INTRODUÇÃO: A vacina pneumocócica conjugada 10-valente (VPC10) foi introduzida no calendário de imunização infantil do Programa Nacional de Imunizações em 2010. Este estudo analisou as taxas de incidência de Meningite Pneumocócica (MP) no Brasil, por faixa etária e unidade da federação (UF); a distribuição espacial das taxas de incidência de MP em menores de cinco anos no Brasil, por UF, no período pré (2005-2009) e pós-vacinação (2011-2013); e associações com variáveis socioeconômicas e cobertura vacinal. Foram analisadas a distribuição espacial das taxas de incidência de MP em menores de cinco anos, por município do estado de São Paulo (SP), no período pré e pós-vacinação, e a existência de aglomerados espaciais e espaço-temporais. Através de estatística espacial, foram analisadas associações das taxas de incidência de MP, por microrregiões do estado, com variáveis socioeconômicas e cobertura vacinal. MÉTODOS: Estudo ecológico de base populacional, que utilizou dados do Sistema de Informação de Agravos de Notificação. Cobertura vacinal e o Índice de Desenvolvimento Humano (IDH) foram utilizados na análise do Brasil. Na análise de SP, as unidades ecológicas foram municípios e microrregiões, e a variável socioeconômica foi o Índice Paulista de Responsabilidade Social (IPRS) da Fundação Sistema Estadual de Análise de Dados. Foram construídos mapas temáticos para as taxas de incidência de MP em menores de cinco anos, nos períodos pré e pós vacinação, cobertura vacinal e IDH, por UF. Também foram construídos mapas temáticos das taxas de incidência de MP em menores de cinco anos, por microrregião de SP, nos períodos pré e pós-vacinação, cobertura vacinal e IPRS, utilizando o software QGis 2.6.1. Para SP, foi utilizada a técnica de varredura (software SatScan 9.2) para analisar aglomerados. O modelo Gaussiano latente Bayesiano com modelos inflados de zeros de Poisson, através da aproximação de Laplace aninhada e integrada (INLA), foi utilizado na análise espacial para avaliar associações entre taxa de incidência de MP, cobertura vacinal e variáveis socioeconômicas. RESULTADOS: De 2005 a 2013, foram notificados 10.769 casos de MP. Crianças menores de cinco anos foram as mais acometidas. No período pós-vacinação (2011-2013), a taxa de incidência de MP diminuiu nos menores de cinco anos, especialmente nos menores de um ano (de 10,42/100.000, em 2005, para 4,13/100.000, em 2013). No Brasil, maiores taxas de incidência de MP ocorreram nos estados com maior IDH. Em SP ocorreu o mesmo, sendo encontrados, no período pré-vacinação, dois aglomerados de municípios - um de baixo risco para MP, no noroeste do estado (RR, 0,45, p=0,00025), e outro de alto risco no sudeste, englobando a capital do estado, (RR, 1,62, p=0,0000051). No período pós-vacinação, apenas um aglomerado de maior risco se manteve na mesma região (RR, 1,97, p=0,057). Na análise Bayesiana, riqueza foi identificada como fator de risco para MP (RR, 1,026, IC: 1,002-1,052) no período pré-vacinação. Cobertura vacinal, longevidade e escolaridade não foram significativas. CONCLUSÕES: Maior IDH e maior riqueza foram fatores de risco para MP, sugerindo necessidade de maior investimento na capacidade diagnóstica de MP nas áreas estudadas, avanços na qualificação da vigilância e notificação da doença / INTRODUCTION: The 10-Valent pneumococcal conjugate vaccine (PCV10) was introduced into the childhood immunization schedule of the Brazilian National Immunization Program in 2010. This study analyzed Pneumococcal Meningitis (PM) incidence rates in Brazil, by age group and federation unit (FU), the spatial distribution of PM incidence rates in under-5 children in Brazil, by FU, in the pre (2005-2009) and post-vaccination (2011-2013) periods, and associations with socioeconomic variables and vaccination coverage. We conducted spatial analysis of PM incidence rates in under-5 children, by municipality in SP, in pre and post-vaccination periods, and evaluated the existence of spatial and spatial-temporal clusters. Spatial statistics was used to test associations of PM incidence rates with socioeconomic variables and vaccine coverage, by state micro regions. METHODS: This is a population-based ecological study using data from the Sistema de Informação de Agravos de Notificação. Vaccine coverage and the Human Development Index (HDI) were used in the Brazilian analysis. In SP analysis, the ecological units were municipalities and micro regions, and the socio-economic variable was the Índice Paulista de Responsabilidade Social (IPRS) of the Fundação Sistema Estadual de Análise de Dados. Thematic maps were built for PM incidence rates in under-5 children, in the pre- and post-vaccination periods, vaccine coverage and HDI, by FU. Thematic maps were also built for PM incidence rates in under-5 children by SP micro regions, in the pre- and post-vaccination periods, vaccination coverage and IPRS using QGis 2.6.1 software. The scanning technique (SatScan 9.2 software) was used to analyze spatial and spatiotemporal clusters in SP. A Bayesian latent Gaussian model with zero-inflated Poisson model through the integrated nested Laplace approximation (INLA) was used in the spatial analysis to evaluate associations of PM incidence rates with vaccine coverage and socioeconomic variables. RESULTS: From 2005 to 2013, 10,769 PM cases were reported. Under-5 children were the most affected in the whole period. In the post-vaccination period (2011-2013), PM incidence rates decreased among under-5 children, especially among infants (from 10.42/100,000, in 2005, to 4.13/100,000, in 2013). Higher PM incidence rates occurred in states with higher HDI. The same occurred in SP, where two municipalities clusters were found in the pre-vaccination period - one of low risk for PM in the northwest of the state (OR, 0.45, p=0.00025), and another of high risk in the southeast, including the state capital (OR, 1.62, p=0.0000051). In the post-vaccination period, only one cluster of higher risk remained in the same area (RR, 1.97, p=0.057). In Bayesian analysis, wealth was identified as a risk factor for PM (RR, 1.026, CI: 1.002-1.052). Vaccination coverage, longevity and education were not important. CONCLUSIONS: A higher HDI as well as greater wealth were risk factors for PM. This result highlights the need to improve the diagnostic capacity of PM in studied areas, advancing in the surveillance quality and disease notification

L’Évaluation d’Impact sur la Santé (EIS) est un instrument de support à la décision, pour juger une politique quant aux effets potentiels sur la santé et leur distribution (équité). C’est encore souvent une approche qualitative.L’objectif principal est de montrer l’utilité de méthodologies statistiques quantitatives multivariées pour enrichir la pratique d’EIS, améliorant la compréhension des résultats par des professionnels non-statisticiens.Les futures réformes des systèmes de santé déplacent le centre d’évaluation des services de santé des fournisseurs aux citoyens (besoins, préférences, équité d’accès aux gains de santé), exploitant big data associant information de soins aux données sociales, économiques et de déterminants de santé. Des méthodologies statistiques et d’évaluation innovantes sont nécessaires à cette transformation.Les méthodes de data mining et data science, souvent complexes, peuvent gérer des résultats graphiques compréhensibles pour amplifier l’usage d’EIS, qui deviendrait ainsi un outil précieux d’évaluation de politiques publiques pour amener les citoyens au centre de la prise de décision. / Health Impact Assessment (HIA) is a decision-making support tool to judge a policy as to its potential effects and its distribution on a population’s health (equity). It’s still very often a qualitative approach.The main aim here is to show the usefulness of applying quantified multivariate statistical methodologies to enrich HIA practice, while making the decision-making process easier, by issuing understandable outputs even for non-statisticians.The future of healthcare reforms shifts the center of evaluation of health systems from providers to people’s individual needs and preferences, reducing health inequities in access and health outcomes, using big data linking information from providers to social and economic health determinants. Innovative statistical and assessment methodologies are needed to make this transformation.Data mining and data science methods, however complex, may lead to graphical outputs simple to understand by decision makers. HIA is thus a valuable tool to assure public policies are indeed evaluated while considering health determinants and equity and bringing citizens to the center of the decision-making process. / A Avaliação de Impacte na Saúde (AIS) é um instrumento de suporte à decisão para julgar política quanto aos seus efeitos potenciais e à sua distribuição na saúde de uma população (equidade). É geralmente ainda uma abordagem qualitativa.O principal objetivo é mostrar a utilidade das metodologias estatísticas quantitativas e multivariadas para enriquecer a prática de AIS, melhorando a compreensão dos resultados por profissionais não-estatísticos.As futuras reformas dos sistemas de saúde deslocam o centro da avaliação dos serviços de saúde dos prestadores para as necessidades e preferências dos cidadãos, reduzindo iniquidades no acesso à saúde e ganhos em saúde, usando big data que associam informação de prestadores a dados sociais e económicos de determinantes de saúde. São necessárias metodologias estatísticas e de avaliação inovadoras para esta transformação.Métodos de data mining e data science, mesmo complexos, podem gerar resultados gráficos compreensíveis para os decisores. A AIS é assim uma ferramenta valiosa para avaliar políticas públicas considerando determinantes de saúde, equidade e trazendo os cidadãos para o centro da tomada de decisão.

Addressing complex global health challenges, including the burden of noncommunicable diseases (NCDs), will require change in sectors outside of traditional public health. Contemporary regional trade and investment agreements (RTAs) like the Trans-Pacific Partnership (TPP) continue to move further ‘behind-the-border’ into domestic policy space introducing new challenges in the regulation of health risk factors. This dissertation aimed to clarify the pathways through which RTAs influence NCDs, and to explore points along those pathways with the intent of improving the existing evidence base and supporting policy development. This work develops a critical theoretical framework exploring the ideas, institutions, and interests behind trade and investment policy; it also develops a conceptual framework specifying how trade and investment treaty provisions influence NCD rates through the effects of trade and investment on tobacco, alcohol, and ultra-processed food and beverage products, as well as access to medicines and the social determinants of health. Using health impact assessment methodology, three analytical components were designed to examine pathways of influence from RTAs to health outcomes as mediated by the interests of transnational corporations (TNCs). The first component explored the influence of industry during the TPP negotiations and how its health-related interests were reflected in the final TPP text. The second component examined the role of trade and investment liberalisation in health-harmful commodity markets, finding a rise in TNC sales after a period of liberalisation. The third component demonstrated how investor rights and investor-state dispute can challenge the state’s right to regulate if it damages the profits of TNCs, which may threaten effective health regulation, and provides opportunities to strengthen the right to regulate. The work in this dissertation provides support for the thesis that trade and investment policies are a fundamental structural determinant of health and well-being, which are highly influenced by TNCs that guide such policies in the interest of maximising their profits and protections, often to the detriment of public policy and population health. This work identifies the need for more robust health impact assessments of RTAs before future agreements are ratified, as well as an imperative to challenge vested interests that entrench neoliberal policy preferences that have hindered sustainable and equitable development.

Chemical (gaseous) emissions from wastewater treatment plants (WWTP) are perceived as odour and can generate undesirable health reactions. These could affect the quality of life to those regularly albeit intermittently exposed, although regular exposure may result in tolerance and loss of recognition of the odour. Consequently, research is necessary to establish the role of Environmental Impact Assessment (EIA) in ensuring that facilities such as WWTP do not pose a threat to local communities, particularly in relation to housing developments for poor communities. The aim of this study was to investigate if the EIA regulations were effectively implemented before building of the Residential Development Proiect (RDP) houses at KwaMathukuza, in Newcastle South Africa. Furthermore, the study intended to determine the impact the (WWTP) has on the health of people who reside close to the site and to establish if the families staying close to the WWTP have any common ailments such as handicapped babies, early deaths within families. Findings would also indicate if RDP houses should be built in close proximity to such sites. Data was collected through a survey of 85 residents living within five kilometers and ten residing beyond five kilometers from the WWTP. The participants were selected randomly. Municipality employees who worked with the WWTP were also sampled together with healthcare givers. Data included information about health of the community, their perception of the gases released from the WWTP as well as EIA meetings and processes related to the construction of the RDP houses. The outcome of this study revealed that EIA regulations were not adequately implemented and that this residential site was unsuitable for the construction of RDP houses. It emerged that the residence regularly experience unpleasant odours released from the WWTP. These could be associated with ill-health of the residents as well as v degradation the environment in general. The study also found that the effect of the WWTP affects even those communities who reside beyond five kilometers from the WWTP. The researcher argues for more stringent mechanisms to ensure that EIA regulations are properly implemented to prevent adverse effects on the communities. / Environmental Sciences / M. Sc. (Environmental Management)

Made available in DSpace on 2011-05-04T12:42:06Z (GMT). No. of bitstreams: 0 Previous issue date: 2010 / Este trabalho foi concebido a partir da reflexão e discussão sobre as principais oportunidades para a abordagem integrada de Saúde Pública e Meio Ambiente, considerando o atual cenário energético mundial, a introdução de novas tecnologias de produção e a necessidade de construção de uma referência conceitual e prática que considere a interação entre Saúde, Meio Ambiente, Responsabilidade Social das Empresas e Desenvolvimento Sustentável. Considerando os inúmeros desafios para o controle dos impactos ambientais e de saúde, através da gestão efetiva de responsabilidade social de empresas potencialmente poluidores, particularmente da indústria de petróleo e energia, essa tese tem como principal objetivo discutir como os aspectos de saúde podem ser incorporados ao licenciamento ambiental do setor petróleo. Para isso foi utilizada uma abordagem metodológica exploratória, por meio de exaustivo levantamento bibliográfico na literatura científica e institucional deáreas empresariais e governamentais responsáveis pela obtenção e concessão de licenças ambientais. A partir da seleção e análise de documentação oficial, particularmente dos estudos e relatórios de impacto ambiental (EIA/RIMA) de empreendimentos nacionais de exploração, produção e refino do petróleo, avaliousea existência de evidências e oportunidades para a inserção de questões de saúde nos respectivos projetos de investimento desse setor. Como ferramenta de avaliação dessas questões nos EIA/RIMA dos empreendimentos selecionados, foi utilizada uma adaptação de uma matriz de análise validada para o setor hidrelétrico, cujas categorias analíticas foram baseadas nos parâmetros estabelecidos para os estudos de avaliação de impacto ambiental pela Resolução CONAMA 1/86 e nos principais aspectos da metodologia de Avaliação de Impactos à Saúde (AIS). Como resultado, a matriz empregada mostrou-se um instrumento metodológico com grande potencial para orientar a incorporação dos aspectos desaúde no processo de licenciamento ambiental, instrumentalizando a Avaliação de Impacto Ambiental (AIA) e subsidiando a tomada de decisão, contribuindo, portanto, para uma gestão socioambiental mais proativa do setor petróleo. / This work was conceived from the discussion about the main opportunities for the integrated approach of Public Health and Environment, considering the current world energy scenario, the introduction of new production technologies and the need for a more comprehensive conceptual reference among Health, Environment, Social Responsibility and Sustainable Development. Considering the several challenges for effective management of the environmental and health impacts, particularly caused by oil industry, strategic for the development and economical growth of our country, the main objective of this thesis is discuss how the health aspects could be incorporated to the environmental licensing of this sector. An exploratory methodological approach was used, starting from an exhausting bibliographical research in scientific and institutional literature, produced by companies and government areas related with the environmental licenses. The health evidences and opportunities in the main official documents, particularly studies and reports of the environmental impacts (EIA/RIMA, in portuguese) of national enterprises of oil exploration, production and refining, was evaluated. An adapted matrix, validated for the hydroelectric sector, was used like a tool for assessment of health subjects in the selected RIMA. The analytical categories of this matrix were based on the established parameters for the studies of environmental impact assessment of CONAMA Resolution 1/86, and on the main aspects of Health Impact Assessment (HIA) methodology. As result, this matrix was considered like an important instrument with great potential to guide the incorporation of the health aspects in the environmental licensing, supporting the Environmental Impact Assessment (EIA) and decision makers, contributing to a proactive social and environmental management of the petroleum industry.

Made available in DSpace on 2011-05-04T12:42:06Z (GMT). No. of bitstreams: 0 Previous issue date: 2009 / A proposta aqui apresentada considera a informação gerada pelo setor saúde como resultante da relação entre os processos de destruição e desgaste a que está submetida a população, em seus diferentes estratos, e os processos de resistência e recuperação desenvolvidos pelas mesmas. Dessa forma, a informação gerada, particularmente pelos serviços de saúde, reflete tanto as condições de vida da população como a capacidade de respostas desenvolvidas pela sociedade, naquele momento do seu processo histórico. Para tanto, procura-se descrever a evolução desta informação no tempo e no espaço, procurando identificar os problemas decorrentes de acesso, cobertura e complexidade dos serviços, além de falhas e desvios introduzidos no próprio processo de produção, disseminação e utilização da informação. Este trabalho se organiza em um estudo em três estágios que contribuem para um resultado final com a proposição de uma metodologia ou estratégia de ação. Pretende-se, dessa forma, lançar um debate e contribuir para que os dados do setor saúde possam vir a ser utilizados mais amplamente na construção de indicadores para a gestão do território, tanto por gestores como pela população em geral. O trabalho aqui apresentado envolve o levantamento dos indicadores da relação saúde e ambiente que estão sendo mais utilizados na literatura científica e na definição de prioridades para a gestão do território. Para esse levantamento procurou-se concentrar nas questões que se referem ao saneamento, como sendo as questões mais antigas e gerais da relação saúde-ambiente enfrentadas pela humanidade. Em continuidade, procura-se aplicar uma metodologia, em desenvolvimento, que resgate as informações existentes nos bancos de dados da saúde, buscando identificar situações concretas na população que reside em numa área selecionada, entorno de um ponto, possível foco de contaminação. Esse estudo, realizado em uma cidade do interior do Rio Grande do Norte, onde existe uma planta industrial da PETROBRÁS, inserido em uma pesquisa mais abrangente, funciona como um estudo de caso, onde os dados gerados pelos serviços de saúde puderam ser comparados com outras metodologias de investigação em saúde e ambiente procurando subsidiar a construção de indicadores. Em um segundo estudo de caso, procurou-se aplicar esta mesma estratégia na avaliação de impactos a serem gerados por um projeto de desenvolvimento, na saúde da população, em uma região da Amazônia. Neste estudo também se procurou avaliar o potencial da utilização das informações geradas pelo setor saúde no monitoramento de impactos ambientais, porém em uma região bem mais ampla e de maior diversidade de situações e de estrutura de serviços de saúde. A metodologia empregada parte de um reconhecimento de campo e se utiliza dos dados secundários de saúde disponíveis, que, associados a uma série de entrevistas com informantes-chave e oficinas com gestores, profissionais de saúde e representantes da comunidade, procura traçar os perfis de saúde-doença das localidades, identificando padrões e processos, particularmente, aqueles que poderiam sofrer influência do empreendimento ou processo em questão. Deste estudo, além do perfil de saúde da população, abrangendo a infraestrutura de serviços, acesso e utilização, procurou-se traçar o perfil da própria informação de saúde e seu processo de implantação, desenvolvimento e manutenção dos diferentes bancos de dados. Ao se analisar a informação em saúde num contexto de rápidas transformações e baixa cobertura de serviços, utiliza-se a própria falta do dado, ou baixa qualidade do mesmo, como informação, que, no contexto das formações econômico sociais existentes e no seu processo de desenvolvimento, podem constituir importantes indicadores. Como conclusão, observa-se o potencial da utilização das informações de saúde para o monitoramento e avaliação de impactos de projetos de desenvolvimento e alterações do ambiente. Destaca-se a não utilização das informações, tanto por parte dos setores de planejamento e gestão do território, quanto pelo próprio setor saúde, que se utiliza delas de modo administrativo e burocrático, sem um cuidado com sua qualidade e adequação a usos que, em muito, poderiam auxiliar o planejamento e as ações de saúde, assim como fortalecer o papel da saúde na gestão do território. O adequado tratamento e utilização da informação gerada pelo setor saúde pode promover, a melhoria da qualidade da mesma e o melhor planejamento e utilização dos recursos para a saúde, além de fornecer uma ferramenta de avaliação do processo de desenvolvimento para a sociedade. Esta ferramenta, de baixo custo operacional, permite a introdução de variáveis sobre as condições de vida de diferentes estratos da população na formulação de respostas da sociedade para indagações como: desenvolvimento para que e para quem? / The proposal presented here considers the information generated to the health system as a resultant of the deprived processes to which the population is submitted, in different groups, and the strategies of resistance and recovery developed by them. The generated information, particularly for the health services, reflects the life conditions of the population and the response capacity developed by the society, at that moment of the historical process. The study aims to describe the evolution of the health information over time and space, identifying problems of access, covering and complexity of the services, besides the imperfections and deviation introduced in the own process of production, dissemination and information use. This study is organized in three steps, each of them contributing for a final result proposing a methodology or a strategy of action. There is an intention to launch a debate and to contribute to a widely use of the system health data in the construction of indicators for the management of the territory, touching managers and the population in general as well. The work presented here involves the survey of the indicators related to health and environment that are being used in the scientific literature and in the definition of priorities for the management of the territory. To do this research it was necessary to concentrate on questions related to the sanitation condition, defined as the oldest general questions of the relation health/ environment faced by the humanity. A methodology has been developed, which rescued the existing information in health databases, searching to identify concrete situations in the population living in selected areas that could be a possible focus of contamination. This study, carried through in a city of the interior of the Rio Grande do Norte, where there is a PETROBRAS industrial unity, is part of a larger research and works as a case study, where data generated for the health services could have been compared with other methodologies of inquiry in health and environment, intending to contribute to the construction of indicators. In a second study of case, the same strategy of ―evaluation of impacts in the population health‖ was used in a development project in the Amazon Region. In this study we also intended to evaluate the potential use of the information generated by the system health in monitoring ambient impacts, however in a bigger region with more diversity of situations and structure of health services. The employed methodology starts from a field recognition and uses available secondary data of health that, associated to many key informants interviews and workshops with managers, professionals of health and representatives of the community, showed the local health-illness profiles, identifying standards and processes, particularly those that could have influence the enterprise or the process in question. This study, besides the population health profile, enclosing the infrastructure of services, access and use, aims to trace the profile of the health information itself and the process of implantation, development and maintenance of different data bases. When analyzing the health information in a context of rapid transformations and low covering of services, the lack of information, or its low quality, could be used as an information itself that, in a context of existing socio-economic formations and in its process of development, might constitute important indicators. The conclusion shows the potential of the health information use for monitoring and evaluating impacts in development projects and environment transformation. It is highlighted that the information is not being used in the best way, neither by the part of the planning systems and management of the territory, nor by the health system, which are using them in administrative and bureaucratic way, without taking care to the quality and the adequacy of this utilization that, if used as it should be, could assist the planning and the actions of health, as well as fortifying the role of the health in the management of the territory. The adequate treatment and use of the information generated by the health system might provide, besides the improvement of the quality and the consequent better planning and use of the resources to the population health, an evaluation tool for the society development process. This tool, of low operational cost, allows the introduction of variables on the conditions of different population groups, formulating answers to the society questions as: ―development‖ to what and to whom?

Cette thèse a été réalisée en cotutelle Université de Montréal (UdeM) - Université Libre de Bruxelles (ULB). L'auteur a bénéficié de bourses doctorales provenant du Fonds national de la recherche scientifique (FNRS-Belgique), du Fonds de recherche du Québec-Société culture (FRQSC), et du Centre de recherche Léa-Roback sur les inégalités de santé de Montréal. / Les politiques de soutien au revenu des ménages se déclinent sous formes de mesures variées mises en place dans le cadre du système de protection sociale. Ces politiques influencent considérablement le revenu et les conditions de vie des ménages les plus vulnérables. Elles constituent un levier majeur pour réduire la pauvreté et les inégalités de revenu entre ménages. De ce fait, elles contribuent à améliorer la santé des populations et à réduire les inégalités sociales de santé (ISS) dès la naissance. L’évaluation de l’impact des politiques sociales sur les ISS dans différents contextes constitue une tâche complexe, du fait notamment de la difficulté, voire l’impossibilité, de mettre en place des études randomisées à grande échelle. Les variations des politiques sociales selon les pays constituent des opportunités pour mener des études comparatives sur base d’expériences naturelles. En partant d’un constat sur les limites des études comparatives habituelles, nous avons proposé une démarche de recherche visant à mieux étudier les spécificités des contextes afin d’expliquer les mécanismes par lesquels la combinaison des politiques de soutien au revenu influence la pauvreté des ménages et contribue aux ISS à la naissance à Bruxelles et à Montréal. Ce protocole de recherche a fait l’objet d’un 1er article. Le cœur de la thèse comprend trois parties. La première partie porte sur la comparaison des politiques d’aide sociale et d’allocations familiales en Belgique et au Québec et analyse les impacts sur l’intensité de la pauvreté des ménages à l’aide sociale dans les deux contextes. L’analyse se base sur la méthode des familles-types. Cette méthode consiste à calculer et comparer le revenu disponible de différents types de ménages. L’intensité de la pauvreté des ménages a été estimée selon le nombre d’enfants et la situation de couple. Pour chaque type de ménage, elle correspond à la différence relative entre le revenu disponible du ménage et le seuil de pauvreté relative. Les résultats montrent une intensité de la pauvreté plus marquée au Québec qu’en Belgique. Dans chaque contexte, on constate également que l’intensité de la pauvreté des ménages varie considérablement selon le nombre d’enfants et la situation de couple. Ce travail a fait l’objet d’un 2ème article. La deuxième partie porte sur la description des inégalités de santé à la naissance à Bruxelles et à Montréal. Les hypothèses de travail découlent des résultats obtenus à l’étape précédente. Deux études de cas ont été réalisées et analysées dans une perspective comparative. Les bases de données utilisées proviennent du couplage de données administratives issues des registres de naissance et des données de sécurité sociale. Les résultats ont donné lieu aux 3ème et 4ème articles. Le 3ème article concerne la population générale. Dans chaque région, des modèles de régression logistique ont été élaborés afin d’étudier l’association entre les issues défavorables de la grossesse (faible poids à la naissance, prématurité) et le statut socioéconomique (éducation de la mère et revenu). L’ampleur des inégalités de santé est plus marquée à Montréal qu’à Bruxelles et celles-ci diffèrent également selon l’origine de la mère. Le 4ème article porte spécifiquement sur la population bénéficiaire de l’aide sociale. Il compare l’association entre le faible poids à la naissance et la composition de ménage dans chaque région. On constate que les inégalités face au FPN varient selon le nombre d’enfants et la situation de couple entre les deux contextes, dans le même sens que les différences observées au niveau de la pauvreté. La troisième partie explore davantage les différences constatées à l'étape précédente selon l’immigration. Dans chaque région, elle compare l’impact du SES sur la santé périnatale chez différents groupes d’immigrés et les chez les mères nées en Belgique ou au Canada. Les résultats ont donné lieu aux 5ème et 6ème article de la thèse. L’analyse souligne l’importance de tenir compte des enjeux liés à l’immigration pour mieux expliquer la contribution des politiques de soutien au revenu aux ISS à la naissance. Cette thèse constitue une contribution unique. Dans deux régions où les taux de pauvreté et les prévalences des issues de la grossesse sont comparables dans la population générale, on constate des différences notables quant aux inégalités de santé à la naissance. Les politiques de soutien au revenu dans les deux contextes contribuent à expliquer ces différences. L’analyse démontre la nécessité de remédier aux insuffisances de ces politiques dans les deux contextes. Finalement, elle souligne les défis de la réduction de la pauvreté. Ces défis touchent à différents domaines, notamment la conciliation travail-famille, le marché du travail, l’immigration et les inégalités économiques. Mots-clés : Pauvreté, Inégalités sociales de santé, Politiques sociales, Etat-Providence, Evaluation d’impact en santé, Expériences naturelles, Inégalités de revenu, Faible poids à la naissance, Santé périnatale, Politiques de soutien au revenu. / Income support policies significantly influence the income and living conditions of the most vulnerable households. They constitute a major lever for reducing poverty and income inequalities between households. As a result, they contribute to improving the health of populations and reducing social inequalities in health (SIH) starting from birth. Assessing the impact of social policies on SIH in different contexts is a complex task, due in particular to the difficulty, if not impossibility, of setting up large-scale randomised studies. Varying social policies in different countries provide opportunities for comparative studies on the issue, based on natural experiments. Starting from an observation on the limitations of the usual comparative studies, we have proposed a research approach aiming to better study the specificities of contexts, which would allow us to explain the mechanisms by which the combination of income support policies influences household poverty and contributes to SIH at birth in Brussels and Montreal. This research protocol was the subject of a first article. This core of this thesis consists of three parts. The first deals with the comparison of welfare and family allowance policies in Belgium and Quebec and analyses their impact on the intensity of poverty of welfare households in both contexts. The analysis is based on the model family method, which consists of calculating and comparing the disposable income of different types of households. The intensity of household poverty was estimated according to the number of children and marital status. For each household type, the intensity of poverty corresponds to the relative difference between the household's disposable income and the relative poverty threshold. The results show a higher intensity of poverty in Quebec than in Belgium. It is also found that, in each context, the intensity of household poverty varies considerably depending on the number of children and marital status. This work was the subject of a second article. The second part of this thesis focuses on the description of health inequalities at birth in Brussels and Montreal. The working hypotheses are derived from the results obtained in the previous stage. Two case studies were carried out and analysed in a comparative perspective. The databases used come from a combination of administrative data from birth records and social security data. The results led to the third and fourth articles. The third article concerns itself with the general population. Logistic regression models were developed for each region to study the association between adverse pregnancy outcomes (low birth weight, prematurity) and socioeconomic status (mother's education and income levels). The magnitude of health inequalities is greater in Montreal than in Brussels and also differs according to the mother's origin. The fourth article focuses specifically on welfare recipients. It compares the association between low birth weight and household composition in each region. We can see that inequalities in LBW vary according to the number of children and marital status in both contexts, much like the differences observed in terms of poverty. The third part further explores the differences observed in the previous stage according to immigration. It compares the impact of SES on perinatal health among different immigrant groups and among mothers born in Belgium or Canada. The results led to the fifth and sixth papers of the thesis. The analysis underlines the importance of taking the specific issues linked to immigration into account to better explain the role that income support policies play in SIH at birth. This thesis is a unique contribution. There are notable differences in health inequalities at birth between two regions with similar poverty rates and levels of prevalence of unfavourable pregnancy outcomes among the general population. The impact of income support policies in Belgium and Quebec on the intensity of household poverty helps explain these differences. Our analysis demonstrates the need for public policies that address the inadequacy of the current income support measures in both regions. Lastly, it emphasises that the causes of poverty are interdependent and touch on various issues, including work-family balance, job insecurity, immigration and economic inequalities. Keywords: Poverty, Social inequalities in health, Social policies, Welfare state, Health impact assessment, Natural experiments, Income inequalities, Low birth weight, Perinatal health, Income support policies.