Article Text

Download PDFPDF

P79 Ten-years risk of all-cause mortality in the ELSA-brasil cohort: the intersection of race/skin color and gender
  1. SM Barreto1,2,
  2. FEG Oliveira1,
  3. L Giatti1,2,
  4. RH Griep3,
  5. LV Camelo1,2,
  6. D Chor4
  1. 1Social and Preventive Medicine, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
  2. 2Post Graduate Program of Public Health, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
  3. 3Laboratory of Health and Environment Education, Fundação Oswaldo Cruz, Rio de Janeiro, Brazil
  4. 4National School of Public Health, Fundação Oswaldo Cruz, Rio de Janeiro, Brazil


Background Historical, persistent and systematic disadvantages in health and socioeconomic indicators are observed for Brown and especially for Black individuals in Brazil when compared to White ones. Gender inequities are also present, being different for men and women, as men die more e more prematurely than women, while women generally present worse socioeconomic and morbidity markers than men. Analysis based on the intersectionality of race and gender have contribute to apprehend and deepened the understanding of the multiple and inseparable dimensions of the existing race and gender disparities in social and health factors. Although the number of studies using the intersection of race and gender are growing, few are longitudinal.

The main objective of this work was to investigate the role of race/skin colour and gender, as well as the intersection of these categories on the risk of death in the multicentric and multiracial Brazilian Longitudinal Study of Adults Health (ELSA-Brasil).

Methods A total of 14.365 civil servants from six Brazilian cities participants of the ELSA-Brasil cohort (95.1% of original cohort) were followed up from study baseline (2008–2010) up to July/2018. Deaths were identified by annual interviews and death certificates. The associations between race/skin colour and gender separately, and between the intersection of race/skin colour and gender on all-cause mortality were estimated using Cox proportional models. Adjustments for age (confounder) and socioeconomic, behavioural and health factors (potential mediators) at baseline were considered.

Results In total, 441 deaths (incidence: 3.1% ; 44/10.000 person-years) were identified. After age-adjustments, greater risks of death were observed for Black and Brown individuals in comparison with White ones (HR:1.95; 95%IC:1.53–2.49 e HR:1.64; 95%IC:1.32–2.03,respectively), and for men in relation to women (HR:1.96; 95%IC:1.62–2.38). The intersectional analysis, considering White women as the reference category, Black women (HR:1.62; 95%IC:1.13–2.41), White men (HR:1.71; 95%IC:1.28–2.28), Brown men (HR:3.04; 95%IC:2.24–4.12) and Black men (HR:3.74; 95%IC:2.65–5.27) showed greater risk of death. After considering the adjustment for potential mediating factors, the magnitudes of the associations were attenuated, but the risk of death remained statistically greater for White men, Brown men and Black men, especially for the latter group.

Conclusion Race/skin colour and gender inequalities in the risk of death were confirmed and the extent of these inequities was found to be greater when social subgroups established in the intersectionality of race/skin color and gender are considered.

  • Race inequality
  • gender inequality
  • intersectionality analysis

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.