Racial disparities and maternal mortality in Brazil: findings from a national database · Silva AD, Guida JPS, Santos DS et al.

METHODS

We conducted a retrospective cross-sectional study to assess the relationship between skin color and maternal deaths in Brazil. According to the World Health Organization (WHO), maternal mortality is the annual number of female deaths from any cause related to or aggravated by pregnancy or its management (excluding accidental or incidental causes) during pregnancy and childbirth or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy.

We analyzed two public databases maintained by the Brazilian Ministry of Health. The first database was the Brazilian Ministry of Health Panel of Living Births (LB), and the second was the Brazilian Ministry of Health Panel of Maternal Mortality (MM). According to local regulations, all LB and maternal deaths (MD) must be reported to the authorities, and these data are anonymized and made public in the panels mentioned above. We evaluated data on Maternal Mortality Rate (MMR) from 2017 to 2022 for each year and the entire period. MMR is defined as the division of MD by LB, expressed in every 100,000 LB.

The two above-mentioned Panels contains data that werecollected from two mandatory reporting instruments in Brazil: the Live Birth Certificate and the Death Certificate. These forms are provided to all healthcare services by local health departments, and their information are verified by respective state health departments, with the Ministry of Health compiling and anonymizing all data. We collected data on skin color, maternal age, geographic area, and cause of maternal death from these databases. Access to both Panels is public through the Internet.

The Panels mentioned above and the Brazilian Institute of Geography and Statistics (IBGE) classify the Brazilian population based on skin color self-declaration as White, Black, Pardo, Yellow (East Asian), Indigenous, or Ignored. In Brazil, ethnicity is particularly complex due to great miscegenation, and we choose to use the word Pardo (instead of Brown or Mixed Brown, as in other studies), as it comes from the self-declaration of the women and is used in Brazilian Portuguese. In this study, we initially presented the MMR for all categories. For bivariate analysis, we compared women with self-declared skin colors as Black, Pardo, and White, comparing White women versus Black women and Pardo women versus Black women. We decided to exclude Indigenous and Yellow women from the comparative analysis for two reasons: first, other factors are related to inequities among Indigenous peoples, and second, the low prevalence of Yellow women in the general population (1.16%)14.

We compared MMR among self-referred women of Black, Pardo, and White skin color according to regions of Brazil (North, Northeast, South, Southeast, and Midwest), maternal age (0–19, 20–29, 30–39, 40 or more), cause of MM (direct causes, including hypertension, hemorrhage, infection, and abortion; and indirect causes, those that cannot be categorized among the direct causes), and in periods prior (2017–2019) and during the Coronavirus pandemic (2020–2022). For statistical analysis, the Chi-squared test and prevalence ratio (PR) and confidence interval (CI) were obtained using EpiInfo 7.0. A significance level of 5% was adopted.

We decided to adopt the group of White and Pardo women as the reference to compare maternal deaths among Black women because we understand that Black women suffer from the effects of Racism in their health outcomes in comparison to the other two groups.

The analysis of this study used a public national database available online and therefore does not require a consent form, as per the Declaration of Helsinki as amended in Hong Kong in 1964.

According to Resolution 510/2016 of Ethics in Research, the following will not be registered or evaluated by the system of the National Committee for Ethics in Research in Brazil: “Research that uses publicly accessible information, under the terms of Law No. 12,527, of November 18, 2011; research that uses information in the public domain”.