HEALTH SECTOR REFORM IN BRAZIL 1

Transcription

1 HEALTH SECTOR REFORM IN BRAZIL 1 Celia Almeida 2 Tatiana Baptista Claudia Travassos Silvia Porto GENERAL BACKGROUND Brazil is a middle-income country of continental proportions (8.5 million sq. km.), with a total population of nearly 160 million (1999). The country is divided into five geographical regions - North, Northeast, Midwest, Southeast, and South - with different demographic, economic, social, cultural, and health characteristics, constituting a mosaic with extreme inequalities. As a Federal Republic, it includes 27 states and 5,507 municipalities. The great majority (90%) of the municipalities have populations of less than 50,000, but 10% of them account for 60% of the total population: approximately 50% have fewer than 10,000 inhabitants and account for around 9% of the total population, while 0.5% have more than 500,000 inhabitants and account for around 27% of Brazil s population. The urbanization rate varies from 62.35% in the North region to around 89.29% in the Southeast region, which is the most highly developed and holds 43% of the country s total population (IBGE, 1997). In the 1990s, the population growth rate was 7.54%, against 22.7% in the previous decade. Meanwhile, the percentage of children (0 to 4 years old) declined while the numbers of the elderly (over 60 years old) increased. In 1995, the economically active population comprised 72 million workers (approximately 46% of total population), 53 million (74%) of which were unskilled and 18 million (26%) skilled (IPEA, 1997). In the first quarter of 1999, the overall unemployment rate stood at around 8.0%, although it has increased recently and the figure is much higher in urban areas 3 (IBGE, 1998). In addition, estimates indicate that, in 1998, the informal sector accounted for some 48.5% of those individuals with some kind of occupation 4 (MTb, 1998). Nonetheless, market studies admit that the level of informal occupation has grown so quickly in recent years that measurements are jeopardised and understated. The 1999 Gross Domestic Product (GDP) was the equivalent of US$ 556,83 billion (Central Bank - O Globo newspaper, February 13, 2000). Between 1990 and 1999, GDP grew at an average 1.39% per year, less than half the average rate for (2.96%) and a far cry from the average of 8.78% achieved in the 1970s 1 This narrative is based on the following studies: Almeida, C.M.; Travassos, C.; Porto, S; & Baptista, T.W.F. (1999) A Reforma Sanitária Brasileira: em busca da Eqüidade (Health sector reform in Brazil: the search for equity), Research in Public Health, Technical Report, Washington: PAHO (89 pp.).; Almeida, C.M.; Travassos, C.; Porto, S. & Labra, M.E. (2000) Health Sector Reform in Brazil: a case study of Inequity. International Journal of Health Services, vol. 30, n. 1, 2000; as well as on data from the study by Almeida, C.M. (2000) A reforma Sanitária Brasileira: atores, regulação e eqüidade, ongoing research. 2 Celia Almeida, MD, MPH, Ph.D, Associate Professor and Senior Researcher in Health Policy and Health Sector Reform, DAPS/ENSP/FIOCRUZ. Mailing address: Rua Leopoldo Bulhões 1480/715, Manguinhos, , Rio de Janeiro, Brazil. Phone/fax: (+55-21) or calmeida@ensp.fiocruz.br. Claudia Travassos, MD, MPH, Ph.D, Senior Researcher, Health Care Services Research, Health Information Department, Center of Scientific and Technological Information (DIS/CICT/FOCRUZ). Phone: (+55-21) , claudia@procc.fiocruz.br. Silvia Porto, Mathematician, Master in Public Administration, Ph.D in Public Health, Associate Professor and Senior Researcher in Health Policy and Planning, DAPS/ENSP/FIOCRUZ. Phone/fax: (+55-21) sporto@marlin.com.br. Tatiana W. de F. Baptista psychologist, Master in Social Medicine, assistant researcher in Social Medicine Institute/State University of Rio de Janeiro, ernanisb@mandic.com.br 3 In the metropolitan region of Greater São Paulo, in the first trimester of 1999, the open unemployment rate was 8,81%, representing 16.3% of the Economically Active Population. 4 In 1997, the informal market accounted for a volume of trade equivalent to 8% of the GDP (Sebrae, 1999).

2 (FGV, 1999). Per capita GDP was U$3,396 in 1999, lower than in 1994 and a sharp drop of 32.6% from the 1997 value (Table 1). Table 1 Main Economic Indicators Indicators Years GDP (U$ billion) 496,183 (1) - 569, , , , ,83 GDP per capita 3,322 (1) 3,569 4,554 4,920 5,037 4,798 3,396 (2) Source: IBGE apud FGV. Conjuntura Econômica (1) Database, FGV/Rio de Janeiro. (2) IMF apud O Globo newspaper, Sunday, February 13, The 1997 Human Development Index (HDI) a calculation based on life expectancy at birth, level of schooling, income and access to resources published in 1999, was 0.739, which ranks Brazil 79 th in the world. According to the HDI, only the states of the South and Southeast regions may be considered as attaining a high level of development; the other states (particularly in the Northeast) are either average or under-average (UNDP, 1999). The UNDP Report indicates that the main problem in Brazil is the gap separating off the upper stratum. Brazil is the Latin American country where inequalities among socio-economic groups are most extreme: data of the 1996 National Household Sample Survey (Pesquisa Nacional de Amostra por Domicílios (PNAD, 1996) showed that the wealthiest 20% of the population account for 65% of national income, while the poorest 50% obtain 12% of income (PNAD, 1996; UNDP, 1999). The average income of the wealthiest 10% is around 30 times larger than the average income among the poorest 40%. The fraction of income appropriated by the wealthiest 20% grew by 11% between 1960 and 1990, while the income of the poorest 50% fell by 6% and that of the intermediate classes remained almost unaltered (UNDP, 1999). The total number of Brazil s poor was estimated at 42 million at the start of the 90s, account for around 30% of the total population, a proportion that is higher in the North and Northeast regions and concentrated mainly in the metropolitan regions 5. The latest (1999) estimate made by the Institute for Applied Economic Research (IPEA) is that 45% of the population live in conditions of poverty; that is, 69 million survive on less than US$ 66 per capita per month (O Globo, June 27, 1999). Figures from the Brazilian Geographical and Statistical Institute, drawn from the 1996 PNAD, reveal a total of 27% of the population with income of less than half a minimum wage (US$ 33, at present day values), with the Northeast and North regions accounting for the most alarming figures - 50% of the population of the Northeast and 33% of the North with incomes of less than half a minimum wage). The illiteracy rate in Brazil is high (14.7%) and the average level of schooling is low, with considerable variations between regions and between urban and rural areas: in 1997, the highest illiteracy rate was in the Northeast region (29.4%), concentrated in the rural area (45.83%) (IDB 98). 1. HEALTH NEEDS Brazilian society has undergone important demographic and epidemiological changes that, taken together with the economic and social context, have resulted in changed health conditions for its population. From 1960 onwards, population growth slowed considerably, with fertility declining by more than 50% in thirty years 6. Note that at the start of the decade the average fertility rate in urban areas was 2.5 children per woman and, in rural areas, was 4.4 children per woman (IPEA, 1997), highest rates being in the Northeast (IDB 98). In the 90s, the average fertility rate continued to decline, and in 1997 the rate was 2.4 per woman. 5 The figures are confirmed by 1996 UNDP Human Development Report and by government research institutes - Institute of Applied Research (IPEA) and Brazilian Census Bureau (IBGE): 30% are poor (39% in rural areas and 56% in urban areas), and are to be found mainly in the metropolitan regions. The Northeast region accounts for 46% of this population (41% in urban areas and 42% in rural areas) and the Southeast region has the second highest number of poor - 34% (53.3% in metropolitan regions and 32.3% in urban areas) (UNDP, 1996; IPEA, 1998; IBGE, 1996). 6 The total fertility rate declined from 6.1 in the 40s to 2.9 in the 80s and 2.4 in the 1997.

3 The pattern of inter-regional differentiation persisted, although the gaps closed somewhat: the lowest fertility rate was recorded in the Southeast region (2.13 children per women in 1997), against 3.28 children per woman in the North. Nonetheless, the population pyramid is still relatively young, while the relative number of elderly is increasing, the most recent figure being 7.87% of the population aged 60 or over (IDB 98). Health indicators have improved significantly in recent decades (Duchiade, 1995; Almeida et alli, 2000): life expectancy has increased and the general and cause-specific death rates have diminished, but regional differences and variations by income level persist. Life expectancy at birth illustrates this point quite clearly: it is lower in the lower income groups (62.2 years in Alagoas) and in poorer regions (64.83 years in the Northeast) and higher in high income groups (71.03 years in Rio Grande do Sul) and in the richest region (70.34 in the South). The differences between urban and rural areas remain, life expectancy being lower in the rural areas in both regions and markedly lower in the Northeast. Another differential are the average values among women and among men. Among women, life expectancy at birth is years, while the average expectancy among men is years, which also confirms a world trend. Proportional mortality rates by cause of death have changed significantly, with a marked rise in diseases of the circulatory apparatus (32.40% of deaths in 1997); external causes - accidents and violent deaths in general (15.5% of total deaths in 1997) and neoplasias (13.88% of total deaths in 1997), across all the macroregions in Brazil (Table 2). The death rate from infectious and parasitic diseases continues high, with AIDS (21% of IPD deaths) as the main cause in this group, affecting mainly the women. Table 2 - Proportional death rate by groups of causes - Brazil 1997 Groups of causes North Northeast Southeast South Mid-West Total Region Region Region Region Region Total 100% 100% 100% 100% 100% 100% Infectious and 8.10% 7.80% 5.97% 4.45% 7.49% 6.22% parasitic diseases Neoplasias 11.23% 10.58% 14.22% 17.48% 12.32% 13.88% Diseases of the 24.40% 30.35% 33.24% 34.77% 29.43% 32.40% circulatory apparatus Diseases of the 9.10% 9.03% 11.44% 11.90% 10.27% 10.91% respiratory apparatus Perinatal Ailments 10.97% 6.84% 4.22% 3.29% 5.53% 4.89% External Causes 20.03% 17.15% 14.88% 13.17% 20.00% 15.51% Other Causes 16.17% 18.25% 16.02% 14.94% 14.97% 16.18% Source: MS/FNS/CENEPI/SIM In 1997, infant mortality rates ranged from per 1,000 live births in the Northeast Region to per 1,000 live births in the South Region (MS/FNS/CENEPI/SIM 2000). In 1990, these rates ranged from 71.4/1000 live births in the Northeast to 30.9/1000 live births in the South (IBGE, 1997); that is, an improvement of 18.3% in the Northeast and 22.2% in the South. The infant mortality rate declined more slowly in the 1980s, especially in the poorer regions ( 15.28% in the North and 17.41% in the Northeast versus more than 35% in the Southeast, South, and Midwest), further increasing inequalities among regions and income groups (Simões, 1992). Rates of infant deaths due to diarrhea are also lower in the more developed regions of the country. The incidence and prevalence of epidemic and endemic infectious diseases are on the rise, spreading across the country as well as between social groups. That is to say, regional health differences persist, infectious and parasitic diseases are still the second cause of death among infants in the North (10.6% of infant deaths) and Northeast (13.4% of infant deaths), with high incidence of diarrheas (between 55 and 65% among the infectious causes), while in the South and Southeast, IPDs are the fourth cause of death, responsible for 7% of infant deaths. In the South and Southeast regions, congenital malformations rank second among the causes (accounting for 15.7% and 12.5% of deaths by region, respectively), indicating a shift in the epidemiological profile of the regions. Certain indicators highlight the regions differences. In 1997, there were cases of dengue per 1000 inhabitants; in the Northeast, the rate was 43.28, while in the South it was Malaria is another good example: new cases were recorded in the North Region in 1997, against very few new cases in the

4 Southeast, mainly due to migration. The tuberculosis rate is increasing, and remain practically equal over all regions, except the South and Mid-West, as can be seen in Table 3. Groups of causes North Region Table 3 Incidence of Transmissible Diseases - Brazil 1997 Northeast Region Southeast Region South Region Mid-West Region Total Cholera 0,41 6, ,91 Leprosy 8,45 3,21 1,62 0,85 6,99 2,82 AIDS 0,41 0,45 1,85 1,52 1,21 1,25 Malaria 361,98 6,37 0,11 0,15 13,02 29,07 Dengue 19,11 43,28 3,37 0,30 12,04 15,97 Tuberculosis 5,82 5,30 5,84 3,61 3,86 5,22 Source: MS/FNS/CENEPI Little information is available as regards morbidity indicators for chronic and degenerative diseases, and information systems have not been developed for these records. The results of inquiries in nine Brazilian state capitals (1988) suggest that the average prevalence of diabetes mellitus in the population between the ages of 30 and 69 is around 8%, and the prevalence of hypertension in the over-20s is estimated at 10% (MS/FNS/CENEPI/2000). One source of information that may be used to infer morbidity trends is the hospital admissions record (SIH/SUS - MS/DATASUS). In 1997, this source indicated that pregnancy, childbirth and puerperium represented the main cause of admission (26.42%), followed by diseases of the respiratory apparatus (17.32%) and of the circulatory apparatus (9.71%). At the same time, there can also be seen an increase in the proportion of admissions for external causes (5.9%) and a constant rate of admission for neoplasias (2.97%). Infectious and parasitic diseases, accounting for 7.96% of admissions that year (1997), were most present in the North and Northeast regions, averaged 12% of admissions (MS/FNS, 2000). Interregional differences can thus be seen to exist in hospital morbidity. 2. HEALTH SYSTEM REFORM The movement for health sector reform in Brazil dates from the 1970s, but the landmark was the new Constitution approved in This Constitution institutionalized the universal social rights, including the right to health, which then became a citizen s right and the state s obligation and responsibility. This extension of rights obeyed a dynamics contrary to global tendencies that were becoming apparent internationally. Health sector reform in Brazil, as idealized and implemented at the end of the 1980 s, was one of the last expansionist reforms of that decade. This reform process was based on ideas that had oriented the organization of health systems in the post-war world and was strongly influenced by the philosophy of the State as service provider. Implementation, however, was fraught with contradictions and conflicts due to both domestic and international political and economic conditions as well as sectorial constraints. The context in which Health Sector Reform was formulated and the early measures implemented in Brazil was characterized by a combination of economic crisis, substantial change in the political regime, revitalization of the party political system and of organized civil society, and social participation in the transition process. This all had considerable influence on government policy agendas and the very way in which social and economic issues were handled. Health sector reform proposed the following: 1) strengthening the public sector; 2) increasing funding and diversifying sources; 3) decentralizing the system; 4) rationalizing service supply and delivery; 5) reorganizing sectorial interests, with a new definition of the relationship between public and private in the way health care was organized. Certain dimensions of the decentralization process, including changes in rights and responsibilities at the different levels of government and in the rules governing financial allocations, acquired specific characteristics in the healthcare area, where important links with the centre were maintained.

5 Delivery of Health Services: components of the Health System The Brazilian health system is a complex network of service providers and service buyers that are simultaneously interrelated, complementary and competitive, forming a complicated public/private mix funded primarily by public resources. The Unified Health System (Sistema Único de Saúde/SUS) was created by health sector reform in 1989 with the intention of favouring the public sector and forming a regionalized system organized by level of competence and type of care, where private care would be complementary. In practice, however, these objectives have not been attained. The system is composed of three main subsectors: 1. The public sector, which comprises publicly financed and provided health services, including services from the federal, state, and municipal levels and the armed forces, which have their own separate health care services. 2. The private (profit and nonprofit) sector contracted by the public sector and paid through reimbursement systems, comprising publicly financed and privately provided services. 3. The free-choice private sector, financed out-of-pocket or by corporate health insurance, comprising privately financed and privately provided services with different levels of insurance premiums and tax subsidies. This system is highly centered on health care and the hospital, characteristics that gained strength in the 1960 s with the development of social security. Unlike most health care systems, general practitioners do not play a preponderant role in the Brazilian Health System, which as a rule is very much directed toward specialized care. The three subsectors of the system of health care services - public, private contracted out and private - make up a disconnected nonhierarchical network that delivers different levels of care to different population groups. The private system is considered supplementary and consists of doctor s offices, specialized clinics, and private hospitals (for-profit and nonprofit) contracted by private health insurance plans and/or paid for directly by the patient. It is estimated that the SUS is responsible for 70 percent of total health expenditures and annual hospital admissions (Almeida et alli, 1998), and 26.8 percent of the Brazilian population is currently covered through private health insurance plans (Almeida, 1998). The public network is composed mainly of primary health care services - Health Units and Health Centers - (95 percent) and emergency services (65 percent), while specialized care (74.5 percent) and hospital care (79 percent) are concentrated in the private network (Table 4). In 1992, the system had a total of 24,096 primary health care establishments, 8,042 outpatient clinics, 8,668 clinics for complementary diagnostic exams, and 7,057 hospitals, amounting to nearly 50,000 registered establishments (Viacava & Bahia, 1996). Table 4 - Proportional Distribution of Public and Private Health Care Services, Brazil, Health Services Type of Service Public (%) Private (%) TOTAL (%) Primary Health Care Units and Health Centers Specialized Health Care Emergency Hospital Source: IBGE/AMS, In: Cohn, A. & Elias, P. Saúde no Brasil: políticas e organização de serviços. Cortez Editora, São Paulo, p.78, The basic health services network underwent marked expansion, especially from 1980 to 1986, mainly in Health Centers (8.6 percent per year). Their numbers increased nearly fivefold as a result of the rationalizing measures that restricted hospital admissions and encouraged outpatient care, following significant changes in health policy aimed mostly at cost containment. Since implementation of the SUS (1990), the municipalization process has been intensified, with a considerable increase in the number of Health Units and Outpatient Clinics (6 to 7 percent per year), despite the reduction in federal public investments from 1990 to This growth has varied across the regions, with higher rates in the North and the South (growth rates 482 percent and 295 percent, respectively) and predominating at the municipal level. Heterogeneity is also evident within a region or state, 604 municipalities concentrated mainly in the states of Amazonas, Pará,

6 Piauí, Maranhão, and Tocantins do not have Health Centers. A more homogeneous distribution of these services is seen only in the South and in São Paulo State (Cenepi, 1992). For basic health care services, in 1992 Health Units (lowest level of complexity) represented 27 percent of the total number of ambulatory services in Brazil, with the largest concentration (47.9 percent) in the Northeast. Health Centers represented 46.5 percent, constituting the main type of ambulatory care in the country as a whole, although concentrated in the Southeast (40.0 percent). A comparison between numbers of ambulatory services in each region reveals the predominance of Health Units in the North and Northeast, meaning that other basic health care services are present in smaller numbers; in the more developed regions, Health Centers predominate, with greater balance in the supply of other types of services at that level of complexity (Cenepi, 1992, p.10-12). Specialized care services represent 25 percent of total ambulatory establishments in Brazil, 83 percent of which are mostly private, with larger concentrations in the more populous municipalities, particularly in the Southeast (56.2 percent) and smaller in the North (2.1 percent) (Cenepi, 1992, p.10-12). Seventy-five percent of these services are concentrated in the richest part of the country (Southeast, South). From 1990 to 1992, public specialized care grew less, while this type of service increased in the private sector. Only 28.9 percent of all specialized care services are managed exclusively by the SUS, and nearly 62 percent are private or are contracted out by private health insurance or provided directly through companies that offer pre-payment plans for their employees. The highest concentration of company-linked clinics is in the Southeast. The situation is reversed, however, for hospital services: 80 percent of the network has contractual links with the SUS, a figure that reaches 91.2 percent in the South. Proportionally, there are more hospitals in the Southeast (33 percent) and Northeast (30.5 percent). Almost 40 percent of all hospitals have contracts simultaneously with the SUS and with private health insurance companies, with the highest concentration in the South (70 percent). Although there are no conclusive data, this overlapping of clienteles generally results in a greater availability of hospital beds for private patients, with higher utilization of services by patients with private health insurance plans. Altogether, these hospitals have a capacity of 452,852 beds (336,966 private and 115,886 public), the distribution of which is even more concentrated, with 44.7 percent in the Southeast (Cenepi, 1992, p.24-25). The SUS segment is made up mainly of contracted private hospitals in all regions of the country. Hospital beds per inhabitant are predominantly private for all regions except the North (Table 5). Classification of hospitals according to the number of beds reveals that taking the country as a whole, 50 percent of hospitals have under 40 beds. The distribution by region indicates that the Southeast, with 26 of the 38 hospitals having over 500 beds, has a greater number of large hospitals than the other regions (Cenepi, 1992, p.27). Table 5 - Percentage distribution of public and private hospitals and beds per 1,000 inhabitants by geographical region, Brazil, No. hospital beds Beds/1,000 inhabitants Region Public Private Total Public Private Total North 48.7% 51.3% 100% Northeast 33.6% 66.4% 100% Southeast 22.2% 77.8% 100% South 17.2% 82.8% 100% Midwest 23.3% 76.7% 100% Brazil Source: IBGE/AMS. Viacava, F. & Bahia, L. Assistência Médico-Sanitária. Figures, Radis, No.20, Table 22, Rio de Janeiro, FIOCRUZ, Private health insurance coverage increased by 73 percent from 1987 to 1996; only 20 percent of the population had this type of coverage in 1987 compared with 26.8 percent in This growth accompanied deteriorating public services. However, the increase in supplementary private coverage was not followed by a reduction in the use of public health care services, particularly highly complex or emergency services; rather, as a rule, utilization

7 overlaps. In other words, clients use available services indiscriminately, according to their means and to the severity of their illness or injury (Almeida et alli, 1999; Travassos et alli, 2000). As in the regional distribution of health care services, physicians are also concentrated in the Southeast (58.8 percent). When one considers the physician concentration outside the state capitals, the picture is more striking: 64 percent are concentrated in the more developed regions. In other words, hinterland towns in the Southeast have a much better supply of physicians than do the less developed regions. This distribution of physicians results in a deficit in the poorer regions and towns and a surplus in the wealthier ones. For the distribution of other health professionals (nurses and dentists) the picture is the same: concentration in the richest parts of the country. Funding Health Care: the social security budget and the health sector reform Health sector financing policies in Brazil are characterized by insufficient resources, regressive collection of revenue, vulnerability of funding to economic cycles, centralization, and inequitable allocation. One of the most important social policy innovations introduced by the 1988 Constitution was adoption of the concept of social security, which not only implied an acknowledgment of health as a citizen s right (health understood as a public good and a universal benefit), but also brought about a radical transformation of the health financing system. In theory, the new Constitution promoted cohesion between areas of social protection by creating the Social Security Budget, and diversified financing sources 7. With creation of the SUS, constitutional dictates found material expression in terms of health service system organization, but its implementation and financing have been deeply affected by resource vulnerability. The new sources of funding for social security are also quite sensitive to economic performance (Porto, 1993). Thus, despite changes in the legislation, the negative characteristics that have historically marked funding for the social security system in general, and the health care system in particular, remained unaltered. Growing criticism and difficulties due to sectorial under-funding led to situations of serious fiscal stress, and proposals for an earmarked revenue source to finance the health system gave new momentum to reintroduction of the CPMF (temporary financial transactions tax). This new source was approved as a provisional tax in August 1996, and again in 1998, becoming effective in 1997 and Criticism of the CPMF as an earmarked tax is well known, but the main point is the extreme unpopularity of yet another tax to be paid and the fact that the tax collected was not used for the health sector alone. In short, despite the changes that have been introduced, the constitutional provisions on social security health funding were not sufficiently definite to reverse the adverse historical pattern. Health system funding policy has jeopardized both the course of reform, with consequences in the attainment of equal opportunity for access to health care services, and the quality of services. Since 1990, when the SUS was regulated, federal health funding has suffered oscillations and instability, and the policy of decentralizing funding has been implemented with many restrictions, as can be seen in the next section. Federal social spending stood at around 12.3% of GDP in At that time, federal health spending accounted for 1.7% of GDP. From 1994 to 1998, federal health spending oscillated around an average of 1.9% of GDP and the average public health spending from 1994 to 1996 was around 3.3%. Federal social spending increased during the first part of the 90s ( ), but public health spending and federal health spending diminished (Table 6 ). Table 6 Federal and Public Social and Health Spending 1994/1998 Indicators Years Federal Social Spending - % of GDP (1) 11.8% 12.2% 12.3% Financing sources are the following: the social contributions from the payroll of formal employees and employers were maintained; a new tax was levied on company net profits (CSLL); two other sources - tax on operating revenues (COFINS) and lottery revenues - previously destined for generic social expenditures, were re-allocated.

8 Public Health Spending - % of GDP (2) 3.32% 3.37% 3.17% - - Federal Health Spending -% of GDP 2.0% 2.1% 1.7% 1.97% 1.7% Sources: (1) IBGE (1997). Brasil em números. (2) Institute for Applied Economic Research (IPEA/DISOC), The Ministry of Health's budget decreased substantially between 1989 and 1992, exactly as implementation of the reform was beginning, going from a total of US$11.30 billion to US$6.57 billion. This loss was partly recovered in 1993 and 1994, followed by considerable increases during the first Fernando Henrique Cardoso government ( ), albeit with annual oscillations (US$16.22 billion in 1995; US$14.20 in 1996; US$17.58 in 1997; and US$14.36, approved and not necessarily executed, in 1998). These oscillations were also reproduced in per capita spending which, in 1989, was US$80.38; falling to US$43.98 in 1992 and rising to US$ in 1997 (Table 7). In brief, both total federal investment in health and expenditure per capita decreased in the early 90s, then increased in the second half of the decade, and went into decline again in Table 7 Federal Financial Resources for Health - Brazil Indicator Years US$ billion , (*) US$ per capita Source: Elaborated with data from Almeida et alli (1999) and Levcovitz, E. (1997). Data from the Ministry of Health. (*) Approved and not necessarily executed. Public resources for health were highly centralized at the federal level, representing more than 70 percent of the total national public resources for health since the beginning of the Eighties, and in some of the following years it reached almost 86 percent (Médici, 1994:22), but decreased around 60% in the first part of the 90s (Table 8 ). In the early 80s, State governments contributed with much less than 10 percent, and, in some cases, it didn t provide any resources for health, financing the delivery of services exclusively through federal funds. The municipalities, on the other hand, expanded their participation, surpassing 15 percent of their budgets in some cases (Médici, 1994:23:10). These figures also changed during the 90s. It should be stressed that the proportion of health spending originating from federal and state governments have diminished from 1994 to 1996, and municipal governments have increased as a result of the health sector reform and decentralisation process (Table 8). Therefore, at that stage, the state level played a less important role in the decentralization process. Table 8 Federal, State and Municipal Level Health Spending 1994/1996 Indicator Years Health Spending Federal % of GDP % of total health area State % of GDP % of total health area 22, Municipal % of GDP % of total health area TOTAL % of GDP ,17 % of total health area Source: Institute for Applied Economic Research (IPEA/DISOC), Decentralization Management Model One positive outcome of the Brazilian reform endeavor has been the decentralization process.

9 Management of the SUS was conceived as a strategy to transform a traditional culture of centralization, bureaucratic authoritarianism and lack of participation and social control in health service affairs, and to eliminate the irrationalities and inequalities resulting from that manner of handling health care. The new organization of the health system and services redefined the roles of both executive and representative bodies, such as the Health Councils, at the national, state and municipal levels. The National Health Conferences (first established in 1947) were also reinstated, to be held every four years as legitimate forums for discussing and formulating national health policies, with ample participation by representatives of organized civil society. The Ministry of Health began to implement the decentralization process in 1991, through the administrative instruments known as Basic Operational Rules (NOBs, enacted in 1991, 1993, and 1996), a set of measures designed to create a planning information system, to establish mechanisms for allocating funding to the state and municipal levels, and to define the managerial model. The NOBS have had a profound influence in inducing, on the one hand, a specific kind of institutionalization of the sector decision-making arena and inter-linking among different levels of government and, on the other, substantive change in the reform agenda from its initial standpoint. In the long term, the managerial model introduced in the SUS aims to attain democratization, accessibility and equity through three principal vectors: 1. Consolidating a comprehensive, regionalized and municipalized health system according to a rationale that harmonizes functions across levels of government and integrates services of varying scope and complexity; 2. Investing authoritative bodies with the power to make all kinds of decisions about the health system and to conclude agreements between the parties involved in the management of services, the crucial concept being formation of social partnerships; 3. Assigning local authorities responsibility for the effectiveness and equity of care provided to the population in each territorial unit, especially at the municipal level. The SUS is organized at the federal, state, and municipal levels. The federal level is legally responsible for formulating and implementing national health policy. It is also in charge of system planning, assessment, and control, as well as funding distribution. At that level the decision-making process involves the following: 1. The National Health Council (CNS), which includes representatives of government agencies, service providers, health care professionals and users (the most representative group in the CNS: 50 percent of total members). The CNS is in charge of formulating strategic planning and monitoring implementation of Brazil s health care policy. 2. The Tripartite Inter-managerial Committee (CIT) whose members include representatives from the Ministry of Health, the National Council of State Health Secretaries (CONASS), and the National Council of Municipal Health Secretaries (CONASEMS). Its main function is to approve rules for operating the system. 3. The Ministry of Health, which manages allocation of federal funds under the supervision and approval of the CNS, and proposes operating rules to be approved by the CIT. This management structure is mirrored at the state level (State Health Secretariat, Bipartite Intermanagerial Committee and State Health Council) and municipal levels (Municipal Health Secretariat and Municipal Health Council), coordinated by the respective Health Secretaries. Functions at the state level involve service coordination, distribution of financial resources, and decisions relating to complex specialized technological interventions. These state-level functions are still evolving because so far the decentralization process has focused predominantly on the municipal level. The municipalities are responsible for handling the delivery of goods and services involved in health promotion, preventive care health care and rehabilitation. The decentralization strategy adopted by the central level through the NOBs is designed to establish the necessary managerial conditions to enable states and municipalities to assume the planned responsibilities and functions. Following NOB 93, two management statuses (intermediate and advanced) were defined for states and three (initial, intermediate, and advanced) for municipalities. Classification into one of these stages involves clearly defined responsibilities and requirements, and financial incentives have been created to spur

10 the process. Municipalities that fail to participate in this qualification process continue as service providers, with the corresponding state assuming responsibility for SUS management in that municipal territory. NOB 96 introduced some fundamental alterations. The central objective of this NOB is to promote and consolidate full exercise of the health care managerial function by municipal government. The main innovation is the priority given to basic care, understood as a package of procedures to be defined locally according to certain parameters and financed through the Basic Health Care Quota (Piso Assistencial Básic, PAB) distributed per capita to municipalities. At the same time, financial incentives are also being put in place to encourage the introduction of two programs, the Family Health Program (PSF) and the Community Health Agents Program (PACS). Two managerial situations (full management of the entire system and management of basic care) are currently in place for municipalities and one (full management of the entire system) for states. NOB 96 also reversed the previous shift toward the municipalities, and restored and strengthened the states role in coordinating and conducting the state health system, of which the respective municipal systems are part. The decentralization process entails a politicization of managerial decisions that demand strong involvement of the actors participating in the process. On the other hand, the process requires far-reaching changes at the technical-operational level, demanding a capacity-building effort, as well as a revision of values, habits, and procedures. A number of recent studies have revealed at least two important problems: a) radical municipalization has led to greater fragmentation of the system, divesting the states of authority (Dain, 1995; Arretche, 1997; Levcovitz, 1997; Costa & Ribeiro, 1998); and b) the very composition of the intermanagerial forums reproduces existing power relations, maintaining historical inequalities in the distribution of funding (Ramminger, 1997; Almeida et alli, 1999). Nevertheless, several innovative experiences are taking place successfully at the local level throughout the country. Fund Allocation in the Unified Health System The systematic distribution of financial resources from the federal level to the states and municipalities consists of several fund flows, the most important of which are destined for a) funding hospital activities and b) funding outpatient activities. The same payment mechanism used for reimbursing hospital admissions provided by private services contracted out by the federal level was also adopted in 1991 to finance hospital admissions provided by public services, thus becoming one of the most important criteria for financial allocation. The allocation of financial resources is accomplished by an Authorization for Hospital Admission (AIH) billed by the different providers - public, philanthropic or private hospitals. The AIH is applied exclusively to the payment of hospitalizations that are reimbursed through a prospective payment system. The payment unit in this system is the procedure ; the value of each procedure is defined previously at the central level, without distinguishing between the different providers (except for university hospitals). In other words, financial resources are distributed through a prospective payment mechanism. Each state has an AIH quota and a financial cap. However, the definition of the AIH quota merely expresses the population size of each state, without taking into consideration distributive factors such as demographic inequalities, epidemiological profile, and regional socioeconomic conditions (Médici, 1991; Rezende, 1992). The financial cap, in turn, establishes maximum limits that tend to perpetuate the existing distributions. In summary, funds are allocated to each state according to AIH billing, always within the corresponding quantitative and financial ceilings. When the difference between AIH billing and the financial cap is positive, the balance is passed on to the local authorities, provided they are classified as intermediate or advanced management. This decision means that the total funds allotted to the state can be passed on only to those municipal authorities that fulfill the requirements and responsibilities established for the more advanced managerial stages; the remaining local authorities receive resources according to billed service production. As described above, the requirements and responsibilities defined for the different managerial stages make up an extensive and growing set of factors that affect the allocation of financial resources and the effective implementation of the decentralization process. Of these factors, three are foremost: the Health Councils, the Unified Health Funds and the need for technical conditions for programming and monitoring

11 service delivery. On the other hand, payment based on service production has set up competition between public and private service providers with different degrees of administrative autonomy. Public institutions cannot decide wage policies, create incentives to increase productivity or realize investments. Outpatient services delivered by the different service providers - public, philanthropic, or private - are reimbursed according to amounts established at the federal level. However, the total resources allocated for each state may not exceed the financial caps, which are based on observed historical expenditure. The crucial problem in this flow is the manner of defining these caps, which rather than diminishing existing disparities have consolidated inequalities (Porto & Ugá, 1992). The list of requirements for receiving the total resources for outpatient services is smaller than that for funding hospital activities: the State Health Secretariats need only be classified as initial management in order to receive the difference between billings and the defined financial cap. Another financial incentive was intended to stimulate the managerial development of local health authorities: states and municipalities could receive additional resources of about 5 percent of the corresponding outpatient cap. However, because of insufficient funds, this incentive to improve management was not implemented and the positive balance between billings and financial caps was not passed on to the corresponding local authorities. In short, the logic behind allocation of funds in NOB 93 was tied basically to service production levels, thus introducing a service buying-selling relationship between the different levels of government and failing to take into account fundamental factors that would permit more equitable interregional distribution of funds. Thus effective equality in access to health services, as guaranteed by the Constitution, cannot be achieved. A look at the estimated distribution of resources by region according to population demonstrates that the poorer areas have the biggest differential rates. Regions with precarious socioeconomic and health conditions continue to receive fewer resources per capita than the more developed regions (Porto, 1997; Almeida et alli, 1999). Finally, as mentioned earlier, in 1998, changes were introduced to the distribution arrangements implemented by the Ministry of Health, especially financing of ambulatory activities through the PAB (NOB 96), which is distributed per capita to municipalities. These norms have become operational very recently and not enough data is available evaluate this process. Health Care Regulation It has required great regulatory activism on the part of the State to given material form to health reform in Brazil, in view of the bring a multiplicity of institutional, technical and social goals and interests into line behind implementation of the SUS, the backbone of which resides in the policy of decentralization. This has entailed developing regulatory capabilities with vast scope for intervention, including relations among powers, among institutions of the Executive, among levels of government and among stakeholders (public and private); and defining rules for funding, for paying expenses of programs and actions, and for overseeing the private health care sub-sector. In parallel, the new regulations on private health insurance constitutes a great advance for the sector. Private health insurance accounts for a significant part of the private health services market in Brazil. It has grown considerably in the last decade, in step with implementation of the Unified Health System (SUS) and the intensification of conflicts among the various actors participating in this arena of the health sector, including users of health plans, accelerating the process of approval of specific regulation on the matter. At present, it is going through a period of transformation that may constitute a significant turning point in relation to its previous history (Almeida, 1998). The new regulation that has been approved provides for important changes, but also supports and legitimates companies selective practices (risk selection) and does not clearly define mechanisms to enforce the measures proposed. It concentrates basically on certain dimensions of private health insurance plan regulations, while leaving substantial lacunas; that is, it aims fundamentally to maximize consumer participation in the market, but offers no effective guarantees for their protection. As regards the stability of the market, and the related subsidies and incentives, the new law leaves innumerable questions unanswered (Almeida, 1998).

12 3. EQUITY IN THE REFORM AGENDA: GENERAL TRENDS The current debate centres on administrative reform at the government level, of the State, and the government proposal is clearly in keeping with the broad international reform agenda, including both privatization and contracting out. This implies a delimitation of institutional scope, as well as the creation, redefinition, and transformation of institutions. But social and health services, among others, are not considered activities exclusive to government, that is, they can be carried out by any organization with private legal statutes, albeit financed by public resources. It is this conception that has given rise to the proposal to transfer these activities to the non-state public sector (known as publicization), to achieve more flexibility within the public sector and to relieve the government of duties in health care delivery. A focus on managerialism and market-oriented reforms directed to responding to consumer demands and the introduction of competitive mechanisms to stimulate better service performance are the main government reform guidelines. Nevertheless, a number of mechanisms to make management more flexible are already in place in various sub-national administrations, and the process of decentralization itself has exacerbated the fragmentation of the system and weakened the legitimacy of the SUS as a national institution. At the same time, parliament and government are engaged in a clash over the regulation of private health insurance, whose lobbies are powerful, well organized interest groups. This dynamics, in addition to the impasse brought about by the economic situation, by the government s reform proposals, and by the 1998 re-election of President Cardoso, challenge the implementation of health reform as it was initially visualized. There was a turning point in the reform agenda with NOB 96, implemented in 1998: the role of the federal government as formulator and regulator of the system is being consolidated; the change in the role of state and municipal governments, from provider to system manager, is being made more explicit; and the state level is being restored as the sphere for co-ordination and articulation of the various municipal health service networks. Financing by automatic per capita transfer (PAB), although more redistributive than the previous allocation arrangement, may introduce new distortions (Almeida et alli, 1999). In summary, a new health care system is being defined which, although decentralized, is still strongly dependent on the federal government. We cannot yet evaluate the results of this new dynamics, which has only recently been inaugurated. While having the merit of encouraging autonomy and strengthening the local level under the co-ordination of the states, possibly making health services more effective, it may also further fragment the system and render service distribution less egalitarian, thus consolidating an increasingly dual system. In any case, the government tends to favour a very narrow concept of social policies focused on the poor. The federal government does have a rhetorical commitment to the principles of the SUS, but the actual implementation does not fulfil the expectation of overcoming inequalities and improving the public health care system on which nearly million Brazilians depend. Although the term equity appears ambiguous in the Constitution, this concept is implicit in the universality of health care coverage and the equivalency of benefits and services for urban and rural populations. Briefly, the concept in the constitutional text establishes equal opportunity of access. Regarding social security in general and health in particular, the constitutional text allows for different interpretations of equality of opportunity. The real objective of this concept, however, can be inferred from the contents of the Organic Health Law (Ministry of Health, 1990), which covers those extra-sectorial factors that determine health conditions - food, housing, water and basic sanitation, income, education, and access to essential goods and services, among others. As such, it can be inferred from the Constitution and the complementary legislation that the concept of equal opportunity of access to health care services refers to equality of opportunity for equal needs and implies a positive discrimination to compensate for what are considered socially unjust, existing inequalities in the determinants of the population s health. The policies implemented, however, have not guaranteed the effective exercise of these rights, but restricted their validity to no more than a formal definition. In other words, the law may be progressive, but the same cannot be said of the practice. One of the main obstacles to greater equity is, without doubt, health sector funding policy, which is characterized by the continuing regressive collection of revenue and allocation of funds. Furthermore, the

13 perverse historical development of the public/private mix, the lack of adequate and effective regulation, limited managerial capabilities, the failure to assign a clear priority to health on the government agenda highlight both the poor capacity for policy enforcement and a lack of accountability regarding citizens right to health. It is fair to assume that this debate will continue to be complex and that the entire reform process will remain extremely conflictive.

14 REFERENCES ALMEIDA, C. (1998) O Mercado Privado de Serviços de Saúde no Brasil: panorama atual e tendências da assistência médica suplementar. Texto para discussão No. 599, Brasilia:IPEA. ALMEIDA, C.; TRAVASSOS, C.; PORTO, S. & FARIA, T. W. de B. (1999) A Reforma Sanitária Brasileira: em busca da eqüidade. Research in Public Health, Technical Report, Washington:PAHO (89 pp.). ALMEIDA, C.; TRAVASSOS, C.; PORTO, S. LABRA, M.E. (2000) Health Sector Reform in Brazil: a case study of inequity. International Journal of Health Service, 30(1). ARRETCHE, M. (1997) O Mito da Descentralização como Indutor de maior Democratização e Eficiência das Políticas Públicas. In: S. Gerschman & M. L. Werneck Vianna (Ed.) A Miragem da Pós-Mordenidade. Democracia e políticas sociais no contexto da globalização, Rio de Janeiro:Ed. FIOCRUZ. CENEPI, (1992) Informe Epidemiológico do SUS, Ministry of Health, Brasilia:National Health Foundation. COSTA, N.R. & RIBEIRO, J.M. (1998) Descentralização e política social: o caso do setor saúde. Unpublished working paper. Rio de Janeiro:National School of Public Health, Oswaldo Cruz Foundation. DAIN, S. (1995) O financiamento da saúde na encruzilhada - financiamento público, universalização e descentralização da saúde. Paper presented at the workshop A construção do Sistema Único de Saúde e os debates da Reforma Constitucional, April 22, Maceió, Brazil. DUCHIADE, M.P. (1995) População brasileira: um retrato em movimento. In: M.C.Minayo (ed) Os Muitos Brasis: saúde e população na década de 80. Rio de Janeiro:Hucitec/Abrasco. FGV. (1999) Conjuntura Econômica, 53(6, June). IBGE. (1997) Brasil em Números, Vol.5, Rio de Janeiro, Centro de Documentação e Disseminação de Informações. IPEA. (1997) O Brasil na Virada do Milênio: trajetória do crescimento e desafios do desenvolvimento. Brasilia:IPEA. LEVCOVITZ, E. (1997) Transição X Consolidação: o dilema estratégico da construção do SUS um estudo sobre as reformas da política nacional de saúde. Ph.D Thesis, Institute of Social Medicine, Rio de Janeiro State University. MÉDICI, A C. (1991) Perspectivas do financiamento à saúde no governo Collor de Mello, Série Economia e Financiamento, 43:43-44, Brasilia:PAHO Publication. MÉDICI, A C. (1994) Gasto com saúde nas três esferas de governo , PAHO Publication, Série Economia e Financiamento, No. 4:22, Brasília, Ministério da Saúde, Brasil (1990), Lei Orgânica da Saúde. Lei 8080, Artigo 2, Brasília, D.F. O GLOBO (1999) Plano Real, as duas faces de uma mesma moeda. Article by Flávia Oliveira. Economy Section, Sunday, June 27, 1999, p.39. O GLOBO (2000) PIB Brasileiro passou de R$ 1 trilhão em Economy Section, February 13, 2000, p.38.

15 PORTO, S.M & UGÁ, (1992) M.A. Avanços e percalços do financiamento do setor saúde no Brasil. In: E. Gallo, E. et alli (org.) Planejamento criativo: novos desafios teóricos em políticas de saúde, Rio de Janeiro:Relume-Dumará (189 pp). PORTO, S.M. (1993) A organização e a gestão do hospital: a gestão financeira, Gestão hospitalar: um desafio para o hospital brasileiro, France:Editions École Nationale de la Santé Publique. PORTO, S.M. (1997) Equidade na distribuição geográfica dos recursos em saúde uma contribuição para o caso brasileiro, Ph. D. Thesis, National School of Public Health, Oswaldo Cruz Foundation, Rio de Janeiro, October, RAMMINGER, I.M.S. (1997) O Orçamento participativo de Porto Alegre: uma experiência local de participação popular e controle social. Master Degree Thesis, National School of Public Health, Oswaldo Cruz Foundation, July, REZENDE, F. (1992) O financiamento da saúde no marco das propostas de reforma do Estado e do sistema tributário brasileiro, Série Economia e Financiamento, 3:12, Brasilia:PAHO Publication. SEBRAE. O Crescimento da Economia Informal. desemprego/economia_informal.html. 03/08/1999. SIMÕES, C.C.S.S. (1992) Aspectos metodológicos das estimativas de mortalidade infantil (mimeo). TRAVASSOS, C.; VIACAVA, F., FERNANDES, C. & ALMEIDA, C. (2000), Desigualdades Geográficas e Sociais na Utilização de Serviços de Saúde no Brasil (Geographical and Social Inequalities in the utilization of Health Services in Brazil). Ciência e Saúde 4(3). UN - United Nations (1999). Internet Site: July/10/1999. VIACAVA, F. & BAHIA, L. (1996) Assistência médico-sanitária, Dados, Radis No. 20, Rio de Janeiro:Oswaldo Cruz Foundation.