Alejandro Zamudio Sosa

Background: In 2026, Brazil marks the 30th anniversary of the Benefício de Prestação Continuada (BPC), a major non-contributory social pension programme supporting older adults and persons with disabilities. We evaluated its health and economic effects over three decades and projected outcomes to 2030. Methods: We used fixed-effects Poisson models to assess associations between BPC coverage and health outcomes across 5,570 Brazilian municipalities (hospitalisations and mortality) and 27 states (YLLs, YLDs and DALYs). Retrospective analyses were complemented by sensitivity analyses, counterfactual simulations, economic evaluation and projections to 2030. Findings: Each one-percentage-point increase in BPC coverage was associated with reductions of 2.9% in hospitalisations, 0.6% in mortality and 3.9% in DALYs. The programme was estimated to have averted 13.0 million hospitalisations and 571,019 deaths over 1996–2025, and 69.6 million DALYs over 1996–2023. The incremental cost was US$2,738 per DALY averted (95% UI: US$2,056–4,068), equivalent to 35.3% of Brazil's average GDP per capita. Monetised mortality benefits were 3.81 times incremental BPC expenditure (95% UI: 2.57–5.07). Expanding eligibility was projected to avert an additional 2.55 million hospitalisations, 43,731 deaths and 12.61 million DALYs (95% UI: 0.42–21.99 million) over 2026–2030. Interpretation: Higher BPC coverage was associated with substantial population health gains in Brazil over three decades, with the economic value of reduced mortality exceeding incremental programme expenditure nearly fourfold. Expanding coverage could yield further health gains by 2030, whereas restricting eligibility could reverse some gains. These findings underscore the importance of weighing health benefits against fiscal costs in NCSP assessments, highlighting its role in advancing health equity, the Sustainable Development Goals, and leaving no one behind.

Background: In 2026, Brazil marks the 30th anniversary of the Benefício de Prestação Continuada (BPC), a major non-contributory social pension programme supporting older adults and persons with disabilities. We evaluated its health and economic effects over three decades and projected outcomes to 2030. Methods: We used fixed-effects Poisson models to assess associations between BPC coverage and health outcomes across 5,570 Brazilian municipalities (hospitalisations and mortality) and 27 states (YLLs, YLDs and DALYs). Retrospective analyses were complemented by sensitivity analyses, counterfactual simulations, economic evaluation and projections to 2030. Findings: Each one-percentage-point increase in BPC coverage was associated with reductions of 2.9% in hospitalisations, 0.6% in mortality and 3.9% in DALYs. The programme was estimated to have averted 13.0 million hospitalisations and 571,019 deaths over 1996–2025, and 69.6 million DALYs over 1996–2023. The incremental cost was US$2,738 per DALY averted (95% UI: US$2,056–4,068), equivalent to 35.3% of Brazil's average GDP per capita. Monetised mortality benefits were 3.81 times incremental BPC expenditure (95% UI: 2.57–5.07). Expanding eligibility was projected to avert an additional 2.55 million hospitalisations, 43,731 deaths and 12.61 million DALYs (95% UI: 0.42–21.99 million) over 2026–2030. Interpretation: Higher BPC coverage was associated with substantial population health gains in Brazil over three decades, with the economic value of reduced mortality exceeding incremental programme expenditure nearly fourfold. Expanding coverage could yield further health gains by 2030, whereas restricting eligibility could reverse some gains. These findings underscore the importance of weighing health benefits against fiscal costs in NCSP assessments, highlighting its role in advancing health equity, the Sustainable Development Goals, and leaving no one behind.

Background. Poverty is a major determinant of premature mortality, yet it persists in low- and middle-income countries (LMICs), largely due to governments’ fiscal constraints, despite the growing concentration of wealth across these countries. We estimated the potential effects of alternative wealth redistribution policies on extreme poverty and mortality in Brazil through 2030. Methods. We conducted a two-stage modelling study. First, we estimated the association between household per-capita income and all-cause mortality using microdata from the Brazilian Census for 19,720,456 individuals. Second, we applied these estimates to large-scale microsimulation models of the same study population, comparing the business-as-usual income distribution with six alternative redistribution scenarios funded by annual taxes on the wealth of Brazilian centimillionaires (1,360 individuals with wealth >US$100 million) over 2026–2030. Findings. In the retrospective analysis, higher household per-capita income was associated with lower mortality (OR:0·798, 95% CI:0·795–0·801). In the forecasting analysis, a 1% wealth tax, generating US$2·7 billion in annual revenue for transfers to 5·3 million people living in extreme poverty in Brazil, reduced the number of deaths in this population by up to 52·2% (95% UI 51·5–53·1). Cumulative deaths averted between 2026 and 2030 ranged from 184,653 (95% UI 180,148–188,999) under the 1% poverty-gap scenario to 224,111 (95% UI 218,879–229,332) under the 3% fixed-transfer scenario. Interpretation. Wealth redistribution policies could contribute to eradicating extreme poverty and substantially reducing mortality in LMICs, thereby supporting progress towards the Sustainable Development Goals. Funding. Medical Research Council (Grant number: MC_PC_MR/Y004884/1).

Background: Social protection programs may influence chronic disease outcomes by shaping living conditions and access to healthcare, yet evidence on their effects on adult non-communicable diseases (NCDs) in low- and middle-income countries (LMICs) remains limited. We examined the effects of social assistance on reported diagnoses of diabetes and hypertension among low-income adults in Brazil and Mexico. Methods: A two-country quasi-experimental study using nationally representative health survey data collected between 2018 and 2019. The study included 21,616 individuals living in low-income households, representing a weighted population of 35,607,974 adults in Brazil and Mexico. Inverse probability of treatment weighting (IPTW)–weighted logistic regression models were used to estimate the effects of social assistance on reported diagnosis of diabetes and hypertension, adjusting for sociodemographic and health-related confounders. Sensitivity analyses produced consistent results. Findings: Receipt of social assistance was associated with lower odds of reporting a diagnosis of diabetes (OR=0·80; 95% CI: 0·73–0·87) and hypertension (OR=0·81; 95% CI: 0·76–0·85). Associations were stronger among individuals in the lowest income strata (OR for diabetes=0·74; 95% CI: 0·66–0·84) and more pronounced in Mexico across overall and stratified analyses by sex and income. Interpretation : Among low-income adults in Brazil and Mexico, receipt of social assistance was associated with lower odds of reported diagnosis of diabetes and hypertension, consistent with a potential effect on chronic disease prevalence. These findings suggest that social assistance may influence NCDs outcomes through pathways related to socioeconomic conditions and healthcare access.

Background: Social protection programs may influence chronic disease outcomes by shaping living conditions and access to healthcare, yet evidence on their effects on adult non-communicable diseases (NCDs) in low- and middle-income countries (LMICs) remains limited. We examined the effects of social assistance on reported diagnoses of diabetes and hypertension among low-income adults in Brazil and Mexico.Methods: A two-country quasi-experimental study using nationally representative health survey data collected between 2018 and 2019. The study included 21,616 individuals living in low-income households, representing a weighted population of 35,607,974 adults in Brazil and Mexico. Inverse probability of treatment weighting (IPTW)–weighted logistic regression models were used to estimate the effects of social assistance on reported diagnosis of diabetes and hypertension, adjusting for sociodemographic and health-related confounders. Sensitivity analyses produced consistent results.Findings: Receipt of social assistance was associated with lower odds of reporting a diagnosis of diabetes (OR=0·80; 95% CI: 0·73–0·87) and hypertension (OR=0·81; 95% CI: 0·76–0·85). Associations were stronger among individuals in the lowest income strata (OR for diabetes=0·74; 95% CI: 0·66–0·84) and more pronounced in Mexico across overall and stratified analyses by sex and income.Interpretation: Among low-income adults in Brazil and Mexico, receipt of social assistance was associated with lower odds of reported diagnosis of diabetes and hypertension, consistent with a potential effect on chronic disease prevalence. These findings suggest that social assistance may influence NCDs outcomes through pathways related to socioeconomic conditions and healthcare access.

Background: Early motherhood remains a major public health and equity challenge in low- and middle-income countries, especially in Latin America. While poverty, education, gender norms, and barriers to sexual and reproductive health are well-established determinants, how household income translates into (or fails to translate into) delayed childbearing remains insufficiently understood—particularly in highly unequal settings marked by informality and fragmented social protection. Objective: To examine the association between household income and early motherhood in Mexico, and to disentangle direct and indirect socioeconomic pathways linking income to early childbearing. Methods: We conducted a population-based, multi-method analysis using harmonised municipal microdata from the 2000, 2010, and 2020 Mexican Population and Housing Censuses ( n = 3.05 million women aged 12–24 years, representing 30.6 million individuals). Total household income was imputed using Quantile Random Forests trained on nationally representative income–expenditure surveys. Our analytical strategy combined: (1) machine learning models to identify key correlates and non-linear individual-level patterns of early motherhood; (2) Bayesian spatial models (INLA) to quantify subnational clustering and contextual heterogeneity; and (3) dynamic structural equation modelling to assess direct and lagged relationships among income, schooling-related indicators, union formation, and early motherhood outcomes. Results: Early motherhood prevalence declined modestly from 21.46% (2000) to 19.31% (2020) and remained concentrated in socioeconomically disadvantaged households. In individual-level models, marital status, age, household structure, and dependency ratio were the dominant correlates; income ranked eighth in feature importance. Partial dependence plots indicated a non-linear inverse L-shaped association: predicted risk increased up to approximately Int$ 1,760 (2018 PPP) per adult equivalent, then plateaued and declined slightly. Parity-specific models showed that income’s predictive relevance weakened with higher parity, with U-shaped patterns at the lowest income levels for second and ≥third births. Municipal analyses revealed persistent spatial clustering after adjustment for key predictors, and residual “place” effects consistent with unmeasured contextual constraints shaping early motherhood. In dynamic models, the direct income–early motherhood path was not statistically significant. Overall patterns were consistent with predominantly indirect linkages operating through schooling- and union-related indicators and reinforcing disadvantage dynamics at the municipal level. Conclusions: In Mexico, early motherhood remains strongly shaped by intersecting social and territorial inequalities. Income gradients are modest and appear to operate mainly through schooling lag and union-formation pathways rather than as an isolated driver. Equity-oriented strategies are likely to be most effective when they integrate school retention, prevention of early unions, and youth-friendly sexual and reproductive health services, prioritising municipalities with persistently elevated risk.

Chronic diseases disproportionately affect certain ethnic and gender groups, but the social determinants driving these disparities in Latin America are not fully understood. In this study, we analyzed data from national health surveys in Brazil, Mexico, and Ecuador (2018–2019), representing a total weighted population of 96,726,891 adults. We used random forest models to predict chronic disease diagnoses based on education, occupation, and access to essential services such as sanitation, drinking water, and garbage collection. Our models performed better for indigenous and afro-descendant people, highlighting significant inequalities. While occupation and education were strong predictors for women, decreasing model performance by 8.57% and 7.36% respectively, occupation was the critical variable for men, decreasing model performance by 19.6% when neutralized. This work highlights the need for public policies adapted to the specific needs of different ethnic and gender groups.

Introduction Conditional economic incentives can improve medication-taking behaviors among populations at risk of contracting human immunodeficiency virus (HIV). However, there are no data on the cost-effectiveness of incentive programs for improving pre-exposure prophylaxis (PrEP) adherence among male sex workers (MSWs) who have one of the highest HIV acquisition rates. Our objective was to assess the cost-effectiveness of incentive programs to improve adherence to pre-exposure prophylaxis (PrEP) among male sex workers Methods We conducted an economic evaluation of the PrEP Seguro randomized pilot trial in Mexico (ClinicalTrials.gov: NCT03674983). Among n = 110 MSWs, those randomized to the intervention received tiered incentives based on PrEP drug levels in scalp hair measured at three clinic visits over 6 months. The intervention led to a 28.7% increase in scalp hair PrEP concentration, consistent with increased adherence (p = 0.05). Here we use a micro-costing approach from the health system perspective to calculate costs. Quality-adjusted life years (QALYs) were estimated from the number of HIV infections averted through sufficient PrEP adherence (tenofovir concentration > 0.011 ng/mg corresponding to greater than or equal to three weekly doses). Incremental cost-effectiveness ratios (ICERs) estimated the cost/QALY gained owing to the intervention. Results The mean cost per patient was US $165.53 and $179.55 among standard care and incentive patients, respectively. Over 6 months of follow-up, 62% of standard care patients and 78% of incentive recipients were PrEP adherent. After the program, the lifetime average QALYs gained per infection avoided were 9.17 (minimum, maximum: 7.5, 10.8) and 9.84 (minimum, maximum: 8.05, 11.6) among standard care and incentive patients, respectively. The 6-month ICER was US $20.92/QALY gained by the intervention, which was highly cost-effective at a willingness-to-pay of US $8655 (Mexico’s 2020 per capita gross domestic product (GDP)). Discussion Using behavioral economics approaches for enhancing adherence to HIV prevention may offer health and fiscal benefits through reduced HIV incidence. Fully powered implementation trials can determine future cost-effectiveness of scaling up incentives for PrEP adherence among high-risk populations.

El propósito del presente estudio fue explorar la relación entre variables cognitivas, emocionales, contextuales y la topología de las redes no digitales en el surgimiento, mantenimiento y extinción de la acción colectiva mediante un modelo basado en agentes. Con ayuda de NetLogo, se desarrollaron diferentes topologías de red, se modeló la identidad social, eficacia colectiva, norma social, enojo, hermetismo del sistema político. Se realizaron 22,000 mil simulaciones variando el número promedio de nodos que conectan a los agentes. Se encontró que el número promedio de días donde se mantenía activa la acción colectiva se relacionó en forma de sigmoide con el número promedio de conexiones entre los agentes. El enojo como variable psicológica motivó fuertemente a los agentes a participar en la acción colectiva. Los modelos basados en agentes ayudan a comprender fenómenos como la emergencia, la transición de fase o la autoorganización en fenómenos sociales complejos como la acción colectiva.

El propósito del presente estudio fue evaluar la relación entre determinantes sociales y la conducta del voto municipal en las elecciones federales en México de 2009, 2012, 2015 y 2018. Se accedieron a datos abiertos y se usó interpolación lineal para obtener un total de 13 determinantes sociales. Mediante regresiones lineales con efectos mixtos y la técnica de eliminación recursiva de características con el algoritmo random forest, se exploraron las relaciones entre la proporción del voto general, por grupos de edad y sexo, y los determinantes sociales. Entre los principales resultados se destacan las asociaciones negativas y significativas entre la conducta del voto y la tasa de homicidios y el índice de rezago social. La población con mayor participación electoral fueron las mujeres de 30 años y más con el 52,69 %, 70,44 %, 60,59 % y 73,41 %, en las elecciones de 2009, 2012, 2015 y 2018, respectivamente. El algoritmo random forest encontró que fueron los años de la elección, la cobertura de Seguro opular y la educación las variables más importantes para predecir la proporción del voto.