used as a measure of overuse and and clinics in the mid-1980s. For example, quality of health care. China experienced a one study estimated that approximately rapid increase in cs when insurance was half of all antibiotic prescriptions in China expanded to cover this procedure at a price

62 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Figure 2 William C. Number of Registered Physicians per 1,000 Chinese Citizens, 1990–2011 Hsiao

Source: Ministry of Health of People’s Republic of China, China’s Health Statistics Yearbook 2012 (Beijing: Peking Union Medical College Press, 2012).

Figure 3 Number of Nurses per 1,000 Chinese Citizens, 1990–2011

Source: Ministry of Health of People’s Republic of China, China’s Health Statistics Yearbook 2012 (Beijing: Peking Union Medical College Press, 2012).

143 (2) Spring 2014 63

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting Table 3 Past Health Comparing Chinese Practice with who Standards Policy Mistakes

Indicators China’s Health Centers who Standard

Drugs prescribed per encounter 2.55–2.6 1.6–1.8 Percent of encounters with an 40.1–45.1 20.0–26.8 antibiotic prescribed Percent of encounters with an 32.4–35.4 13.4–24.1 iv injection prescribed

Source: Yongbin Li, Jing Xu, Fang Wang, Bin Wang, et al., “Overprescribing in China, Driven by Financial Incen- tives, Results in Very High Use of Antibiotics, Injections, and Corticosteroids,” Health Affairs 31 (5) (2012): 1075–1082.

above the surgery’s actual cost. Studies percent by 1.35 percent annually. The rate found that China has a national average of increase has accelerated since 2007. of 46 percent of births by cs, while the During the period of 1990 to 2007, health who standard is only 10 to 15 percent. In expenditure was growing faster annually some areas of China, this ½gure exceeded than gdp by 0.6 percent, but this gap 80 percent, and a few even reached 90 jumped to 4.6 percent for the years be- percent.24 tween 2007 and 2011. This means that an The 2008 Chinese health system reform ever-larger share of government and does not seem to have improved the qual- household expenditures has been going ity of health care because it did not alter to health costs. the fee-for-service payment system, nor the These rising inflation rates are chiefly distorted pricing. A large study of eighty- caused by the previously discussed incen- three counties and cities nationwide found tives given to commercialized public hos- only variable and small changes in the use pitals and clinics: fee-for-service payments of antibiotics, iv injections, and steroid and distorted prices. Facilities can increase hormones between 2007 and 2010.25 their revenues and pro½ts by inducing de- More broadly, a health care system is mand for services, in part by advising pa- not sustainable if increasingly larger shares tients that they need unnecessary or exces- of government and household budgets sive hospitalizations, surgeries, tests, and are being diverted to health care without drugs. As pro½t motives took hold and commensurate bene½ts to patients. The gradually overwhelmed professional med- annual health expenditure per person has ical ethics, physicians and hospitals started been increasing at average rates of 10.85 exploiting patients as cash cows. Kickbacks percent (after adjusting for inflation) from from pharmaceutical and medical device 1990 through 2011: a ½gure that has out- companies created perverse incentives, paced the gdp average growth rate of 9.5 further compounding the problem. When

64 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 China achieved nearly universal health making. Unless China can reestablish the William C. insurance coverage in 2008, hospitals and ethical practice of medicine that charac- Hsiao practitioners faced a moral dilemma of terized a previous generation of medical whether to prescribe unnecessary but practitioners, the quality of health care pro½table drugs, tests, and services, given will remain seriously de½cient and health that insurance would pay a signi½cant expenditure inflation will not be curbed. share of the bill. Based on the available But how can China reestablish ethical be- evidence, we can conclude that China havior for millions of medical practition- currently has an inef½cient health care sys- ers, and how long would that take? tem, characterized by tremendous waste Today, most senior physicians in China and generally poor quality. receive handsome incomes from bonuses, While China has experienced remark- red envelopes (bribes), and kickbacks, able growth in per capita gdp, growth in and they expect that their future income health expenditure per person has out- be at the same level or higher. If the gov- paced it: national health expenditure as a ernment formalized this high level of com- percent of gdp rose from 4 percent in 1990 pensation in the hope of disincentivizing to 5.2 percent in 2011. While this is a rela- corrupt practices, junior doctors, nurses, tively high percentage compared to India’s and other professional staff would demand 3.9 percent, it is a relatively low percentage comparable compensation. Government- compared to upper-middle income nations subsidized social insurance plans and such as Mexico (6.2 percent in 2011) and individual households could not afford to Brazil (8.9 percent). Chinese spending will ½nance this increase in compensation. continue to grow at a fast pace as its pop- Moreover, higher salaries alone may not ulation ages.26 end the unprofessional practices of senior physicians or village doctors. With its 2008 health system reform, Establishing ethical guidelines that may China af½rmed that the government as - go against the self-interest of medical prac- sumes primary responsibility for ½nanc- titioners is particularly dif½cult in present- ing health care. This commitment, along day China, where the pursuit of personal with other new declarations, promises to material wealth is viewed as a widely ac- reduce the serious disparities between cepted, if not glorious, social value. China’s health care in urban and rural communi- unfettered free market, with its crony ties. However, it may take more than a capitalism and widespread corruption, is decade to effectively address the maldis- simply not conducive to the ethical prac- tribution of health professionals between tice of medicine, which may call upon the two demographics. physicians to sacri½ce personal gains for The challenge that China faces now is patients’ welfare. how to address the two fundamental China is also confronted by the challenge drivers of poor quality health care and of governing the public hospitals that rapid health expenditure inflation rates. deliver 90 percent of hospital inpatient The causes are interrelated and reinforc- and outpatient services and absorb 60 per- ing. The ½rst is the disappearance of pro- cent of national health expenditures.27 fessional ethics in medical practices. Under the current system, Chinese public Practicing medicine under a pro½t-driven, hospitals do not transform money into commercialized medical system for nearly ef½cacious services; and more investment three decades has eroded professional in health may not improve health out- ethical constraint in clinical decision- comes. Moreover, because oversight of

143 (2) Spring 2014 65

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting China’s public hospitals is divided between of the two growing states can shed some Past Health multiple governmental agencies, the actual light on Chinese accomplishments and Policy Mistakes governance of the hospitals is character- pains in health care. About a decade ago, ized by contradictory policies of various both governments turned their attention ministries. In total, at least eight ministries to health care with a focus on the poor, control some important part of a public spending an additional 1 percent of their hospital’s operations. Each ministry has country’s gdp on health services. A decade its own policies, which are not necessarily later, 95 percent of children in China and consistent or coordinated with those of 43.5 percent of children in India are fully other ministries. As a result, one agency vaccinated; infant mortality rate is 17 per may demand that physicians prioritize 1,000 live births in China, compared with healing the patient, while another agency 50 per 1,000 births in India; 98.7 percent may provide a perverse incentive to do ex - of births in China and 40.8 percent in India actly the opposite. The Ministry of Health are delivered in institutions; life expectancy is responsible for the population’s health, at birth is 73.5 years in China and 65.5 years but does not control the essential inputs: in India.28 Both nations have been trying ½nancial resources, human resources, or to provide universal health insurance with capital investments. Four different min- shallow bene½ts since 2003. Currently, istries control ½nancial inputs: the Nation- about 95 percent of Chinese and 25 per- al Development and Reform Commission cent of Indian citizens have health insur- sets prices for health services and drugs ance coverage.29 Only one-half of the In - and controls capital investments, the dian poor eligible for free hospital insur- Commission on Personnel sets the num- ance are actually covered today.30 ber of staff that hospitals can employ, the The difference in outcomes can partly ccp Organization Department appoints be explained by the attention China gave the director and other leaders of the hospi- to public health since the mid-1960s, as tal, and the Ministry of Personnel sets well as the different strategies adopted to work and compensation regulations and improve health. The Chinese Communist controls staff appointments. Under these government began prioritizing the provi- circumstances, public hospitals are un - sion of basic health care to peasants in the certain of their functions and social re - mid-1960s, long before the Indian govern- sponsibilities, and they are held account- ment did. China also emphasized preven- able not for quality and appropriateness tion and primary care delivered by mod- of services provided to patients, but only estly trained health workers. The Indian for pro½ts or losses. strategy (except in a few states such as Ker- Reforming the governance structure of ala) relied on public health services staffed public hospitals will require a major re - by fully trained physicians and nurses, but structuring of the power, function, author- due to inadequate public ½nancing and ity, and accountability of many ministries. poor governance, these public facilities Such reform demands agreement among could not meet the population’s basic numerous bureaucratic stakeholders and health needs. Lastly, China has an autocrat- powerful political leaders–a major chal- ic but effective government that can imple- lenge under the best of circumstances. ment its policies ef½caciously, while India’s democratic government struggles to reach Both China and India have achieved consensus and implement its policies. enviable economic growth during the past Indian economist Amartya Sen has ar- decade, and a simple comparative analysis gued that economic growth and increased

66 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 gdp per capita may not necessarily re- and access to quality health services. The William C. duce social inequalities.31 This is clearly policies that China implements to nar- Hsiao evident in both China and India. Between row these disparities will tell us about the their urban and rural residents, both social values that undergird the govern- nations are confronted by huge disparities ment’s actions, as well as the Chinese in health status, health spending per capita, political economy on social policy. supply of facilities, quali½ed medical staff,

endnotes Author’s Note: I would like to thank Mingqiang Li for his superb contribution to this paper. He diligently and thoroughly gathered the data and assisted in the analysis. 1 See Barry Naughton, Growing Out of the Plan: Chinese Economic Reform, 1978–1993 (Cambridge: Cambridge University Press, 1996). 2 The World Bank, “Results Pro½le: China Poverty Reduction,” March 19, 2010, http://www .worldbank.org/en/news/feature/2010/03/19/results-pro½le-china-poverty-reduction (accessed August 8, 2013). 3 See Fu Jun, “World Economic Downturn and Challenges for China,” paper presented at Money and Banking Conference: Lessons and Challenges for Emerging Countries during the Crisis, Buenos Aires, Argentina, August 31–September 1, 2009. 4 See “Chinese Migrant Workers Exceed 260 Million by 2012,” Xinhua News Agency, May 23, 2013, http://news.xinhuanet.com/english/china/2013-05/27/c_132411765.htm (accessed August 1, 2013). 5 National Bureau of Statistics of People’s Republic of China, China Statistical Yearbook 2012 (Beijing: China Statistics Press, 2012); see also previous years’ statistical yearbooks. 6 See Yana Ling, “Gini Coef½cient Release Highlights China’s Resolve to Bridge Wealth Gap,” Xinhua News Agency, January 21, 2013, http://news.xinhuanet.com/english/china/2013 -01/21/c_132116852.htm (accessed August 1, 2013). 7 Martin King Whyte, ed., One Country, Two Societies: Rural-Urban Inequality in Contemporary China (Cambridge, Mass.: Harvard University Press, 2010). 8 See Joseph Kahn, “China’s ‘Haves’ Stir the ‘Have Nots’ to Violence,” The New York Times, December 31, 2004. 9 See Xue-Qiang Wang, Xiao-Tong Wang, and Jie-Jiao Zheng, “How to End Violence Against Doctors in China,” The Lancet 380 (9842) (2012): 647–648. 10 See Therese Hesketh and Xing Zhu Wei, “Health in China: From Mao to Market Reform,” BMJ (British Medical Journal) 314 (7093) (1997): 1543. 11 See Winnie Yip and William C. Hsiao, “The Chinese Health System at a Crossroads,” Health Affairs 27 (2) (2008): 460–468. 12 David Blumenthal and William Hsiao, “Privatization and Its Discontents–the Evolving Chinese Health Care System,” New England Journal of Medicine 353 (11) (2005): 1165–1170. 13 Karen Eggleston, Li Ling, Meng Qingyue, Magnus Lindelow, and Adam Wagstaff, “Health Service Delivery in China: A Literature Review,” Health Economics 17 (2) (2008): 149–165. 14 Data source: Ministry of Health of People’s Republic of China, China’s Health Statistics Yearbook 2012 (Beijing: Peking Union Medical College Press, 2012); see also previous years’ health statistics yearbooks.

143 (2) Spring 2014 67

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021 Correcting 15 See Winnie Yip and William C. Hsiao, “Non-Evidence-Based Policy: How Effective is China’s Past Health New Cooperative Medical Scheme in Reducing Medical Impoverishment?” Social Science & Policy Medicine 68 (2) (2009): 201–209. Mistakes 16 Data source: National Bureau of Statistics of People’s Republic of China, China Statistical Yearbook 2012; and Ministry of Health of People’s Republic of China, China’s Health Statistics Yearbook 2012. See also previous years’ yearbooks. 17 See William C. Hsiao, “The Political Economy of Chinese Health Reform,” Health Economics, Policy and Law 2 (3) (2007): 241. 18 See Wanchuan Lin, Gordon G. Liu, and Gang Chen, “The Urban Resident Basic Medical Insurance: A Landmark Reform Towards Universal Coverage in China,” Health Economics 18 (S2) (2009): S83–S96. 19 See Winnie Yip, William C. Hsiao, Wen Chen, Shanlian Hu, Jin Ma, and Alan Maynard, “Early Appraisal of China’s Huge and Complex Health-Care Reforms,” The Lancet 379 (9818) (2012): 833–842. 20 Wen Chen, “The Analysis of the Characteristics and Defects of China’s Pharmaceutical Price System,” Shanghai Pharmaceuticals 26 (22) (2005): 533–535. 21 Yongbin Li, Jing Xu, Fang Wang, Bin Wang, Liqun Liu, Wanli Hou, Hong Fan, et al., “Over- prescribing in China, Driven by Financial Incentives, Results in Very High Use of Antibiotics, Injections, and Corticosteroids,” Health Affairs 31 (5) (2012): 1075–1082. 22 Winnie Yip, Timothy Powell-Jackson, Wen Chen, Min Hu, Eduardo Fe, Mu Hu, Weiyan Jian, et al., “A Cluster-Randomised Evaluation of the Effect of Capitation with Pay-for-Performance on Primary Care Providers’ Antibiotic Prescribing Practices in Rural China,” working paper (2013). 23 Adam Wagstaff, Winnie Yip, and Meng Qingyue, “Impact Evaluation on Paying for Per- formance in China’s Health Sector (Evaluation of Health XI),” report on the sief Project (2011). 24 Qian Long, Reija Klemetti, Yang Wang, Fangbiao Tao, Hong Yan, and Elina Hemminki, “High Caesarean Section Rate in Rural China: Is It Related to Health Insurance (New Co-operative Medical Scheme)?” Social Science & Medicine 75 (4) (2012): 733–737. 25 Ministry of Health of People’s Republic of China, “Report on the Implementation of National Essential Medicines Policies” (Beijing: Ministry of Health, 2011). 26 World Health Organization, The Global Health Expenditure Database, http://apps.who.int/ nha/database/PreDataExplorer.aspx?d=1 (accessed August 26, 2013). 27 Yip et al., “Early Appraisal of China’s Huge and Complex Health-Care Reforms.” 28 The data on full vaccination and institutional delivery are from International Institute for Population Sciences, National Family Health Survey-3 (2005–2006); and Ministry of Health of People’s Republic of China, China’s Health Statistics Yearbook 2011 (Beijing: Peking Union Medical College Press, 2011). Infant mortality and life expectancy are from the World Bank, World Development Indicators, http://data.worldbank.org/data-catalog/world-development -indicators. 29 See Gerard La Forgia and Somil Nagpal, “Government-Sponsored Health Insurance in India: Are You Covered?” (Washington, D.C.: World Bank Publications, 2012). 30 Ministry of Labour and Employment, Government of India, “Rashtriya Swasthya Bima Yojana: National Summary,” http://www.rsby.gov.in/overview.aspx (accessed August 26, 2013). 31 See Amartya Sen, “Quality of Life: India vs. China,” The New York Review of Books, May 12, 2011, http://www.nybooks.com/articles/archives/2011/may/12/quality-life-india-vs-china/.

68 Dædalus, the Journal ofthe American Academy of Arts & Sciences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/DAED_a_00272 by guest on 23 September 2021