In this paper, we examine whether voters respond to investments in healthcare and access to health services. We focus on the city of Rio de Janeiro, the Brazilian capital with the highest growth in primary healthcare coverage between 2009 and 2012. The newly elected government in 2008 rapidly expanded the Family Health Program (FHP) in the city, a community healthcare program introduced in Brazil at the municipal level and which is currently the largest in the world. We assess whether the FHP expansion affected the mayor’s vote share in 2012 by exploiting extremely fine-grained geocoded variation in access to healthcare and in voting across time and space within the city. The identifying variation comes from changes in the share of voters at the polling booth level residing and enrolled in a FHP catchment area, within polling places fixed effects, triggered by idiosyncratic expansion of catchment areas over the 2008-2012 period. We find that FHP coverage is positively associated with the mayor’s vote share. Yet, the magnitude of the FHP effect is higher for the last year of government and increases with the proximity between FHP facilities and individuals’ home addresses. Our results also suggest that this impact is accompanied by relatively larger increases in vote share when the services are provided in new health facilities compared to the existing units. The evidence suggests that closer contact with services and the visible side of the service provision may be greatly rewarded by voters. Overall, this paper contributes to a better understanding of voter responsiveness of healthcare policies and its mechanisms.

This paper estimates the impact of exposure to infectious diseases during pregnancy on health outcomes at birth. By linking administrative data on births, mortality and mandatory notification of infectious diseases, we are able to assess whether a mother residing in the municipality of Rio de Janeiro was exposed to dengue fever, syphilis or tuberculosis in the nine months prior to birth. Our main empirical strategy consists in estimating linear fixed effects models controlling for neighborhood-specific and city-level trends, as well as mother characteristics and socioeconomic variables in the region of residence. We further complement our analysis with the estimation of semi-parametric survival functions. Consistent with previous evidence in the medical literature, we find that exposure to Dengue fever during pregnancy leads to an increase in preterm birth. As for syphilis, we find that exposure during pregnancy leads to higher infant mortality and lower birth weight; it also increases the odds of fetal death. Exposure to tuberculosis leads to lower birth weight and a higher likelihood of preterm birth. We assess the heterogeneity of our estimates with respect to access to primary health care by exploiting the arguably exogenous expansion of Family Health Clinics (Clínicas da Família) in Rio over the years 2001-2016. We find that primary health care is able to mitigate some, though by no means all, adverse effects of exposure to infectious diseases during pregnancy. Interestingly, there appears to be a “postponement” effect in exposure to syphilis: access to primary health care decreases the effect of exposure on fetal death, though it increases the positive impact on infant mortality. 

Seasonal infectious diseases, such as dengue, have been causing great losses in many countries around the world in terms of deaths, quality of life, and economic burden. In Brazil, this is relevant not only in large cities such as Rio de Janeiro and São Paulo but, according to the Ministry of Health, in another 500 cities throughout the country. Predicting the occurrence of diseases, such as dengue bursts, can be a valuable instrument for public health management as health officials can better prepare and redirect resources to the affected areas. In this paper, we present an explainable machine learning model to forecast the number of dengue occurrences in a large metropolis, Rio de Janeiro. We focus on explainable models, which provide health authorities with the reasons for outbreak predictions, allowing them to plan their actions accordingly. We trained a gradient boosting decision tree algorithm (CatBoost) with data from the National System of Information on Notifiable Diseases (SINAN), weather data, and socio-demographic data from The Brazilian Institute of Geoaraphy and Statistics (IBGE).

This article has the objective to study patterns of interaction between the Executive and the Legislative branches of government, in the City of São Paulo, between the promulgation of the City Council bylaw (1991) and the legislative period ended in 2016. We seek to observe if there are pattern differences’ between the municipal level and the Federal level through the usage of the coalition’s presidentialism model. We use data from all laws presented according to its presenter to work with the concepts of “dominance” and “success” at the municipal level. The main findings are described as: I -the percentage of laws presented and voted inside each legislature period is high; II -the majority of the laws presented came from the Legislative branch; III -The Executive branch has greater success in its propositions even though this power underperforms what is seen at the Federal Level; IV -The Municipal Executive branch tax of dominance is low in comparison to the dominance presented by the Federal Executive, suggesting a greater share of power between powers in the Municipal level.