Racial Disparities in Foot Examination among People with Diabetes in Brazil: A Nationwide Survey, 2019

Diabetic foot is a syndrome derived from biomechanical disorders and neuropathic and microvascular changes that affect patients with diabetes mellitus that are not receiving effective treatment.^1,2  If not treated properly, skin wounds in the diabetic foot can result in infection, neuropathic ulceration, and amputation.³ 

Globally, about 17 million people with diabetes develop foot ulcers each year.⁴  People with diabetic foot have more body pain, poor physical functioning, and low health-related quality of life than do people with diabetes that do not have diabetic foot.⁵  To avoid poor consequences for these individuals, foot examination should be performed regularly with all persons with diabetes to evaluate the patient’s risk and prevent diabetic foot.^1,4 

Adherence to diabetes care can be impacted by social factors such as racism, thus keeping the patient from receiving the best care. Due to systemic racism, racial disparities negatively affect the treatment of Black patients with diabetes.⁶  Those patients face a higher chance of having lower extremity amputation than do White individuals with the same condition.^7,8 

Because much of the Brazilian population endured centuries of enslavement, a significant portion still experiences high levels of inequality and deprivation. This history has had long-lasting effects on society and health care systems. The legacies of enslavement have contributed to persistent racial disparities in health outcomes, including access to quality health care. These historical injustices have compounded over time, leading to systemic issues that disproportionately affect minority populations in Brazil,⁹  manifesting in various aspects of health care, including the management and treatment of chronic conditions such as diabetes.

This study was conducted to assess the prevalence of diabetic foot negligence by health care professionals in Brazil, with a particular focus on understanding how ethnicity influences this health outcome. The aim was to uncover potential disparities in diabetic care across ethnic groups in Brazil, contributing to a deeper understanding of how ethnicity impacts health care access and outcomes for chronic conditions with a high disease burden such as diabetes. This information is important for informing targeted public health interventions and policy changes to reduce racist health disparities.

This cross-sectional, population-based study from the 2019 Brazilian National Survey of Health (Pesquisa Nacional de Saúde; PNS) was conducted by the Brazilian Ministry of Health in partnership with the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística; IBGE).

The 2019 PNS was a household survey with national representation that was conducted to estimate the health status, lifestyle, access to health services, and continuity of care of the population.¹⁰  This periodic nationwide survey makes it possible to estimate the most important indicators at the level of Brazilian states and capitals.¹⁰ 

The questionnaire was subdivided into 3 parts: domicile, all residents, and individuals. The questions pertaining to the domicile and to all residents were answered by one resident who knew about the social, economic, and health situation of all residents. The resident selected for the individual survey also had his/her weight, height, waist circumference, and blood pressure measured by previously trained data collection agents. Data collection agents from IBGE obtained the answers from the selected participants and inserted the data in an electronic device. All answers to the questions were provided by the participants.

As defined by the IBGE in Brazil, ethnicity is categorized into 5 groups based on skin color: Black (very dark skin), White (light skin), Yellow (Asians), Brown (Brazilian mixed race), and Indigenous (descendants of Brazilian native people). The term Black (negros) is used collectively for Black and Brown individuals in Brazil. This categorization was used for identifying and assessing the representativeness of these groups in society, particularly in studies and analyses focusing on racial disparities.

People 15 years or older that had been diagnosed with diabetes that were interviewed in 2019 were eligible to be included our study. We excluded Asian and Indigenous people because of the low frequency of these groups in the sample. The sample selection occurred by group in 3 stages with primary sampling unit (PSU) stratification. In the first stage, the PSU was selected by simple random sampling, maintaining the stratification from the master sample. In the second stage, a fixed number of households were randomly selected from each PSU selected in the first stage. The National Registration of Addresses for Statistical Purposes was used as a database for this selection. In the last stage, in each of the chosen households, 1 participant 15 years of age or older was randomly selected based on the list of residents obtained at the time of the interview. For this selection, a list of residents was made during the interview.¹⁰ 

Diabetic foot negligence was defined by the lack of foot examination by a health care professional. The primary outcome was assessed by means of a specific question in the questionnaire: “When was the last time a physician or healthcare professional examined your feet to check for sensitivity or the presence of injury or irritation?” Possible answers were “Fewer than 6 months,” “From 6 months to 1 year,” “From 1 to 2 years,” “From 2 to 3 years,” “More than 3 years,” and “Never had their foot examined.” People who answered “never” were considered as having the primary outcome of this research.

The independent variables included ethnicity (categorized as White and Black, with Black encompassing both Black and Brown individuals), geographic regions within Brazil (north, northeast, southeast, south, midwest), sex (women, men), age groups (in years: 15–39, 40–49, 50–59, 60–69, ≥70), education level (completed higher education, completed high school, completed elementary school, no formal education), income (multiples of the minimum wage: <1, 1-3, >3), health insurance (yes, no), enrollment in the Family Health Program (yes, no), tobacco use (no, yes); monthly alcohol consumption (none, less than once, once or more often), self-reported hypertension (no, yes), self-perceived health status (good, fair, bad), and years with diabetes (≤4, 5-10, ≥11).

The entire process of weighting and poststratification was led by IBGE to avoid biases given the probabilistic sample design and all the procedures used. All analyses included sampling design and weights.

Selected indicators from the previous PNS, held in 2013, were used to calculate the study size, such as chronic noncommunicable diseases (diabetes, hypertension, depression), violence, use of health services, having health insurance, tobacco use, practice of physical activity, and alcohol consumption.¹⁰ 

For the sampling calculation, the following aspects were considered: desired precision level of 95% confidence; sampling plan effect because of sampling by groups in multiple stages; number of households selected by PSU; and proportion of households with people in the age group of interest.

The main sample limited the PSU sample size; thus, there were different denominators for the calculation.

Because of the complex sampling design and distinct selection probabilities, expansion factors or sample weighting needed to be defined for the selected households and their selected residents. Correction for nonresponses and adjustments to total populations were included in the final weighting, which was calculated by the inverse of the selection probability expressions of each stage of the sampling plan.¹⁰ 

To describe the participants, simple proportions were calculated for the study variables, separated by Black and White individuals. The chi-square test was applied to estimate the differences between Black and White individuals and any differences regarding foot examination conducted by a health care professional.

Poisson regression with robust variance was used to investigate factors associated with never having had a foot examination by a health care professional. This statistical approach is suitable for modeling and analyzing rates and frequencies, such as the occurrence of diabetic foot examinations, and corrects for overdispersion, ensuring more reliable and accurate standard error estimates.^11,12  Four models were used to calculate prevalence ratios (PRs) and 95% confidence intervals (CIs): (1) a bivariate model; (2) a multivariate model with all study variables; (3) a multivariate model exclusively among Black individuals with all study variables; and (4) a multivariate model exclusively among White individuals with all study variables.

No imputation was made for missing data, and a P value of <.05 was deemed significant. The analysis was conducted using Stata 14.2 incorporating the complex sample design, which included design, response rate, and poststratification weights. The 2019 PNS was approved by the National Research Ethics Committee in August 2019 under opinion 3.529.376. The PNS ensured that participation was voluntary and anonymous and that withdrawal from the study was permitted. The interviews were conducted after the interviewees or their legal guardians had read and signed the consent form, ensuring participant confidentiality and anonymity.

A total of 94,111 adults were selected for the 2019 PNS. Among these, 90,846 answered the interview (96.5%), and 7374 individuals self-reported having diabetes (8.1%). Of these, 1158 were ineligible for the study: 1033 individuals did not have information about foot examination conducted by a health care professional (14.0%), 70 individuals were Yellow (0.9%), and 55 were Indigenous (0.7%). Of the 6216 adults with diabetes that were included in the study, 3800 (61.1%) self-reported as Black and 2416 (38.9%) self-reported as White.

Among the patients with diabetes, 52.1% (95% CI: 50.0%-54.2%) reported never having had a foot examination by a health care professional. A higher frequency was observed in Black individuals (55.3%; 52.5%-58.1%) than in White individuals (48.2%; 45.0%-51.5%).

In socioeconomic variables, White individuals had higher representation in the more favorable categories than did Black individuals, such as higher education (13.7% versus 5.5%), income >3 times the minimum wages (17.7% versus 6.7%), and having health insurance (37.6% versus 20.1%). More Black individuals with diabetes accessed the Family Health Program (68.6% versus 61.1%), abstained from alcohol (77.1% versus 69.2%), had hypertension as a comorbidity (65.6% versus 61.3%), had worse self-reported health status (20.4% versus 16.9%), and had been diagnosed with diabetes up to 4 years ago (38.6% versus 31.8%) (Table 1).

Black participants aged 50–59 (42.0%) and 60–69 (36.6%) with diabetes were significantly more often neglected for foot examination (P=.014), whereas among White patients, the age groups of 40–49 (22.3%) and 50–59 (56.2%) were predominant (P<.001). The distribution of the outcome significantly differed according to income, health insurance, and time since diagnosis of diabetes in both Black and White participants (P<.020). For both groups, the lower the income, the fewer the patients who had had their foot examined (57.8% for Black patients and 55.9% for White patients at the lower income strata). For instance, lack of health insurance was also a factor in the absence of foot examination for both groups: 57.3% (P=.002) of Black and 52.7% (P<.001) of White patients. When comparing time since diagnosis, the group with the highest percentage of absence of foot examination was >11 years (42.6%, P<.001) in White patients but ≤4 years (39.1%, P<.001) in Black patients. For White patients, lack of formal education led to a higher rate of ignored foot examination (52.8%, P=.003). Among Black patients only, smoking (65.1%, P=.015), higher frequency of alcohol consumption (63.3%, P=.027), and fair health status (59.4%, P=.012) were more closely related to neglect of foot examination (Table 2).

In 5 Brazilian states (Roraima, Piauí, Pernambuco, Bahia, and Espírito Santo), the frequency of never having had their feet examined by health care professionals was higher among White patients. In the 22 remaining states, this frequency was higher among Black patients (Figure 1).

After adjustment, never having had the feet examined was more prevalent in individuals of ages 15–39 (PR = 1.34, 95% CI: 1.14-1.57) and 50–59 (1.24, 1.11-1.38), with lowest education level (1.37, 1.13-1.65) and income (1.23, 1.02-1.48), with higher alcohol consumption (1.18, 1.06-1.31), with self-reported fair health status (1.11, 1.01–1.21), and with less time since the diagnosis of diabetes (1.42, 1.28-1.57) (Table 3).

The multivariate analysis restricted to the Black population indicated that patients ages 15–39 (PR = 1.28, 95% CI: 1.06-1.54), with tobacco use (1.16, 1.01-1.33), with fair health status (1.13, 1.01-1.28), and with less time since diabetes diagnosis (1.33, 1.18-1.50) had less access to foot examination. Being registered with the Family Health Program was a protection factor (0.88, 0.80-0.98). Among White patients, never having had the feet examined was more prevalent in individuals ages 15–39 (1.41, 1.07-1.86), 40–49 (1.47, 1.15-1.88), and 50–59 (1.27, 1.06-1.53), those without formal education (1.42, 1.09-1.85), and those more recently diagnosed with diabetes (1.57, 1.32-1.86).

Half of Brazilians with diabetes lacked foot examinations by a health care professional in this 2019 nationwide assessment. Frequency of this negligence was higher for Black individuals, a trend consistent across most Brazilian states. Age and recent diabetes diagnosis emerged as common factors increasing the likelihood of lack of foot examinations in both Black and White patients. In Black individuals, smoking, alcohol use, and fair health status were linked to higher chances of not receiving foot examinations. Among White patients, lower education levels were associated with this outcome. Access to primary care was protective against such neglect in Black Brazilians with diabetes.

Our study is based on self-reported data, which can lead to recall bias, because participants might have forgotten whether they have had their foot examined, and potentially to underestimation of the prevalence of lack of foot examinations in the sample. Because of the cross-sectional nature of this study, a causal relationship between the variables and outcome cannot be established.¹³  However, the large sample size and countrywide representation of the 2019 PNS strengthen the confidence in these results.¹⁰ 

According to our findings, half of Brazilians with diabetes have never had their feet examined, in conflict with the health promotion and protection, harm prevention, diagnosis, treatment, rehabilitation, and maintenance of health guidelines included in the National Program of Diabetes to prevent diabetic foot among Brazilians with diabetes.¹⁴  In the previous Brazilian nationwide survey, the 2013 PNS, a higher risk of ignored foot was found among women (59%), individuals in rural areas (67%), Brown individuals (59%), individuals without a partner (57%), those without education (58%), and those without regular doctor’s visits for diabetes (67%).¹⁵ 

Black participants were more likely to have their feet ignored. In a cross-sectional study of data from the 2013 Medical Expenditure Panel Survey in the United States with 35,086 participants, Black patients with diabetes had fewer glycated hemoglobin tests, eye exams, and flu vaccinations, but no difference was observed in foot inspection.¹⁶  In another study conducted with only African American women, perceived systemic racism was cited by those women as a major impediment to satisfactory management of diabetes.⁷  A review focused on ethnic inequalities in health care in the National Health System of the United Kingdom revealed that patients who experienced discrimination in health care were more likely to postpone care, less open to guidance on chronic diseases, and less likely to follow medical recommendations.¹⁷ 

Inclusion in the Family Health Program, the main primary care strategy of the Brazilian Unified Health System (Sistema Único de Saúde; SUS), was a protection factor against neglect of foot examinations among Black individuals with diabetes. Primary care is a widely recognized strategy to prevent diseases and promote health in communities.¹⁸  A more comprehensive chronic care model in primary care reduces both the burden of health service and the risk of cardiovascular diseases and improves the management of diabetes and other chronic diseases.¹⁹  Continuity of care is the essential component of primary care that creates links between patients and both health care professionals and health care services and promotes self-care and chronic disease management. In a Brazilian study, continuity of care was also higher in Black individuals, women, and poorer individuals; those with continuity of care had physician and dentist consultations, fewer depressive and anxiety symptoms, and higher quality of life than did individuals without continuity in primary care.²⁰  Investment in primary care would mitigate the pernicious effects of inequality and racism in health care in Brazil.

Lack of a formal education increased the poor health outcome in White individuals. This situation can affect other issues, such as the chances of getting a job and thus limiting income and access to health care resources.²¹  Along with ethnicity, class is a major driver of discrimination, especially in high-inequality settings such as Brazil. Inverse associations between glycated hemoglobin levels and education in people never diagnosed with diabetes and between this indicator and income in people with diabetes were found in a cross-sectional study with 1267 non-Hispanic Black and White people living in greater Nashville-Davidson County, Tennessee between 2011 and 2014.²²  Lower levels of education may negatively influence diabetes onset, and once the disease is established, income affects diabetes management.²²  Lack of literacy can also result in poorer diabetes management because patients might not be able to read and understand health information, such as prescriptions.⁶ 

Age seemed to affect both White and Black participants regarding appropriate foot examination. In the United Kingdom, a longitudinal study conducted from 2009 to 2019 with approximately 40,000 households revealed that participants <30 years of age had greater effects on their health due to racism.²³  Although older people are less likely to report racism, this discrimination affects all age groups and impacts mental health.²³ 

Time since diagnosis was a determinant factor for both White and Black individuals lacking foot examinations. Lack of foot care in patients with diabetes has social, family, and economic impacts, including loss of personal income, lack of productivity, and consequently family instability.²⁴  Ensuring proper care and appropriate information about self-care and monitoring of diabetes, especially at the onset of the disease, is central for a better quality of life of people with diabetes.

Tobacco and alcohol use increased the probability of Black participants not having had their foot examined. In the United States, perceived racism led to higher rates of both alcohol use and regular smoking in the National Survey of American Life in which 4462 adults were assessed from 2001 to 2004.²⁵  In this survey, adults with lower education who faced daily discrimination had higher alcohol consumption, and adults <45 years of age were more likely to smoke regularly.²⁵  Income was not assessed; however, alcohol use and regular smoking could be a reflection of societal status, which converges with our findings.

Our study is unique because of its focus on Brazil, a country often underrepresented in global health research especially in specific contexts, such as diabetic care. The study was designed specifically to addresses diabetic foot care negligence, a critical yet frequently overlooked aspect of diabetes management. The statistical approach to data analysis made it possible to provide an accurate picture of the factors influencing diabetic foot care in different ethnic and socioeconomic groups.^11,12  This unique combination of regional focus, the specific health care aspect, and proper methodological analysis contributed significantly to the understanding these health care disparities in Brazil.

In conclusion, Brazilians with diabetes have often had foot examinations ignored by health care professionals, and this lack was higher among Black individuals. Primary care was a protection factor for Black individuals with diabetes, indicating that the SUS could mitigate this scenario of racism in health care. Appropriate training of health care professionals to avoid racial prejudice in patient care, including examination of the feet of people with diabetes, is necessary in Brazil. An effort should be made to include this aspect of diabetes management in undergraduate curricula.

Tais Freire Galvao received a productivity scholarship from the National Council for Scientific and Technological Development (grant 313431/2023-0). Data used for this research is openly available at the IBGE website (https://www.ibge.gov.br/estatisticas/sociais/saude/9160-pesquisa-nacional-de-saude.html?=&t=microdados). Codes used for statistical analysis are available upon request from the corresponding author.