Regional Inequalities in Infant Mortality and Primary Health Care in the State of Minas Gerais, Brazil

Espace populations sociétés Space populations societies 2014/2-3 | 2014 PopulationsetterritoiresduBrésil Regional Inequalities in Infant Mortality and Primary Health Care in the State of Minas Gerais, Brazil Les inégalités régionales de la mortalité infantile et des Soins de Santé Primaires dans l'État du Minas Gerais, au Brésil RivaldoMaurodeFariaandPaulaSantana Electronicversion URL: https://journals.openedition.org/eps/5852 DOI: 10.4000/eps.5852 ISSN: 2104-3752 Publisher Université des Sciences et Technologies de Lille Brought to you by Faculdade de Letras da Universidade de Coimbra Electronicreference Rivaldo Mauro de Faria and Paula Santana, “Regional Inequalities in Infant Mortality and Primary Health Care in the State of Minas Gerais, Brazil”, Espace populations sociétés [Online], 2014/2-3 | 2014, Online since 23 April 2015, connection on 06 October 2022. URL: http://journals.openedition.org/eps/ 5852 ; DOI: https://doi.org/10.4000/eps.5852 This text was automatically generated on 3 April 2021. Creative Commons - Attribution-NonCommercial-NoDerivatives 4.0 International - CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/ Regional Inequalities in Infant Mortality and Primary Health Care in the State of Minas Gerais, Brazil Les inégalités régionales de la mortalité infantile et des Soins de Santé Primaires dans l'État du Minas Gerais, au Brésil RivaldoMaurodeFariaandPaulaSantana 1.Introduction 1The decline in infant mortality, and the concomitant reduction in territorial inequalities, is one of the results highlighted by the Brazilian government in the sphere of the national policy for expanding access to Primary Health Care (PHC) under the Family Health Teams (FHT) model [Brazil, 2012]. Infant mortality is known to be very sensitive to PHC [Frankenberg, 1995; Shi et al., 1999; Starfield, 1992; Santana, 2005]. Indeed, the main causes of infant mortality in the world, such as diarrhoea and pneumonia [Lawn, Cousens & Zupan, 2005; Lopez et al., 2006], could in fact be prevented with appropriate medical care. However, there is perhaps no other health indicator that is so dependent upon poverty-related social conditions, thereby reflecting social and geographic inequalities [Goldani et al., 2001; Navarro & Shi, 2001; Szwarcwald, Andrade & Bastos, 2002; Marmot, 2005; Santana, 2005]. This explains the spatial concentrations of high infant mortality rates in regions of the world where the populations are poorer and have less access to healthcare [Lopez et al., 2006]. 2PHC is defined in Brazil as a combination of actions of both an individual and collective nature, including particularly the promotion of health and prevention of disease [Brazil, 2012]. Since the advent of the Single Healthcare System (SHS), which was celebrated in the Brazilian Constitution as free and universal, the right of all and duty of the state [Brazil, 1989], Brazil has made great efforts to expand access to these services, particularly in view of its strategic role for the integration of the healthcare system, and in order to improve health indicators, particularly as regards women and RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 1 children. Hence, the Family Health Programme (FHP) was created in 1994 in response to the need for a healthcare model that was less hospital-oriented and more decentralized, and also to take control of flows in the PHC. 3The FHP has gradually acquired a strategic dimension for health in Brazil, particularly in relation to basic indicators such as the reduction of infant and maternal mortality, increase in vaccination cover and reduction of hospitalizations. To give an idea of its significance, 6.5% of the Brazilian population were covered by the programme in 1998, but this figure had gone up to 40% by 2004 and to around 60% by 2012 (SIAB, 2012). Therefore, the National Basic Healthcare Policy (NBHP), which is the document that determines the way PHC is organized in Brazil, no longer considered to be a programme (because a programme is understood as a package of short-term solutions), but rather as a strategy for the consolidation of PHC through the implantation of Family Health Teams (FHT) [Brazil, 2012]. 4Therefore, the family is the smallest geographical unit with which PHC facilities work in Brazil. But each team, generally consisting of a doctor (general practitioner), nurse, technicians, ancillaries and community health agents, have to be responsible for the health of a geographically defined group of families. Problem-solving is one of its most important principles [Brazil, 2012]. Thus, according to Mendes (2009), these teams should be equipped to attend to and overcome most of the population’s health problems. 5Reducing regional inequalities in infant mortality was one of the Brazilian government’s main objectives in developing FHTs [Brazil, 2012]. This is very important because, despite the fact that the infant mortality rate has been decreasing over the last twenty years in Brazil (from 33.9‰ in 1994 to 13.6‰ in 2011) [SIAB, 2012], there are still enormous regional inequalities in this indicator, with very high levels in the more deprived areas of the North and Northeast of the country [Victora et al., 2011; Faria, 2013). But also on the interregional and even intraurban scales, there are considerable variations in the infant mortality rates. That is the case with the state of Minas Gerais, whose location at the intersection between the Northeast, Southeast and Centre-West regions of the country, combined with historical processes of occupation and the different forms of economic and productive life, has enormous social and regional inequalities and consequently enormous variations in infant mortality rates. 6The state of Minas Gerais is the fourth largest in Brazil in territorial terms, with almost twenty million inhabitants, distributed across 853 municipalities and 66 administrative microregions [IBGE, 2010]. This state also has some specific public health policies, including in the domain of PHC. In fact, the government of Minas Gerais, through the State Secretary for Health, and in association with policies already created by the Ministry of Health, has given broad technical and financial support to the expansion of FHT coverage throughout the state. Through its structuring project “Health at Home” [SESMG, 2005], which cuts across other public health policies, the government achieved an FHT coverage of almost 72% of the population in 2012, which is above the average for the country as a whole. 7However, there have been very few studies that assess the impact of this increased PHC coverage on population health indicators, particularly on infant mortality. It is worth mentioning the work of Raselha et al. (2013), specifically as regards the relationship between the increased PHC cover and the reduction in infant mortality. Although its main theme is rent transfer programmes, the authors show how those programmes, RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 2 associated to greater public health cover, are important to improve population health indicators. From a different perspective, though no less important, Lourenço et al. (2014) also show significant positive correlations between the increase in PHC and the reduction in infant mortality in the state of São Paulo. Although these studies have made an enormous contribution, there are still numerous questions to be discussed on the subject, as regards both services and social inequalities. Hence, this study aims to analyse the regional impacts of PHC on the infant mortality rates (IMRs) in the state of Minas Gerais, taking into account specific socioeconomic contexts. Infant mortality was analysed for the periods 2002-2006 and 2007-2011 in the 66 microregions of the state, with relation to three operative dimensions: the indices of families receiving PHC assistance; indexes of prenatal care; the Index of Sociomaterial Deprivation (ISD). 2.Materialsandmethods 8The study was implemented on the level of the 66 administrative microregions of Minas Gerais for the years 2002 to 2011, grouped into two periods (2002-2006 and 2007-2011) in the case of infant mortality and prenatal care, and at intervals in the case of the FHT coverage and social deprivation indicators (2002 and 2011). The data were collected from secondary official sources, namely the Brazilian Institute of Geography and Statistics (IBGE), Live Births Information System (SINASC), the Mortality Information System (SIM) and the Basic Healthcare Information System (SIAB). The following indicators were constructed from the database in the light of the desired aims: infant mortality rate (IMR), clustered for the years 2002-2006 and 2007-2011; index of population registered with the FHTs, calculated for the years 2002, 2006 and 2011; index of women that did not have prenatal care during pregnancy; Index of Sociomaterial Deprivation (ISD) in 2002 and 2011. 9The ISD was constructed in accordance with the methodology suggested by Cartars & Morris (1991) (cited by Santana, 2005), involving the strategic selection of three social indicators: illiteracy amongst women of reproductive age; domiciles without indoor bathrooms; low-wage families (i.e. those earning less than half the minimum wage). 10Following Santana (2005), those variables were standardized using the z-score method so each would have the same influence on the final result of the index. The ISD thus represents the sum of those standardized variables. 11Pearson’s correlation coefficient was also used to calculate the interdependence of the variables of the study and their ratios, particularly as regards regional dynamics. 12After the construction and modelling of the indicators in accordance with the previously defined spatial and temporal scales, a thematic cartography was constructed using the Geographic Information System (GIS). For this, the IBGE cartographic base was used at a scale of 1:50,000, and the ArcGis Programme from the company Esri. i. ii. iii. i. ii. iii. iv. i. ii. iii. RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 3 3.Results 3.1.Infantmortality:temporalvariationandspatialdistribution 13As in the rest of the Brazil, the state of Minas Gerais, influenced by the policies of the country, has seen a reduction in infant mortality. In 2002-2006 the rate was 18.0‰ which declined to 13.1‰ in 2007-2011. It is, therefore, lower than the Brazilian average (13.6‰ in 2011). There was also a drop in infant mortality in the microregions generally (Figure 1) over the period, though this did not occur in all microregions. There are still enormous spatial inequalities in infant mortality in Minas Gerais, as some microregions have low rates while others are higher than the Millennium Development Goal for infant health [WHO, 2005] (the 4th MDG determines the reduction of the IMR in Brazil to 15.7‰ by 2015). 14Figure 1 shows the spatial concentrations of high IMR in the microregions in the Northeast, and also, though with lower values, in the microregions of the “Forest Zone” and Centre. In contrast, the microregions located between the Northwest and the “Mining Triangle”, and between Belo Horizonte and South have lower IMRs in the two periods, though more markedly between 2007-2011. Special attention should be given to the microregions located in the extreme Northeast: four microregions had a very high IMR, and of those, two actually worsened, while the other two experienced only a very slight reduction. 15As Table 1 shows, there was spatial variation in IMRs between microregions in accordance with population size. The ten most populated microregions, where almost half the population of the state is concentrated and where the largest cities are located, including the capital Belo Horizonte, are those with the lowest IMR. On the other hand, the ten least populated regions, which account for only 3.7% of the population of the state and which have only small cities, have higher IMRs on average, located mostly in the Northeast of the state. It should be pointed out that the IMRs indicated in the Table 1 are the average calculated between the ten most populated and the ten least populated microregions. The others are grouped separately. It can be seen that the ten most populated microregions have IMRs that are close to the state average, accompanying the general trend towards reduction seen not only in Minas Gerais but also in Brazil as a whole. The same cannot be said of the ten least populated microregions. RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 4 Figure 1. Infant mortality rate (IMR) by regions of the State of Minas Gerais, Brazil, for the years 2002-2006 and 2007-2012. Table1.ComparisonofIMR,livebirthswithoutprenatalcare,ISD,andFHTcoveragefortheten mostandleastpopulatedmicroregionsandothersinthestateofMinasGerais,Brazil IMR (‰)* LBWPC (‰)** ISD*** FHTs (%)*** Population (%)*** Urbanization Index*** 10 + populated 13.6 10.4 -1.7 63.0 48.9 89.9 10 – populated 16.6 10.6 2.4 93.9 3.7 65.6 Others 15.01 9.96 -0.16 80.4 47.4 77.9 *Referring to the aggregate period 2007-2011 ** Live births without antenatal care in the aggregate period 2007-2011 ***Referring to the year 2011 Source: Datasus, Ministry of Health, 2013 (constructed and adapted by the authors). 16Table 1 also shows variations in IMR in accordance with the urbanization rate in the most and least populated microregions. The former are the most urbanized as they include the biggest cities of Minas Gerais (as mentioned above), including its capital, Belo Horizonte. The opposite occurs with the least populated microregions. The highest IMRs were found in the least urbanized microregions and diminished as the urbanization index increased. In fact, the ten most populated and most urbanized microregions have IMRs similar to the state average, and some are even lower than that average. For example, the two most populated microregions in the state, Belo Horizonte and Uberlândia (Mining Triangle), have IMRs of 11.6‰ and 11.8‰ respectively. Therefore, with regard to the aggregated period of 2007-2011, their rates are lower than the state average. On the other hand, the least populated and least urbanized microregions have IMRs well above the state average. RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 5 17Nevertheless, it is relevant to analyse these rates in relation to other indicators, particularly the indicators of PHC cover and usage. This is also important to clarify a discourse that has already been presented by some authors [e.g. Sastry, 1997; Souza et al., 1999] according to which urbanization is associated to a reduction in infant mortality. 3.2.Expansionofprimaryhealthcareprovisionusetoantenatalcare 18With the FHT model, there was an increase in PHC coverage throughout all the microregions of Minas Gerais (Figure 2). The state average, which was 50% in 2002, went up to 64% in 2006 and 71% in 2011. The microregions located in the North, Northeast, Centre and “Forest Zone” presented higher levels of coverage in the three years analysed, while the opposite happened in the microregions located in the South, “Mining Triangle”, Belo Horizonte and Northwest, where the FHT coverage was lower. In fact, the first group had levels of coverage that were higher than the state average, while the second was generally lower than the state average, despite the increase in the supply in the period analysed. 19In any case, the data indicate the enormous advances in FHT coverage in Minas Gerais. For many microregions, particularly those of the North and Northeast of the state, provision was universal by 2011. Considering only the microregions with a provision of over 80%, there is a total of 35 microregions, which represents over half. It can also be seen that the most populated and urbanized microregions (Table 1), where the major cities in Minas Gerais are located such as Uberlândia (“Mining Triangle”), Juiz de Fora (“Forest Zone”), Belo Horizonte (Metropolitan Region), Poços de Caldas, Pouso Alegre and Varginha (South) and Montes Claros (North), are those with the lowest levels of coverage, averaging 63% of the population. In contrast, in the less populated and urbanized microregions, where the small cities are located, the level of coverage exceeds 93%. It should be mentioned that the expansion of services in the less populated microregions was one of the goals of the National Basic Healthcare Policy [Brazil, 2012]. Thus, in accordance with the principle of equity, the microregions that are less dense in public health services and consequently have greater health needs should have a greater supply of services per inhabitant. RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 6 Figure 2. Coverage of Family Health Teams (in percentage of population) by regions of the State of Minas Gerais, Brazil, in 2002 (A), 2006 (B) and 2011 (C) 20However, if we compare the way services are distributed (percentages per population), as in Figure 2, there is clearly a tendency for a homogenization of supply; by 2011, there was a predominance of microregions with over 61% provision, and a reduction of microregions with less than this. In fact, in 2002, there were 44 microregions (accounting for 66.9% of state’s population) with less than 60% provision, while in 2011 this figure had dropped to 9 microregions (19.9% of the population). Similarly, in 2002 there was 80% coverage in only 5 microregions (5.3% of the population) and by 2011, it existed in 35 microregions (over 28%). 21Thus, it can be concluded that the policy of expanding PHC through the establishment of FHTs has yielded numerically important results, in keeping with the policies produced for the sector. This is despite the fact that the National Basic Healthcare Policy [Brazil, 2012] does not stipulate the figure to be achieved, but merely aims for an expansion of coverage in the least populated/most deprived areas. 22It should be added, however, that the existence of services, or rather the expansion of provision, does not mean that these health services are properly accessed and used, nor does it automatically produce good health outcomes. It is necessary to analyse the impact of the services upon the population’s living conditions and health. In this case, it is worth reflecting on the population’s access to and use of services, and on health outcomes, particularly for infant mortality. There are numerous indicators of access/ use of primary health services by the population. In relation to child health and impacts upon infant mortality rates, prenatal care is important, given its potential for the prevention of disease and infant mortality through the promotion of maternal health. Therefore it is to be hoped that the increased FHT coverage in Minas Gerais will have had a positive impact by providing greater access to prenatal care, and consequently helping reduce infant mortality. RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 7 23In accordance with the World Health Organization [WHO, 2002], there should be at least six prenatal consultations during pregnancy. The Brazilian Ministry of Health has recommended that these should take place every four weeks, and every 15 days after the 36th week of pregnancy [Brazil, 2006] As this is an important indicator of maternal and child health, the Ministry of Health itself monitors the number of prenatal consultations that have taken place during pregnancy in relation to the numbers of live births. However, the official data do not distinguish between consultations that have taken place in the public and private healthcare systems. Thus, only data relative to live births without any prenatal care during pregnancy have been collected. In other words, what is assessed is the vulnerability of women and foetuses when prenatal consultations do not take place. This is shown in Figure 3 for the years 2002-2006 and 2007-2011. Figure 3. Live births without prenatal care (in ‰) between the years 2002-2006 and 2007-2012 by regions of the State of Minas Gerais, Brazil 24Figure 3 shows that there was a reduction in the number of live births without prenatal care in almost all the microregions of Minas Gerais. The state average, which was 14.4‰ in 2002-2006, dropped to 10.1‰ in 2007-2012, a reduction of almost 30%. These values are below the national average of 36.7‰ in 2002-2006 and 26.9‰ in 2007-2012 (a reduction of 26.5%). Therefore, not only is the average in the state of Minas Gerais better than the country as a whole, its performance was also better in this indicator. However, the indexes of live births without prenatal care are still high in some microregions, compared with the state and national averages. Moreover, the spatial concentration of these indices is visible in microregions located between the Centre and Northeast of the state. In fact, though there has been a reduction, there continue to be spatial concentrations of higher rates of live births without prenatal care over the two periods. RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 8 NAVARRO V., SHI L. (2001), The political context of social inequalities and health, Social Science & Medicine, vol. 52, nº 3, pp. 481-491. RASELLA D. et al (2013), Effect of a conditional cash transfer programme on childhood mortality: a nationwide analysis of Brazilian municipalities, The Lancet, vol. 382, nº 9886, pp. 57-64. RAVALLION M. (2009), Comparative Perspective on Poverty Reduction in Brazil, China and India, Washington, DC, World Bank. SANTANA P. (1995), Acessibilidade e Utilização dos Cuidados de Saúde. Ensaio Metodológico em Geografia da Saúde, Lisboa, ARSC/CCDRC. SANTANA P. (2005), Geografia da Saúde e do Desenvolvimento: Evolução e Tendências em Portugal, Lisboa, Almedina,. SASTRY N. (1997), What explains rural-urban differentials in child mortality in Brazil? Social Science & Medicine, Vol. 44, n° 7, pp. 989–1002. SESMG – Secretaria de Estado da Saúde de Minas Gerais. (2005), Resolução SES Nº 661, de 22 de março de 2005, disponível em: www.saude.mg.gov.br, acessado em: dezembro de 2012. SHI L. et al. (1999), Income inequality, primary care, and health indicators, Journal Family Practice, vol. 48, nº 4, pp. 275-284. SIAB - Sistema de Informação da Atenção Básica. (2012), Cadastramento Familiar, Brasília. SOUZA A. C. Terra de (1999), Variations in infant mortality rates among municipalities in the state of Ceará, Northeast Brazil: an ecological analysis. International Journal of Epidemiology, vol. 28, nº 2, pp. 267-275. STARFIELD B. (2004), The Medical Home, Access to Care, and Insurance: A Review of Evidence, Pediatrics, vol. 113, nº 4, pp. 1493-1498. STARFIELD B. (1992), Primary care: concept, evaluation, and policy, New York, Oxford University Press. SZWARCWALD C. L., ANDRADE C. L. T., BASTOS F. I. (2002), Income inequality, residential poverty clustering and infant mortality: a study in Rio de Janeiro, Brazil, Social Science Medicine http:// www.sciencedirect.com/science/journal/02779536, vol. 55, nº 12, pp. 2083-2092. VERDOM J., CAO H. (2011), Health Care Access and Regional Disparities in China, Espace populations Sociétés, vol. 2011, n° 1, pp. 63-78. VICTORA C. G, et al. (2011), Saúde de mães e crianças no Brazil: progressos e desafios, The Lancet, vol. 377, nº 977, pp.32-46. WHO - World Health Organization (2002), WHO antenatal care randomized trial: manual for the implementation of the new model, Geneva, WHO. WHO - World Health Organization (2005), Health and the Millennium Development Goals, Geneva, WHO. WONG C. (2006), Indicators for urban and regional planning, London, Routledge. ABSTRACTS Infant mortality has declined in Brazil in recent years, partly as a result of increased Primary Health Care (PHC) coverage and better access to it. However, there are still enormous regional RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 15 inequalities in these indicators, and therefore spatial variations in infant mortality. In a country like Brazil that is of continental proportions, it is important to analyse/assess health policies in the context of regional inequalities in order to ensure equity. This study thus aims to analyse the regional impacts of PHC and assess the importance of socioeconomic contexts upon infant mortality rates (IMRs) in the state of Minas Gerais. This is the fourth biggest state in Brazil, with a population of almost 20 million, distributed over 66 microregions. The IMR was assessed for the periods 2002-2006 and 2007-2011 in the 66 microregions, considering three operative dimensions: i. indexes of families aided by PHC; ii. indexes of prenatal care; iii. and the Index of Sociomaterial Deprivation (ISD). The data were collected from secondary official sources, and modelled on temporal and spatial scales defined and worked in a Geographical Information System. The ISD, which is a composite indicator used to express social and regional inequalities, was constructed on the basis of three indicators: i) illiteracy; ii) wages; and iii) homes without indoor bathrooms. The results showed a reduction in IMR in almost all the microregions of Minas Gerais between 2002-2006 and 2007-2011. However, there were inequalities in this decrease, with a continuing concentration of mortality in microregions with higher ISD. The expansion of PHC is more marked in the microregions with higher ISD, which is a positive sign from the Brazilian Health System. However, the expansion of these services as an isolated sectorial policy, though important, has not on its own been able to reduce the geographical inequalities in infant mortality in Minas Gerais. Hence, it is necessary to consider the social and economic contexts of families when attempting to improve indicators that go beyond mere healthcare intervention. In other words, infant health is the result of integrated public policies (eg. housing, employment/wages, education/training) and multi-level decisions and actions (local, municipal, regional, federal). La mortalité infantile a diminué au Brésil au cours des dernières années, grâce, entre autres, à l'extension de la couverture et l'accès aux soins de santé primaires (SSP). Cependant, d'énormes inégalités régionales persistent en ce qui concerne ces deux derniers indicateurs qui ont un impact sur les variations spatiales de la mortalité infantile. Dans les pays de la taille d’un continent comme le Brésil, il est pertinent d'analyser / d’évaluer les politiques de santé dans le contexte des inégalités régionales, dans le but d’assurer plus d’équité. Ceci a motivé la présente étude, dont l'objectif est d'analyser les impacts régionaux du SSP, en incluant également l'analyse des contextes socio-économiques, sur le taux de mortalité infantile (TMI) dans l'État du Minas Gerais, au Brésil. Cet état est le quatrième plus grand État du Brésil, sa population s’élève à près de 20 millions d'habitants, répartis en 66 microrégions. L'analyse de TMI a été réalisée pour les périodes 2002-2006 et 2007-2011 dans chacune de ces microrégions, en relation avec trois dimensions opérationnelles de l'analyse: i. le taux de familles assistées par SSP, ii. les taux de couverture des soins prénataux iii. et l’indice de défavorisation matérielle et sociale (IDMS). Les données ont été recueillies à partir de sources secondaires officielles, adaptées à des échelles spatiales et temporelles définies et travaillées avec un Système d’Information Géographique. L’indice de défavorisation matérielle et sociale, indicateur composite utilisé pour exprimer les inégalités sociales et régionales, a été construit sur la base de trois indicateurs i) l'analphabétisme, ii) les revenus et iii) le nombre de ménages sans installations sanitaires au sein de leur foyer. Les résultats ont montré une baisse du taux de mortalité infantile dans presque toutes les microrégions du Minas Gerais, entre 2002-2006 et 2007-2011. Toutefois, on constate des inégalités dans cette baisse avec une mortalité plus forte dans les régions ayant un IDMS plus élevé. L’expansion des SSP est plus importante dans les régions qui possèdent un IDMS plus élevé, ce qui est un signal positif pour le Système de Santé brésilien. Cependant, l’expansion des SSP en RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 16 tant que politique sectorielle isolée, malgré son importance, n’est pas suffisante pour réduire les inégalités spatiales de la mortalité infantile à l’intérieur du Minas Gerais. Pour cela, il est nécessaire de prendre en compte le contexte social et économique des familles quand l’objectif est d’améliorer les indicateurs au-delà d’une simple intervention sanitaire. Pour le dire autrement, la santé de l’enfant est la résultante des politiques publiques intégrées (exemples : logement, emploi/revenu, éducation/instruction,) et des décisions et actions à différents niveaux (local, municipal, régional, fédéral). INDEX Keywords: primary healthcare, regional inequality, infant mortality Mots-clés: soin de santé primaire, inégalité régionale, mortalité infantile AUTHORS RIVALDOMAURODEFARIA Institute of Geosciences State University of Campinas Unicamp, Brazil University Federal of Santa Maria, Brazil [email protected]. PAULASANTANA Centre for Studies into Geography and Territorial Planning Department of Geography University of Coimbra, Portugal [email protected]. RegionalInequalitiesinInfantMortalityandPrimaryHealthCareintheStat... Espacepopulationssociétés,2014/2-3|2014 17