Racial inequalities in mental healthcare use and mortality: a cross-sectional analysis of 1.2 million low-income individuals in Rio de Janeiro, Brazil 2010–2016

Discussion

This study assessed the racial/colour inequalities in PHC usage, hospitalisation and mortality for mental health disorders in a cohort of 1.2 million low-income individuals in Rio de Janeiro between 2010 and 2016. Racial/colour inequalities in mental health outcomes were generally large, with black and pardo individuals having the lowest healthcare usage rates and highest levels of mortality. These racial/colour inequalities persisted after adjusting for a wide range of socioeconomic factors. Males, black and pardo individuals, and those with lower income had the lowest PHC usage rates for mental health, whereas males, white individuals and those with higher income (of low-income individuals) had the highest hospitalisation rates. PHC usage was predominantly for anxiety and mood affective disorders, over two-thirds of hospitalisations were for psychotic syndromes, and nearly 90% of deaths were attributable to substance abuse/dependence disorders and suicide. Individuals of black race/colour, lower educational attainment and lower household income had the highest mortality rates from mental disorders. Black and pardo race/colour, and lower socioeconomic status (eg, lower education or low income) interacted to confer additional disadvantage in mental healthcare outcomes, particularly mortality.

The racial/colour inequalities in PHC usage identified in this study are highly concordant with international and Brazilian research demonstrating the lower healthcare usage for rates of Afro-descendants and darker-skinned individuals.23–25 47–58 Despite greater coverage of ESF for black and pardo Brazilians, barriers to accessing health services remain due to the uneven distribution of resources, limited health literacy and unconscious bias against non-white individuals in healthcare practice which leads to a lack of trust in healthcare professionals.5 57 This may explain the lower rates of PHC usage and hospitalisations observed in non-white individuals in this study. Evidence from health services in the USA, identified that white individuals were more likely than black individuals to get a psychotherapy appointment, and white males were twice as likely to get appointments compared with black males.58 A randomised cross-sectional study in the USA found that of patient callers offered primary care appointments, black and Hispanic callers were offered appointments further in the future despite identical insurance as their white counterparts.59 Similarly, the National Audit Office in England found that white patients found it easier to get an appointment within general practice compared with other ethnic groups across all health conditions.60 61 Other evidence suggests that structural racism can adversely affect mental health due to differential access to resources and services, and poor living conditions.5 11 62–64 Furthermore, experiences of discrimination can induce physiological and psychological reactions that can lead to adverse changes in mental health status.62 63 However, in this study, adjustment for PHC usage in hospitalisation and mortality outcomes did not change the estimates by racial/colour groups, suggesting that PHC usage was not a key determinant of racial/colour inequalities in mental health hospitalisations or mortality.

Limited Brazilian research exists regarding mental disorder hospitalisation and mortality by race/colour—one study in Salvador, Bahia found higher hospitalisation rates for black and pardo individuals compared with white, but there was a high level of missing data for race/colour.32 However, studies in high-income countries (HICs) document higher hospital admissions for mental disorders in non-white racial/ethnic groups.65 66 One potential explanation is that non-white individuals are more likely to end up in crisis care for poor mental health rather than accessing primary care, as reported in the United Kingdom and United States of America (USA).65 67 This is often due to reduced access to PHC, stigma, and distrust of health systems among non-white individuals compared with white individuals, leading to delayed treatment and therefore, increased risk of inpatient hospitalisation.68

In this study, black individuals were nearly 70% more likely to die from mental disorders than white individuals, a finding supported by a recent study in São Paulo where ‘unnatural’ causes of death, including suicide, were associated with non-white race/colour and substance-related disorders.29 This differs from HICs where White individuals with mental illnesses often have higher ‘unnatural’ mortality rates.69 70 The different findings between HICs and evidence from Brazil in relation to mental health hospitalisations and mortality may stem from multiple factors including persisting healthcare access issues beyond PHC, such as difficulties in receiving treatments and referrals to specialists, differences in quality of care, as well as substantially higher levels of deprivation and poverty for non-white individuals in Brazil.56 71 72

Socioeconomic status is highly interlinked with race and ethnicity and explains many, but not all of the inequalities across racial and ethnic groups. In many countries, darker-skinned individuals and ethnic minorities are more likely to be of lower socioeconomic status and exposed to chronic stressors and mental disorder risk factors than other racial/ethnic groups, such as financial insecurity, family instability and drug/alcohol consumption.73–75 However, this study found that racial/colour inequalities in mental health outcomes persisted after controlling for a wide range of socioeconomic variables, similar to other studies in both Brazil and HICs suggesting factors, other than socioeconomic status, explain these inequalities.76–80 Brazil has a long history of structural racism that reduces healthcare access for black, pardo and indigenous Brazilians.5 11 64 In the USA, internalised stigma and limited mental health literacy surrounding mental health is also linked to a lack of willingness to seek care for mental health and poor treatment adherence in racial/ethnic minorities.71 Differential treatment, either intentional or unintentional, by healthcare providers can reduce care quality for individuals of non-white race/ethnicity, leading to poorer mental health outcomes.5 81 US studies identify implicit bias in mental healthcare providers, with biased treatment recommendations, poorer communication and care quality.72 82 Further studies in the USA have shown that black and Latino patients have decreased confidence in mental health services and a lack of trust in healthcare professionals and treatment, contributing to lower service use.83 84 Aversion to mental health treatment has also been observed in Brazil, particularly for low-income individuals.85

This study identified that being of black or pardo race/colour interacted with measures of lower socioeconomic status (low income and education) to exacerbate poor PHC usage and mortality risks for mental health. These findings align with intersectionality theory and the multiple jeopardy hypothesis, whereby the negative effects of simultaneously belonging to multiple marginalised or disadvantaged groups is greater than the total negative effects of belonging to each social group.86 These predisposing socioeconomic factors form interlocking systems of oppression, particularly in Rio de Janeiro where income and gender inequality, violence and national issues associated with structural racism and limited employment and education opportunities are widespread.5 11 87 However, international studies, arguably in settings with lower racial/ethnic and social inequality, have shown contraindicating and inconsistent results regarding intersecting axes of race/colour inequalities in both a general health and mental health context.88–93 This warrants further research to strengthen the empirical evidence of intersectionality in healthcare and health outcomes.

There are key policy recommendations that stem from this work. There is a need to focus on the structural drivers of racial/ethnic inequalities in Rio de Janeiro, in Brazil, and internationally to identify factors amenable to intervention. One potential method to support black and pardo Brazilians is by driving Community Health Worker (CHW) engagement with mental healthcare in Brazil to improve trust in the health system and mental health services. CHWs play a crucial role in the Brazilian health system, particularly in lower-income communities, bridging health services and communities with consideration of socioeconomic, cultural, and racial and ethnic factors.94 95 Promoting access to mental health services across the city and country’s regions is also vital and identifying and addressing socioeconomic and cultural barriers to access. In Brazil, this means strengthening both healthcare and social assistance services as part of the Rede de Atenção Psicossocial (Psychosocial Care Network), which is a gateway for individuals with mental distress and disorders, particularly for those with substance abuse disorders.96 This underlies the importance of continued and equitable financial investment in mental healthcare, not only in Brazil, but globally. From the service perspective, there is a need to strengthen capacity of community-based mental healthcare to identify poor mental health, prioritising prevention and reducing costs to services, simultaneously tackling income inequalities in healthcare access.97 Overall, the findings indicate that policies must address mental disorder risk factors and the wider social determinants of mental health. This includes reinforcing protective mental health factors such as strengthening support networks and building social relationships, as well as improving access to education and employment to improve emotional and financial stability.8 98

A key strength of this study was the use of a dataset of 1.2 million individual individuals with linked health and healthcare data, and rich, detailed socioeconomic data. While linkage of datasets could have introduced bias, this was minimised by separately linking Cadastro Único to mortality datasets and primary care records, reducing interference. Further, linkage was determined as high quality with a 99% precision, as detailed elsewhere.99 This study used clinical diagnoses from electronic medical records (both ICD-10 and ICPC codes), However, under-reporting and under-diagnosis of mental disorders in LMICs is common,1 potentially underestimating the true mental disorder prevalence and as under-reporting is likely associated with race/ethnicity, inequalities may have been underestimated.100 Socioeconomic data was self-reported, potentially leading to misclassification101; however, this bias was unlikely to be sizeable enough or correlated with skin colour sufficiently to invalidate the findings.

Missing data for race/colour was reported as ‘other’, which could have affected the results, but this was less than 2.5% of individuals and also included individuals of amarelo (Asian) or Indigenous race/colour. Additionally, in recent years, there has been an increase in the number of Brazilians identifying as black/pardo and fewer identifying as white,12 102 which means that rates of PHC usage, hospitalisation and mortality could change which is important to acknowledge. The demographic profile of the cohort has implications for generalisability of the findings. The racial/colour distribution of the cohort is different to that of the general population of Rio de Janeiro, which may affect the generalisability of the results. In this study, white, black and pardo individuals represented 29.6%, 17.5% and 50.7% of the cohort, respectively and 29.3%, 17.6% and 51.1% of those ESF-registered. Yet in the city of Rio de Janeiro in 2010, white, black and pardo individuals represented 51.2%, 11.5% and 26.5% of the population, respectively.13 Furthermore, the cohort was composed of individuals claiming government welfare through the Cadastro Único. Welfare recipients often face significant socioeconomic disadvantage, barriers to healthcare access, financial instability and food insecurity which contributes to poorer mental health outcomes and can exacerbate mental disorder risk.8 However, low-income populations also face challenges in accessing high-quality healthcare and may not receive mental health diagnoses. The findings of this study should not be directly extrapolated to the general population without caution.

There are other limitations from a lack of data from other mental healthcare services in the city. Although this study addressed multiple mental disorders, PHC usage for mental health was predominantly for common mental disorders including depression and anxiety, whereas individuals with more severe mental disorders would be most likely to access the specialist Centros de Atenção Psicossocial (Psychosocial Care Centres; CAPS). Unfortunately, data from CAPS services was not available. Therefore, the hospitalisation and mortality estimates exclude conditions diagnosed/treated in certain specialist services, meaning the true burden is likely to be underestimated. Finally, this study did not adjust for prior hospitalisation and family history of mental disorders, or violence which may affect pathways to healthcare seeking or exacerbate mental disorder risk factors, and the results should be interpreted with appropriate caution.

While the specific historical and cultural factors driving racial/colour and ethnic inequalities in Brazil may differ from those in other countries, this study provides important insights that can be applied globally. Racial discrimination is not limited to Brazil; populations such as migrants, ethnic minorities and Indigenous Peoples face discrimination worldwide. Racial/ethnic disparities are widespread and unjust in many countries, with common underlying causes. While much of the evidence on inequalities comes from the USA and Europe, there is a pressing need for research from other parts of the world, particularly from LMICs and in marginalised areas like favelas. This study contributes to a better understanding of the root causes and magnitude of mental health inequalities and can serve as a platform for similar research in diverse settings, fostering a global dialogue on addressing these inequalities.