Efeito dos programas de proteção social na saúde cardiovascular em países de baixa e média renda.

Abstract

Introduction: Cardiovascular diseases (CVD) are the main causes of death in Brazil and are closely related to living conditions, particularly affecting individuals in situations of social vulnerability. In this sense, social protection programs, such as income, food and input transfer programs, and social housing have been identified as interventions capable of reducing social and health inequities, including those associated with cardiovascular health. Evidence regarding the effects of social programs on cardiovascular health is scarce, and even more so are studies on the effects of social housing programs on CVD mortality, especially in low- and middle-income countries (LMICs). Thus, the present work had the general objective of studying the effect of social protection programs on cardiovascular health in low and middle-income countries. The central theme of this Thesis was approached through two articles and a research protocol. The article entitled “Impact of social protection policies on cardiovascular health in low-and-middle-income countries: a systematic review” corresponds to a systematic review of the effects of social protection programs on cardiovascular health in low-and-middle-income countries; the protocol corresponds to the research description, entitled “Evaluating the health effect of a Social Housing programme, Minha Casa Minha Vida, using the 100 million Brazilian Cohort: a natural experiment study protocol” to study, through a quasi-experimental study, the impact of Minha Casa Minha Vida (MCMV) program on health outcomes, using data from the 100 Million Brazilian Cohort. This protocol supported the elaboration of the third article, entitled “Effect of social housing programme, Minha Casa Minha Vida, on the risk of premature cardiovascular mortality among vulnerable and underprivileged: a population based nested case-control study”. In this, we investigate the influence of the MCMV program on CVD mortality. Methods: For the systematic review study, articles published until July 31st, 2020, in MEDLINE, Scopus, Lilacs, Web of Science and Google Scholar were considered, observing the effect of social protection programs on health behaviours (diet, physical activity and use of alcohol and tobacco), cardiometabolic risk factors (hypertension, type 2 diabetes, dyslipidemia, and excess weight) and CVD mortality. All peer-reviewed studies published between 1990 and 2020, conducted in any LMICs, that quantitatively assess the effect of any social protection policy or program on the cardiovascular health of adults were included and synthesized in tables. The review article was conducted according to the reference document Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA). For the protocol, a quasi-experimental study was designed, from 2007 to 2015, after linking the cohort data with the base data of the MCMV program (2009-2015), and of CVD mortality (2007-2015), leprosy (2007-2015) and tuberculosis records (2007–2015). The exposed population was defined as individuals who signed the contract to receive the MCMV social benefit, and the unexposed group included comparable individuals within the cohort who did not sign a contract to receive social housing. The impact of MCMV on health outcomes will be estimated using different propensity score approaches to control observed confounders. Individuals will be followed up until the occurrence of the specific outcome, date of death or end of follow-up (December 31st , 2015). The analyzes will be stratified by follow-up time, age group, race/ethnicity, gender and socioeconomic status. The protocol was developed according to the guidelines suggested by the Template for Intervention Description and Replication for Population Health and Policy (TIDieR-PHP). For the empirical article of the Thesis, premature death from CVD was the only outcome studied. Also, we opted for a case-control study design paired and nested with the Cohort, with 102,882 individuals aged 30-69 years. The cases, premature deaths due to CVD, were defined considering the date of occurrence of death and paired with two controls, selected at random and without replacement in the period of death`s occurrence. Variables such as age, sex, race, education, region and receipt of Bolsa Família programme were used to match cases and controls. Receiving social housing from the MCMV program was the main exposure of the study. This was defined, considering the date of signature of the new housing contract, being necessary to be before the date of death. Conditional logistic regression models, adjusted for household characteristics and the size of the municipality, were used to model the odds ratio for premature CVD mortality. Sensitivity analyzes considering only individuals residing in municipalities with a high Human Development Index (HDI) were performed. Results: For the systematic review article, 34 studies were included, distributed in 15 low and middle-income countries. In twenty-two studies, results on health behaviours (61.8%), 12 on cardiometabolic risk factors (35.3%) and one on CVD mortality (2.9%) were identified. Results point out that social protection programs are associated with a healthy diet (N = 17/26, 65.3%), increased physical activity (N = 2/2, 100%), lower prevalence of hypertension (N = 2/3, 66, 6%) and type 2 diabetes (N = 1, 100%). These programs were associated with excess body weight (N = 8/11, 72.7%), especially among beneficiaries of unconditional food transfer programs. We found no association between social protection programs and the use of tobacco and alcohol (N = 4/7, 57.1%) or mortality from CVD (N = 1, 100%). Regarding the empirical article, the results showed that during the follow-up period (2010-2015), 34,294 cases of premature deaths from CVD were registered in the Cohort of 100 Million Brazilians, with a median age of 58 years (50- 63), who were paired with 68,588 controls, with a median age of 55 years (50-61). Most cases (59.72%) and controls (57.14%) are from large municipalities, with more than one hundred thousand inhabitants. However, those who died from CVD (cases) had higher proportions of inadequate access to water supply (9.02 vs 8.77%) and households without electricity or meter (7.21% vs 6.11%), when compared to the control group. Beneficiaries of social housing had a higher risk of premature death from CVD compared to non-beneficiaries of the social program, after adjusting the models for family housing conditions and the size of the municipality (OR = 1.18; 95% CI = 1.07- 1.30). The same pattern of association remained after analyzing the cases and controls residing in municipalities with a high HDI (ORaj = 1.16; 95% CI = 1.03-1.30). Conclusion: Social protection programs have a positive influence on improving the quality and diversity of the diet. However, they were associated with an increase in the prevalence of overweight, especially among beneficiaries of non-conditional food transfer programs. Limited evidence on its effects on other indicators of cardiovascular health (type 2 diabetes, hypertension, physical activities and dyslipidemia) and mortality from CVD, highlight the importance of further studies on the subject. Besides, we have not found studies on the effects of housing interventions on cardiovascular outcomes, especially mortality, in low and middle-income countries, despite housing being an important social determinant of health. The results of the empirical study, unexpectedly, a greater risk of premature death was observed among beneficiaries of the MCMV program compared to non-beneficiaries. However, the findings should be interpreted with caution. The construction of an enterprise in areas devoid, in particular, of health promotion service, leisure area, healthy food and adequate public transport, combined with the loss or weakening of the support and social support networks, associated with the relocation of families in new territories may have contributed to such unfavourable results. On the other hand, some study limits need to be considered, such as the short time of exposure to the program, the undereporting deaths, and the beneficiaries' previous health condition (we do not have such information), which may have influenced the results. It is also argued that the PMCMV was initially designed to meet only housing demands, not addressing issues associated with the characteristics of the neighbourhood and access to housing developments. It is important to note that other analyzes are planned (see protocol), which makes the results presented here still preliminary.