Contexto urbano e hipertensão arterial: abordagem multinível entre cidades da América Latina

Abstract

Introduction: Despite the global interest in the social determinants of arterial hypertension (AH), researches in urban areas and regions with a high prevalence of AH, such as Latin America, are limited and results are inconsistent. Objectives: To analyze how individual-level and area-level socioeconomic status are associated with AH in adults from 230 cities in eight Latin American countries and examine also examined the extent to which these associations vary by sex and across countries (Article 1), as well as to analyze the association between individual and contextual socioeconomic conditions with AH, according to sex, among older adults living in urban areas in Brazil (Article 2). Methods: To contemplate the first objective harmonized data were analyzed of 109,184 adults (57.8% women), aged between 18 and 97 years (mean: 42.7±16.4 years), from the SALURBAL project (Salud Urbana en America Latina). AH was self-reported. Individual-, sub-city- (administrative units nested within cities) and city-level education were used as proxies of socioeconomic status. Three-level multilevel logistic regression models were used (individuals, sub-cities, and cities), adjusted for age and country and stratified by sex. To contemplate the second objective, data from 6,767 participants living in urban areas (54.4% women), aged 50 or over (mean: 61.6±9.5 years), from the baseline (2015-2016) of the Brazilian Longitudinal Study of Aging (ELSI-Brazil) were analyzed. AH was self-reported. The measure of individual socioeconomic condition was the individual's education and the measure of the context was the Brazilian Deprivation Index (IBP, acronym in Portuguese), used at the census level. Two-level multilevel logistic regression models (individuals and census tracts), adjusted for age and stratified by sex were used. Results: Article 1 showed that education at the individual level was inversely associated with greater odds of AH among women (university education or higher versus lower than primary: odds ratio [OR] = 0.67, 95% confidence interval [CI] = 0.61-0.74), while for men it was positively associated (university education or higher versus lower than primary: OR = 1.65, 95% CI = 1.47-1.86). For both genders, living in sub-city areas with higher educational achievement was associated with higher odds of AH (OR per standard deviation [SD] = 1.07, 95% CI = 1.02-1.12; OR = 1.11 per SD, 95% CI = 1.05-1.18, for women and men, respectively). Furthermore, the association between city-level education and AH varied across countries. In Peru, there was an inverse association (OR for SD = 0.79, 95% CI = 0.64-0.97; OR for SD = 0.77, 95% CI = 0.59-0.99 for women and men, respectively), while no association was observed in Argentina, Brazil, Chile, Colombia, Mexico, El Salvador, and Guatemala. Furthermore, the inverse association of individual education with AH became stronger (in women) or emerged (in men) as city or sub-city education increased. Article 2 showed that the prevalence of AH by individual education and by IBP at the census tract level differs between men and women. In women, the higher the level of education, the lower the chance of AH (≥ 9 years versus ≤ 4 years of education: OR = 0.62, 95% CI = 0.52-0.74) and living in census tracts with greater deprivation was associated with greater odds of AH (OR per SD = 1.04, 95% CI = 1.01-1.09). In men, no association was observed. Conclusions: The social pattern of AH differs by gender and by analyzed context. Our results suggest that public policies aimed at dealing with the burden of AH in Latin American countries should adopt strategies that are sensitive to gender and the scope of the context in which these individuals reside.