China's health care system : policies, organization, inputs and finance

World Bank Reprint Series: Number 392 Dean T. Jamison tmas eanwreystcnm Policies, Organ1ization, Inputs, Reprinted with permission from Good Health at Low Cost, edited by Scott B. Halstead, Julia A. Walsh, and Kenneth Warren, published by the Rockefeller Foundation, New York, October 1985.  China's Health Care System: Policies, Organization, Inputs and Finance* by Dean T. Jamison, Chief, EdLucation Policy Division, The World Bank, Washing,ion, D. C 21  China's success in improving the health of its people far mariies salient rural-urban differences to show that progress exceeds what could be expected at its stage of develop- in improving health conditions has been far from uniform. ment as illustrated in Figure 1. Comparative health data be- tween China and othercountries (SeeTable5) furtherconfirm that China's success in increasing life expectancy (and reduc- 1. Health Policies ing fertility) goes far beyond what would be expected given Health policy is formed along a number of dimensions. One the levels of other indicators of development' Much of concerns the extent to which resources will be made avail- China's success has been attributed to the national health able and how theywill be financed. A second is the relative service delivery system, and both it and the policies on emphasis on preventive and curative measures. A third is the which it is based are described in this paper. Other influen- extent to which responsibility for health rests with highly ces on health-watersupply and sanitation, food availability, trained professionals, with lower-level staff and with the education and fertility levels - are discussed in another paper individual. A fourth concerns the way health programs are for this conference. Most of the discussion in this paper addressed - directly, through the health care system, or abstracts from the major inter-provincial and rural-urban dif- indirectly by improving water supply and sanitation, educa- ferences that remain very important in China. Table 1 sum- tion or nutrition levels. A fifth important dimension is the f;,caore 1 Life Expectancy in Relation to Income: Developing Countries, 1978 Life Expectancye 80 Cuba Flong Kong Sri Lanka * Costa Rica * Argentina Malaysia r hl ,Singapore China Mexico * Thailand ~ rzlNormr for 36 Developing Countrifes Philippines Egvpt o Algeria P'akistan Noii 5. India Indonesia o Nigeria *Congo *Ivory Coast Bangladesh N'sepal a 40 Senegal Ethiopia 0 50 10 0 $1,50s 0 $2,000 $2,500) $3,000 $3,500 - GNP' Per Petrson, 1978- * . ,iws tan,r, at tt h Ith he t1 itse 1\,rilg(i r I I onbr .il a dr- Al IwI. "A * l hI , I ( pe(4 tt-ed I .i - I1 (U rrent *ondm n, vIrc ,to ( ontitiut throtuLighl t his LI tII or her i(itile - e l id;tii l ;rod n .n I: r N I;.i '1IAt.ir-:, :ir *- 11 i 1 rliI,( I ,wI-I r sw. A if aIr) Si 'II" 'f l- o ,itr I 404 \ i ., i urlr l)ui, | 1t98(f ifgiLre 1- 38 rt rv rO( i(t e(t tr)nllI . it it. *1 H1'-ft i4 gtnn Ii ( Populo atiorn Reret I Bureau D I b)ereitier 1'180 `we; \trW'{o iis il , ,rJ kd!iC tor Lnt explanation t h ith P0 ri( 0(r01- un't tlI ) dernv th lint-e shovni China alnd t bi t i ot- IN ki d i nw h ttle lstrnitm I i th t hut i il . . - i- . j,dt di i iti ther it(' tvs7p ( It7ut \ 1no C N I' Vper pe rson Thi-, paper Ie drawt lirint ipati tron lp hpter i t0 I 1, 0i - lO st r i'- Ilt 1 More detailed statistics on health conditions in China and their improvement over l'arneu I f 5 n.I Kali~ I irre,r N ton tt inld A Prit WiN tm . time mabefoundin XuSu-enspaperforthis conference, HealthStanisticsofthe I ' h 7fkrlo Bank I tl84 \ wvv, vxprt-t,(ir in1 the i- I jrt; tho,ve ot the iUthLr People's Republi( of China. .inli i tiit n1t eshnlt f-a ilet't i 100 it The Wo orld Bank. 2. See Chen C'hunming 'The Development of Health Systems in 'China and the Con- ducive Factors.' 22  'rable 1 China: Urban-Rural Differences Related to Health Urban Rural National Health Status Crude Death Rate, 1 981 (per thousand) 4.8 7.1 6.6 Life Expectancy, 1975 (years) 72 57 59 Nutrition Status Percent of 7-Year Old Boys Nutritionally Stunted, 1979 2.6% 12.7% NA Fertility Levels Crude Birth Rate, 1981 (per thousand) 13.5 24.8 20.9 Total Fertility Rate, 1981 1.5 2.9 2.6 Health Service Availability (per 1,000 population) Doctors of Traditional Medicine 0.38 0.27 0.29 Senior Medical Staff 2.05 0.28 0.67 Middle-Level Medical Staff 2.99 0.73 1.16 Annual Health Expenditure (dollars per capita) State Subsidies $13.1 $1.4 $3.8 Private Expenditure 1.4 2.5 2.3 Total Expenditure 16.2 4.8 7.2 Annual Food Subsidies (dollars per capita) $48 Negligible NA NA: Not available extent to which a broad range of health services will be in the uneven quality of rural medical services. Even in rural integrated into a common delivery package for a specific areas, however, the resources available for health care are geographic area, orwhetherseparate"vertical" (orcategori- substantial compared to rural areas in many other low-in- cal) programs will address specific diseases. come countries. The Chinese have taken relatively consistent po- sitions on these policy issues, though emphasis has changed Preventive Emphasis: from time to time. This section discusses general directions Compared to virtually all other countries, China has strongly of health policyin China in terms of howthe governmenthas emphasized public preventive over curative health serv- responded to these policy issues. ices.1 Majorcampaigns were mounted shortlyafterl 949 to improve environmental sanitation; to eliminate the (then) R c Mn ."four pests" - rats, flies, mosquitoes, and bedbugs; to vac- ResorceMobrzahrl:cinate against and cure infectious diseases; and to control China's leadership places great value on improving health c a a c the vectors of major endemic disorders such as malaria and and has made a substantial volume of resources available to s o . M schistosomiasis. Mass mobilization played a key role in the do so - accounting, in 1981, for about 3.3 percent of GDP. ' success of these campaigns. The allocation of health resources Financing is in part by the State (for government and enter- t p a ( to preventive activities (and to curative services In urban prise employees), in part by communal or subcommunal levels of organization and in part by the individual. In rural fr east minimalyurae srovices and rmaceual area an imotn mehnslooiiigrsucs(n for at least minimal curative services and pharmaceutical ara nipratmcaimfrmblzn eoucs(n availability In the rural areas, and health policies increasingly pooling risks) has been the cooperative health insurance sys- responded to thea cnern. hilethe vigoos preventiv tem; this system, however, is being eroded by the rural pro - res uoubtcedl erusessfuu inefecting duction responsibility system. While great variation in finan- initi morb ando motaliy e reduccts,fth dn efo cing arrangements exists throughout the country, in general, curtive care cud be aignoreduonlynup to apnand is the State and communal levels combined with self-help . . . , i finance the greater part of the preventive measures, leading receiving increased emphasis. The preventive approach to to relative uniformity of their distribution around the coun- 1. This orientation is deeply imbedded in Chinese tradition. A recentWestern review try. Rural curative services, on the other hand, tend to be of medical policy in China begins by noting that "2The keystone of the Chinese financed at the individual and brigade level, and the major medical tradition is the notion that medicine is the art of maintaining health, not curing disease." (See D.M. Lampton, The Politics of Medicine in China, Westview nationwide income inequalities at these levels are reflected Press, 1977.) 23  Figure 2 Organization of Health and Birth Planning Services . Clinic-Based Medical Services Birth-Planning Programs Planning Vertical - Preventive Programs Coty Urban Responsible for: Budgeting County District o Planning and monitoring Unified administration with General General ' of birth planning staff of the provincial, Hospitals Hospital 0 Information education protectural and county ; I and communication Specialized levels (anti-epidemic * v r v Liaison with clinical Hospitals stations, and maternal and Commune-Clinic Street and services child health centers) (middle-level Workplace Responsible for: physicians and Health * Infectious disease auxiliaries) Three- Stations - surveillance, prevention 1 Tiered t A 1 School and cure (including Production System Lone vaccinations) Brigade- Health * Hygiene and public Brive Staith health campaigns Cooperative Stations * Monitoring food and (3 or 4 barefoot L Research environmental sanitation doctors) * School hygiene I + O Industrial hygiene Production Team * Maternal and child (part-time t health health aide) Referral Supervision I - , _ , _ , _, and Training Sot,rcc. World Bank problems of communicable disease does not, moreover, Multifaceted Strategy: seem to be playingnearlyas prominent arolein dealingwith China's concern with prevention has also led to a health today's problems of chronic disease, strategy that reaches well beyond the health system per se. In particular, improved nutrition,' provision of safe water Manpower policy: supplies, sanitary and convenient means of waste disposal, China's policy toward highly professionaiized versus lower- and fertility reduction and widespread educational im- level staff follows naturally from its emphasis on public pre- provements have been major policy objectives. Underlying ventive services and on providing at least minimal levels of all facets of this health strategy have been the administra- curative care, Training requirements for thoracic surgery dif- tive capacity and political will of the Chinese government, fer substantiallyfrom requirements forgivingvaccinations or which have, despite occasional setbacks, been essential assisting with environmental sanitation, to take extreme ex- to success. amples, Preventive measures also place greater respon- sibility on communities and individuals for their own health conditions. As the basics of hygiene and commrunity involve- Program integration: ment in preventive health become more established, how- A final policy issue is that of integrated versus vertical prou- ever, it will become necessary to upgrade the skills of the grams. Figure 2 provides a schematic overview of the struc- barefoot doctor to undertake (and be trusted with) more ture of health care deliver in China, although it should be complicated tasks. The Chinese Ministry of Public Health emphasized that there is substantial local variation in the (MOPH) plans, within a decade, to have upgraded 50 to 60 structure and organization of the service. The three main percent of barefoot doctors and chronic care workers to the boxes shown in Figure 2 are the "Vertical Preventive Pr- level of'countryside' doctor. It should also be stressed that grams," "Clinic Based Services," and "Birth Planning Serv- the intensive and active involvement of the (non-profes- 1. A detailed discussion of food policy and nutritional change in China maybe found sional) political leadership at every level is a key element in in "Food Policy and Nutritional Status: China, 1949-82". by D.T. Jamison and A implementing China's health policies. The Army has also Piazza. ,P. Gittinger, J. Leslie and C Hoisington iEds.), in Food Policy. Integrating Supply, Distribution and Consumption, John Hopkins University Press, forth- played a role in providing manpower for rural medical care. coming. 24  ices." Much of preventive medicine in China is public and typically has some assistant doctors and sometimes full doc- organized vertically, with responsibility for controlling spe- tors in addition to paramedical staff. All brigades, except cific communicable diseases centrally located. In addition, those very close to an urban or commune center, have a hygiene promotion (through the National Patriotic Health brigade health post. In Sichuan in 1981, for example, there Campaign Committ" es) operates in coordination with politi- were 65,446 brigade health posts for75,570 brigades, an 87 cal authorities in a centrally directed fashion. The impor- percent coverage. Even brigades without health posts gen- tance of these centrally run activities can, however, be erally have a barefoot doctor. overemphasized since the preventive, vertical programs At the county level, the top of the three-tiered heavily utilize the clinic-based services for their implementa- system has three health care units - the county general hos- tion and would certainly be far less effective without them. pital, the county epidemic prevention station and the county Nonetheless, the strong emphasis placed on categorical MCH station. At this level and above, facilities are owned by programs and health campaigns (except during the Cultural the state and funded from its budget Underthe direction of Revolution) likely played an important role in China's suc- the county health bureau, each of the three units supervises cess in dramatically reducing morbidity and mortality rates. and supplies technical support (including in-service training) Other countries seeking to learn from China as they extend to the commune health centers within the county in its primary health care should thus pay particular attention to specialized field. Typically well-staffed by college-graduate China's centrally initiated and directed vertical programs. An physicians, the county hospital serves as the general referral importantfeatureofthe organization of thevertical program hospital for the entire county, receiving and treating cases is that instead of having separately administered programs referred to it by commune, urban and enterprise health cen- for malaria, smallpox, leprosy, etc., most vertical programs ters. While data are unavailable for more than a few coun- are the responsibility of the epidemic prevention stations of ties, it appears that 50 to 75 percent or more of the patients the provincial and county health bureaus. (Tuberculosis is served by county hospitals are urban residents from the perhaps the most important exception to this generaliza- county town. The county hospital also provides training for tion and is underaseparatestructure.) Interestingly, the clini- the medical staff atthe commune level. The countyepidemic cal services associated with family planning were initially prevention station, with three to ten staff members, is re- provided through the MOPH; in the past fewyears, however, sponsible for preventive health and communicable disease the 5tate Family Planning Commission and its implementing control for the entire county. It provides technical support units have increasingly assumed this responsibility. for and supervises the preventive health section of the com- mune health centers. 2. Organization of Health Service Delivery The Three- Tiered System: Urban Health Services: The primary health care system that exists today in rural Urban health services, like those in rural areas, operate on China operates on three levels: the brigade cooperative three levels, although the system of finance differs. Corres- medical center, the commune health center and the county ponding to the brigade and commune health facilities are general hospital (Figure 2). This system is based upon the street' and 'lane' health stations, staffed by paramedicals, idea that, in a country where it is impossible to have a fully that provide preventive and simple curative services. Street qualified doctor in each town or village, it is feasible to have clinics play a role in urban areas similar to that of commune one ormore paramedical workers within walking distance of health centers in rural areas. They have few beds, their staff most communities, even many that are poor and remote. generally includes doctors as well as other health workers The two lower levels of the three-tiered system - the com- and they play an important role in public health work Refer- mune and the brigade - are financed principally by the rals from these stations go to district,' municipal, provincial peasants through various combinations of fee-for-service, or specialized hospitals. brigade-level cooperative health insurance and work-point Urban collective organizations also exist and subsidy for barefoot doctors.1 About one-third of the com- include some hospitals, especially hospitals of traditional mune health centers are state-financed. These lower levels C:hinese medicine, and street clinics. Such organizations are constitute the rural collective health system and provide officially distinct from lane health stations or hospitals run most of the medical care received by the Chinese people. In directly by municipal and district health departments, but in addition, production teams often have part-time midwives practice the distinction is unimportant. They are operated in and health aides who link the production teams to the much the same ways as government units. They pay similar barefoot doctors in the brigade health post. Almost all com- salaries and, in contrast to enterprise facilities or rural collec- munes have one commune hospital or health center, with tive facilities, which can obtain resources from enterprises ten to forty beds, which is responsible for routine curative and communes, brigades and teams depend entirely for work, preventive medicine and family planning. The center funds on the state or on operating income. 1. Astheruralresponsibilitysystembecomebmorewidespread,thework-pointsystem Distncts' are the urban political unitsatthe same administrative level as counties. is losing prominence. Fee-for-service and, in' some cases, government subsida of barefoot doctors' saidnes, replace the former system. This change has the dual dis- advantages ofleaving many individuals uninsured and of eroding thefinancial base for preventive and promotive activities. 25  State Enterprise Organization: improving public health as increases in the quantity and Side by side with the network of national, provincial and quality of water and sanitation facilities. local facilities, which serves the whole population, is a net- The key to the success of the health campaigns work of industrial and other state -nterprise hospitals and has been a very effective organizational network at all levels, facilities that provide free services to workers in those enter- closely linked to the network of political, administrative and prises. Relatively little information is available on these facili- economic organizations. At each level - from the national ties, but those data that are available indicate that in 1981 through the provincial and prefectural to county, commune they provided about 25 percent of medical care in China1 and brigade - is a small health campaign full-time staff that Although enterprise facilities report administratively to the reports directly to the health campaign committees, which health sections of various industrial departments, they also also exist at all levels and include high-ranking party and receive technical guidance from provincial and local health government officials and representatives of mass organiza- departments. Depending on the size of the enterprise, corn- tions such as the youth league and the women's federation. plicated cases are treated there or referred to the govern- (Premier Zhou En-lai was for many years chairman of the ment hospital system; if a patient is referred to a district National Patriotic Health Campaign Committee.) The heaith hospital, expenses are reimbursed entirely by labor insurance. campaign committee at the brigade level, which is often Large enterprises handle almost all their own cases, and the chaired by the brigade barefoot doctor and reports to the Railway Ministry and Army, to take extreme examples, even brigade management committee, mobilizes people to carry have their own medical colleges as well as hospitals. out both the programs initiated from above and any pro- grams devised by the brigade health committee itself. Although much has been achieved through the Patriotic Health Campaigns: health campaigns, the need for them still remains. This is As indicated previously, the Chinese Government has made partly because, unless campaigns are repeated, problems considerable use of health campaigns as a means of attain- may recur - for example, areas previously cleared may be- ing some of its health objectives. The campaigns take the come reinfested with snails. New campaigns also may be form of the mass mobilization of people for preventive and necessary to tackle the diseases that are becoming relatively health promotion activities. The first campaign was launched more prevalent. This is already happening, and the current in 1951, and over the past 30 years, there has been an campaignobjectivesincludefoodhygieneandoccupational average of four or five campaigns a year. These campaigns and environmental pollution. Somewhat surprisingly, how- have been conducted under the leadership of an organ- ever, there are as yet no campaigns to promote the diet ization known as the "National Patriotic Health Campaign exercise and anti-smoking efforts2 that would help prevent Committee." the newlyeemerging chronic diseases. The planning of future Although it is difficult to separate the effects of campaigns vw ili also have to take account of recent changes the health campaigns from those of other programs, there is in the organization of economic activities in rural areas which, little doubt that without these campaigns, China's achieve- by giving households much more autonomy, make it more ments in health over the past three decades would have difficult for the brigade committees to mobilize labor for been much less substantial. The reduction in the prevalence health campaign work of schistosomiasis, for example, owes much to the mass mobilization of people forwork, which in some areas entailed not only diverting water channels from infested rivers and Traditional Medicine: canals and burying the snails, but also collecting and killing Traditional medicine, as is well known, plays an important individual snails. Similarly, the reduction in the incidence of role in China. Several important types of traditional medicine malaria has been due largely to campaigns, a key aspect of are practiced - Including Chinese, Mongol, Tibetan, Ugyour which has been to fill in stagnant ponds and eliminate other and others. Because over93 percent of China's population mosquito breeding sites. People are mobilized to clear the is of the Han ethnic group, Chinese traditional medicine pre- streets and remove rubbish and nightsoil, and the disposal dominates, although official policydoes not discriminate in of the waste is carefully supervised. Another objective of the favor of any particular system of traditional medicine. campaigns has been to teach people to boil drinking water All forms of traditional rnedicine share two char- and notwash tools and containers contaminated byexcreta acteristics. The first is that their theoretical and diagnoDstic in rivers that provide drinkingwater. Such campaigns to pro bases are not (at least for the present explicable in terms of mote better hygiene have probably been as effective in modern biology.3 The second is their wide variety of thera.- 1. Dataontheavailabilityofhealthfacilities underotherministres" inselectedyears suience is well put in a do ument prepared for The World Bank by the Beiling have recently been made available: these show that the rate of growth of personnel College of Traditional Chinese Medicine: Traditional Chinese medicine is imbued and facilities in the enterprise system has beensubstantiallymore rapid than forthe with rich lnical experience through ages and it is a unique theoretical system health care system as a whole. difficult to understand because its theory [is] confined by the development of science in ancient China and is combined with the ancient philosophy and explained 2. As this document goes to press. it appears that campaigns against smoking are b' the philosophical terms It has not integrated with modern science and technol- being initiated. ogy and some of the theories still carnot be made clear by modern science. For this reason, it is difficult for the students to understand since there are no oblective 3 The current relation of the theoretical system of Chinese medicine to biological parameters to reveal the essence of Chinese medicine." 26  Table 2 Seleded Health-Care Workers - Education and Role Position Education Typical Role Non-Salarild Workers Barefoot Doctor Highly variable, primary e-Aucation Service in brigade health station (or more) plus 3- to 6-month special course, plus continuing education Rural Doctor Upgraded barefoot doctor Service in brigade health station Middl-Level Salaried Statf Assistant Doctors (of Western and Specialized senior secondary school Play assisting roles in c ity hospitals, roles Chinese medicine) of greater leadership in commune health centers Nurses Specialized senior secondary school Support role in hospitals Senior Salaried Staff Western Doctor Varied over time; currently 5 to 8 County general hospitals; urban and years post-secondary enterprise hospitals Doctor of Chinese Medicine Currently 5 years post-see ondary County and distric t general hospitals; hospitals of Chinese medic ine Public Health Doctor Same as We,stern do(tor, except for Staff and manage county and provincial less clinical training and more anti-epidemic stations public health \o(lt ThI re are niam more t\ro% tI I ior O e( l 0ln I .I ( li (,I t tIe .i(tgorw I ited im I tr 1 do tI(hIiII Ol., t( '14! RT m ei f nam t % t A .i (l \hAu( 'Aork( r, ar, hlo\s ('V(r. o IIn( I the k1wp Itiled hte re peutic measures. These include acupuncture and moxibus- opment. The priority accorded to both traditional medicine tion in Chinese traditional medicine' and a wide variety of and rural health care is reflected in the fact that traditional herbal and other medicinal preparations in each of the practitioners are almost as numerous in proportion to pop- traditional systems. Important efforts are now underway to ulation in rural as in urban areas - there are.26 practitioners improve biochemical knowledge of the active agents in per thousand in rural areas and .46 in urban areas. In con- these preparations, and attempts to initiate scientific efforts trast, senior Western medical staff are about seven times as to assess the therapeutic efficacy of traditional procedures numerous in proportion to population in urban areas as in enjoy strong support, at least from the Ministry of Public rural areas. Health. Ancient Chinese medical writings also contained accurate dietary advice concerning avoidance of important 3. Inputs to Medical Services m icronutrient deficiency diseases (beriberi, xeropthalmia Section 5 of this paper provides estimates of the and goiter), and current dietary and exercise guidelines from fraction of health care expenditures devoted to the inputs to traditional medicine accord well with research findings on medical services personnel, facilities, equipment and phar- prevention of chronic disease. maceuticals; it suffices here to note the predominance of In recent years, the Government has encouraged expenditure on pharmaceuticals-49 percent of total expen- traditional medicine, and the number of hospital beds for ditures are on Western and 9 percent on traditional drugs. traditional medicine increased by35 percentbetween 1979 Each major class of inputs to medical services is discussed and 1981, compared with a 6 percent increase in general below. hospital beds and a 1 percent decrease in commune clinic beds. An increase in the number of doctors of traditional Personnel: medicine is one of the priorities for health manpower devel- There are three levels of salaried staff. Higher level personnel are trained in medical colleges, with specialized under- 1. A particularly clear exposition of what is known about the analgesic and therapeutic gaut rgas oeo oeo h ulcso et effects of acupuncture and moxibustion may be found in Science in Traditional graduate programs in one or more of the subjects of West- China, by J. Needham, Chapter 4 (Harvard University Press, 1981). ern medicine, public health, high-level nursing forensic medi 27  Table 3 Availability of Health Care Personnel - China and Other Countries Number Per 1,000 Population' Developing Developed Personnel Category China (1980) Countries Countries Doctors of western medicine .45 .33 1.9 Medical assistants (in China; assistant doctors of traditional medicine) .45 .11 NA Traditional medical practitioners (in China: including pharma(ists of traditional medicine) .38 .35 NA Nurses and assistant nurses .71 .65 4.6 Dentists .007 .06 .42 'Workl .aerage,, .1roLUnd lt3-5 N.A. Not apaila.hlv cine, pediatrics, stomatology, pharmacy, traditional Chinese of health manpower, and Table 4 shows the growth in num- medicine and traditional Chinese pharmacy, and massage bers of hospital beds of selected types between 1979 and and acupuncture. From 1985, undergraduate programs will 1981. The decline in the number of commune clinic beds is be added in hospital nutrition, health economics and man- at variance with government policy, which is to increase the agement and health statistics. Middle level health personnel proportion of cases that can be handled at the commune are trained in secondary medical schools as assistant doctors level, without referral to county facilities. It is, however, con- of Western or Chinese medicine or public health, nurses, sistent with an analysis suggesting that the smaller a com- assistant stomatologists, assistant pharmacists, laboratory mune health center the less well utilized it is.' A decline in technicians, radiographers, etc. Primary level health workers the number of commune clinics could thus be interpreted as and barefoot doctors are trained through refresher courses a result of decreased demand for their services, reflecting a and in-serviceexperience in countyhospitals, epidemic pre- tendency to bypass them in favor of the county hospital. vention stations and maternal-child health stations. Primary Efforts to establish and strengthen "major commune health health workers and assistant doctors with several years' ex- centers" should help alleviate this problem. perience in practice may take refresher courses and exam- courses and examiniations to qualify as middle level and Pharmaceuticals and Medical Equipment seniordoctors respectively. Table 2 summarizes the differ,.nt Pharmaceuti als and medical equipment in China are prin- categories of health personnel and their typical training. Pro- cipally produ. ed and distributed by three corporations form- motion based on experience and examination is an impor- ing part of the State Pharmaceutical Administration of China: tant alternative to formal training in a med;cal college as a the China National Pharmaceutical Corporation (280,000 source of doctors, particularly in rural areas. employees), the China National Herbal Medicine Corpora- There has been a rapid increase in the numberof tion (80,000 employees) and the China National Medical health care workers of most types, and, quantitively speak- Equipment Production Corporation (80,000 employees). ing, the health manpower situation is rather good. The total value of the annual output of the three producing Table 3 compares the availability of health care corporations is about 3.5 billion dollars peryear. Adding di- workers in China with the WHO estimates of availability in tribution costs, the total comes to slightlyover 5 billion dollars other countries. As that table clearly indicates, the situation ayearforpharrnaceuticals and medical equipment, of which in China is relatively very good for most categories of person- about 90 percent is for pharmaceuticals. As distribution nel, one important exception being dentists. markups (typically 15 percent) cover the operating costs of the National Drug Corporation2, and each of the three pro- Facilities: duction companies returns a small profit to the state, drug The growth in the availability of facilities has paralleled that production and dist'ibution is notsubsidized. China imports -h 4 very little in the way of pharmaceuticals. hinlItHe 4 Each province has its own pharmaceutical bureau, China: Increase in Number of Hospital Beds, 1979-81 which receives technical guidance from the Pharmaceutical Number of Beds Administration but is under the administrative control of the 1 See " Commune Health Care in Rural China,' by J. van der Gaag, Technical Note Type of Hospital 1979 1981 ' Increase GEN 20, Population, Health and Nutntion Department The Word Bank 1984. 2 The National Drug C orporation 1310,000 employees) distnbutes pharmaceutical Gel 907998 968041 6 2 products throughout the country. Distnbution is through 5500 pharmaceutical eneral Hospil 9,9 9, . wholesale departments at county and district level, typically) and over 50.000 Commune Clinic 7,71,231 7 63,11 4 -1 .(1 drugstoresbelowthecountylevel. Hospitalssometimespurchasepharmaceuticals directly from a factory rather than from wholesale distributors. 28  province. The Pharmaceutical Administration also operates (ii) For cancer mnortality, on1 the other hand, income three colleges (the Shenyang College of Pharmacy, the Nan- increases are associated with increases in age-ad- jing College of Pharmacy, and the Shanghai College ot Medi- justed mortality rates for cancer. To the extent that cal Equipment; in addition, there are 13 technical middle this adverse relation between income and mor- schools of pharmacy operated by provinces. tality (age-adjusted) holds for cardiovascular and The MOPH is responsible for drug quality con- other chronic diseases as well as for cancer (and it trolandforlicensingnewdrugs.TheMOPH produces serums is found to hold for hypertension morbidity), the and vaccines at six regional institutes of biological products. estimated postive effects of income on life expec- Serums and medicines used in curative medicine are dis- tancy understate the beneficial effects of income tributed through the National Drug Corporation; those used improvements on communicable disease control. in anti-epidemicworkare distributed bythe MOPH. In 1982 (iii) There appearto be no significant improvements in the six institutes produced 140 different biological products, life expectancy that are associated with higher per including 500 to 600 million doses of vaccines; the total capita availability of barefoot doctors, hospital beds, value of output was about $25 million. or other health resources in a province. This find- Drug production is growing at a present rate of 6 ing must be viewed as very tentative, although it is to 7 percent per annum. Current production is about 14. consistent with some (but not all) of the literature times greater than 1957, and prices have remained stable. from other countries. One explanation for the Domestic sales increased seven-fold between 1955 and findingwould be that preventive activities account 1981, and exports increased twenty-fold. The Cultural Revo- for much of the mortality improvements, and that lution slowed the rate of growth of production and weakened these activities are more uniformly distributed than quality control. In recent years there has been a very rapid curativefacilities. Barefootdoctorsaswellassalaried increase in production of drugs for the treatment of cancer health workers and hospital facilities may, none- and cardiovasculardisease. Further, new antibiotics are being theless, have important roles in providing access introduced in the place of old. to care and in reducing morbidity and its adverse consequences. The geographical distribution of 4. The Impact of Health Resources future increases in the quantity of health resources The preceding sections have discussed the policies guiding will, of course, determine the extent to which they health resource allocation in China and the growth in avail- contribute to further improvements in access to ability of hospital facilities and health personnel. The generally care. accepted reason for China's successes in the health sector The analysis suggests that only limited further has been the manner in which resources were deployed- at gains in life expectancy are to be expected frorn quantitative least as much oreven more than the extent to which doctors increases in health facilities and manpower of the quality and hospitals were available. Specifically, China's emphasis available at the time data were gathered for this analysis; on prevention has been viewed as particularly important as further, as the disease profile moves more toward chronic has been geographical deployment that reaches rural areas and away from communicable disease, income improve- relatively well. Analysis of the correlates of various measures ments are likely to cease having the beneficial effects on life of morbidity and mortality tend to support this view.- expectancy that they have had in the past This suggests the Data on life expectancy, crude death rate and priority forseeking affordable public policies for postponing age-adjusted cancer mortality rate by province were used to the onset and managing the consequences of chronic dis- examine factors influencing mortality. The following con- ease in the population, with concomitant qualitative change clusions emerge from the analysis: in the education and function of health personnel. (i) A strong positive relation exists between provin- cial income levels and life expectancies estimated 5. Health Sector Financing and Expenditures2 for 1973-75; a 10 percent increase in income is Finance and Expenditures: estimated to correspond to an increase of about 8 This section analyses the mobilization arid allocation of re- months in life expectancy. Consistent effects are sources in the Chinese health sector. Total health expendi- found in the analysis of 1981 crude death rates. ture in 1981 is estimated at approximately $7.50 per capita. There is also evidence that, controlling for income, This represents3.3 percentof GDPpercapita in 1981. Figure provinces with higher indicated education levels 3 shows the results of dividing expenditure by source of have longer life expectancies. When the effect of finance, by resource inputs and by delivery system. The top urban income is assessed separately from that of frame of Figure 3 shows that financing comes in about equal rural income, however, a slight negative impact of amounts from three main sources - private outlays (32 per- urban income on life expectancy is observed. cent, labor insurance (31 percent and state budget expend- 1. The discussion in this section is based on the analysis reported in "The distribution 2. An expanded treatment of the subject of this section may be found in "Health Sec- and impact of health resource availability in China", by N. Prescott and D Jamison. tor Finance in China," tFinancement du Secteur Sanitaire en Chinaj, by N Prescott International Journal of Health Planning and .Management 1985, vol. 1, pp.1-12. and D.T. Jamison, World Health Statistics Quarterly, 1984, V.c. 37, pp. 397-402. 29  itures (30 percent). Production brigades finance the residual Figure 3 7 percent. The bottom frame of Figure 3 shows that the struc- Recurrent Expenditure on Health, 1981: Sources ture of health services delivery is essentially tripartite. The of Finance, Resource Inputs and Delivery Systems rural collective system, principally brigade and commune Sources of Finance health facilities, delivers the largest share of health services, valued at approximately 40 percent of the total; the govern- ment system delivers about 32 percent; and the enterprise Brigades (including system, which benefits enterprise employees and depend- State Budget rural cooperative ents exclusively, delivers 25 percent of total services. Private 30% health insurance) 7% medical practice has recently been authorized and some exists, but it is currently of little significance. Of all resource Labor \inputs into the health sector, shown in the middle frame of lIrsuran'c ~ Figure 3, pharmaceuticals are by far the most important 31% / accounting for 58 percent of total expenditure. Western r Prvate X, /^ Expenditure figure 4 Urban-Rural Differentials in Recurrent Health Expenditure, 1981 Y32.46 Delivery Systems* Per Capita Other Rural Collective Y3.51 / \ ~System\\ Private Expenditure \\ Y2.76 X Svsem Enterprise /\ 25% \\ L ~\ \\ LMedical Education \ 3% State Incliudes privaLte prao u whu h is negligblck Subsidies \ Y26.19 \ \ Resource Inputs \ \ Y9.64 Traditional Beds | P Capita Pharmaceuticals-/ 13% \ Y1.81 Equipmnent-5 Western E- Pharmaceuticals Y5.07 Brigade Personnel- 4 \, Salaried/ Personnel/ 2 00%i /Y2.76 Urban Rural 30  Table 5 China: Selected Development Indicators, China and Other Countries Growth Rate Population Life Expectancy Daily Adult Per Capita of Per Capita Growth Rate, Total (Years) Per Capita Literacy GNP, 1980 GNP, 1960-80 1960-80 Fertility Gain Energy Supply, Rate, (1980 US$) (% Per Annum) (% Per Annum) Rate, 1980 1980 1960-80 1977 (kcal) 1977 (%) Low-Income Economies (excluding India and China) 230 1.0 2.5 6.1 57 15 2,113 34 India 240 1.4 2.2 4.9 52 9 2,021 36 Sri Lanka 270 2.4 2.0 3.6 66 4 2,126 85 China 290 3.6* 1.8 2.5 67 27** 2,237 66 Pakistan 300 2.8 3.0 6.1 50 7 2,281 24 Indonesia 430 4.0 2.2 4.5 53 12 2,272 62 Thailand 670 4.7 2.7 4.0 63 11 1,929 84 Middle-income Economies 1,400 3.8 2.4 4.8 60 9 2,561 65 Hong Kong 4,240 6.8 2.5 2.2 74 7 2,883 90 Non-Market Industrial Economies 4,640 4.2 0.9 2.3 71 3 3,489 100 Industrial Market Economies 10,320 3.6 0. 9 1.9 74 4 3,377 99 'This figure or China is the growth rate om gross domestic produt t GDP' minus the p)opLulation growth rate. F or the period 190(-8f), ( l.I' and(i e le hii .c gre wat apl)roximateIl the same rate. *1 9h() and the adia ent vear, *vvrr periods of a( ute tframine and turmoil in China, tshcc h restulted In ubhstanti.lK elevate d mortalit0s rates The 2-sear gain in lite (expe( tan rtv portd here is, therefore. based oni an imp)ute'd 1 t) lifel ipe taM thit is the avera.age t tlhe 1 I,)- and 1 () I litte expw( tant( ie. Sooer( v W)R82-kVorldrJet-opmeolt Rei poit 1082 %ashington. D).( tihe World Ban, l 82f WM, ) dtctinvs low-inn(ome tonomiru e as those hating a ptr c(apita in(omit or$41 P or tes itn 1080. 3 3 (lah e(ononlhes are in(fltided in M,f )R82 tahles The tmiddle- in(ome e( onomistt aire thoe Xot il(eteo-tingc ountries that ha le per. apita in onu s et n $41( lintl S4. 5 1: this group olltludes 2 (ountries The "non-market industrial st onoln ie' ha"t e inorime, ranging trom $ fi900 to $ 181 tont thte Inc1usti.l market monoinues ha\ve im onite rangintg trom 54.,880 to $ I6 44ll drugs alone consume 49 percent of total health expenditure than 51.50 in urban areas butover$2.50 in rural areas. Given out!ays.1 the stated emphasis in China on preventive medicine, it is A remarkable feature of the financing profile is noteworthy that less than 5 percent of total health care the high proportion of total expenditure that is mediated resources is directed to prevention and over 95 percent is through insurance schemes. This reflects the high degree of directed to provision of curativeservices. (itshould be noted health insurance coverage, which constitutes one of the that, since virtually all financial outlays for prevention are major achievements of the Chinese health system. It is esti- provided bythe state, much more than 5 percentof thestate mated that only about 30 percent of the population is com- health budget goes for prevention; MOPH estimates that pletely uninsured. Most (but not all) of the uninsured are 13.8 percent was spent on epidemic prevention stations rural dwellers, and considerable urban-rural differentials in and parasitic disease control in 1981.) Tradition,i medical health expenditure exist in China (Figure4). Urban expendi- practice receives a more substantial allocation - perhaps 14 ture is estimated at more than $1 6 per capita, more than tri- percent of the total. ple the rural expenditure. Also striking are the patterns of public expend- itures on health. State subsidies for health in urban areas International Comparisons: exceed by a factor of almost ten those for rural areas - Available data suggest that per capita health expenditure in approximately $13 per capita compared to less than $1.50 China is relatively high for a country at its low per capita per capita. Private expenditures per capita amount to less income level Comparisons with 19 LDCs with 1975 per capita GDP lower than US$2,000 are presented in Table 6. I The cost of unpaid labor time, either voluntarily or otherwise mobilized, is not in- For these LDCs, the relationship between per capita health cluded in the discussion here because virtually no relevant data exist. Anecdotal expenditure anid per capita GDP is very close to log-linear, accounts suggest, however, that substantial amounts of unpaid laborcontribute to the success of many preventive efforts. particularlv those associated with env;ron- with an estimated income elasticity of health expenditure of mental sanitation. For example, it has been estimated that the economic cost of approximately 1.2. At constant 1 975 prices, percapita health voluntary labor mobilized to work in the schistosomiasis control campaign in Shanghai Countywas approximately$4 million between 1953 and 1978. equiva- expenditure in China is estimated at US$7.1, which is 25 lent to nearly double the financial costs incurred for drugs, surgery molluscicides percent greater than the US$5.7 per capita expenditure and salaried personnel. See Chen) et al 19821. "Eradication of Schistosomiasis le Amencan ournal ofPubic Health 1972 (9 Siupplemenc 1;0-S1. level predicted from the LDCsample. 31  labl,' f, China: International Comparisons of Health Expenditures LDC Sample, 1 975* Health Expenditure Per Capita (US$) GDP Per Capita (US$) Health Expenditure % Share of GDP Malawi 2.1 138 1 5 Kenya 8.9 241 3.7 India 3.4 146 2.3 Pakistan 7.2 189 3.8 Sri Lanka 4.1 185 2.2 Zambia 16.6 494 3.3 Thailand 14.8 350 4.2 Philippines 8,7 376 2.3 Korea 16.3 583 2.8 Malaysia 21.1 781 2.7 Colombia 23.2 564 4.1 Jamaica 43.9 1,406 3.1 Syria 16.5 718 2.3 Brazil 44.2 1,149 3.8 Romania 48.0 1,742 2.8 Mexico 64.3 1,465 4.4 Yugoslavia 72.6 1,663 4.4 Iran 48.4 1,587 3.1 Uruguay 66.3 1,308 5.1 Mean 27.93 794 3.3 China (1981) Actual 7.1 214 3.3 Predicted 57*" - - Derived frorm Kravis. et al t19821. .Vorl Produat and I o trJr71 Iti(er)lli)rinaP ( r .ipirisr ut s rr d uriluc( 1, l n itn iore; Johns Hiopkins University Pr.ss. Sumrmari of Multilateral Tables 6-1 and 6-3. Figures in( lude hoth publi and private expenditures. Predi(tion is based on logarithnmi regression of health expenditure on GDP 32