The 2019 national health survey: what does it teach us about the epidemiological profile of stroke in Brazil? Authors: Jussara Almeida de Oliveira Baggio, Bruna Roberta Lopes do Nascimento, Ariana Figueiredo Suassuna, Lucas Almeida Andrade, Marcio Bezerra-Santos, Letícia Januzi de Almeida Rocha Published: 2025 ABSTRACT AbstractBackgroundStroke is one of the leading causes of death in Brazil, as well as a major cause of disability in the population. Because of this, national surveys are highly relevant for understanding the impact and planning public policies. The aim of the study was to assess the epidemiological profile and associated factors of stroke in Brazil from the 2019 National Health Survey (NHS).MethodsWe conducted an ecological and population-based study using data from the 2019 NHS in Brazil. Fifty-one clinical and sociodemographic variables related to stroke were selected for analysis in this study. The descriptive analysis were performed using SPSS software, and we conducted a spatial analysis of the federative units regarding the incidence of stroke and the prevalence of functional dependency per federative unit. We also performed a logistic regression with the aim of identifying the predictor variables of functional dependency in the population.ResultsA total of 1,975 individuals had received a stroke diagnosis, corresponding to a prevalence of 2.1% of the Brazilian population or a total of 4,413,090 people living with a stroke diagnosis in 2019. Most of them were female (52.9%), brown (48.8%), and lived in urban areas (78.1%). The mean age was 63.77 ± 14.93 years, and the mean age for stroke diagnosis was 53.13 ± 18.92 years. The Northwest region of Brazil had the highest number of stroke cases, 731 (37%). We found that 32.4% of individuals were classified as functionally dependent, and male sex, rehabilitation in the last 12 months, use of medication for stroke, and being accompanied by health professionals were predictors of functional dependence.ConclusionThe epidemiological profile of stroke in Brazil reveals a reduction in the number of individuals with severe and very severe functional limitations, however, a high prevalence of functional dependence remains. This burden is unevenly distributed across Brazilian states. Moreover, the majority of individuals did not receive any rehabilitation therapies in the past 12 months. Notably, undergoing rehabilitation and being followed by a healthcare professional were identified as key predictors of functional dependence.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12889-025-25656-y.Keywords: Stroke, Functional dependence, Survey Background A stroke occurs when something blocks blood supply to the brain or when a blood vessel in the brain bursts. It can cause brain damage, long-term disability, and death. Metabolic risks account for 68.8% of the global stroke burden, while behavioral factors such as smoking, poor diet, and physical inactivity account for 35.2%, and environmental risks contribute 36.7% [1]. Importantly, stroke is the second leading cause of death and the third leading cause of death and disability worldwide [2]. In 2019, the Global Burden of Disease (GBD) had a stroke incidence of 12.2 million (11.0–13.6) and a stroke incidence of 101 million patients (93.2–111), with 6.55 million deaths [3]. However, the global distribution of stroke is disproportionate, with 87% of deaths and 89% of disability-adjusted life years (DALYs) occurring in low- to middle-income countries (LMICs) [1]. The disparities in the impact of stroke can be observed in a country of continental dimensions like Brazil. Previous studies have shown a significantly higher lethality rate [4] and a greater proportion of patients with functional dependence [5] in cities in the Northeast region of Brazil. However, these studies analyze specific regions, and national surveys are needed for a better epidemiological understanding of stroke in Brazil. The National Health Survey (NHS), conducted by the Brazilian Institute of Geography and Statistics (IBGE) along with the Brazilian Ministry of Health, is an essential tool for comprehensive epidemiological assessments of the country. This research had two editions, the first in 2013 and the second in 2019, covering self-perception of health status, lifestyles, chronic diseases, and oral health [6]. The 2013 NHS showed a self-reported prevalence of stroke of 1.5% (95% CI: 1.4%–1.7%) in Brazil, representing approximately 2,231,000 stroke survivors. Among them, 25.5% reported severe limitations resulting from the stroke [7, 8]. In the 2019 NHS, the data published so far show an increase in the self-reported prevalence of stroke to 2.0% (95% CI: 1.8%–2.1%) [8]. Therefore, a more in-depth analysis is needed to understand the changes in the epidemiological profile of the population between the two NHS editions. We aimed to assess the epidemiological profile and associated factors of stroke in Brazil using data from the 2019 NHS. Methods Type and study design An ecological and population-based study using data from the 2019 NHS —a household survey with national representation—was conducted. The main objective of the NHS was to produce national-level data on the health status and lifestyles of the Brazilian population, as well as on healthcare access, including service use, preventive actions, continuity of care, and healthcare financing. The survey is structured around three main axes: the performance of the national health system; health conditions and surveillance of diseases and health conditions; and associated risk factors. The target population of the NHS included individuals aged 15 years or older living in private permanent households, meaning homes built exclusively for residential purposes. The questionnaire consists of 26 modules and was administered by agents previously trained by IBGE. It also underwent a validation process and was adjusted as necessary to improve comprehension by the target population. The NHS adopted three-stage cluster sampling, in which the first sample of units was census tracts or sets of tracts; the second sample of units were private households, selected randomly; and finally, the third sample of units was selected for residents aged 15 years or older. The interviews were conducted between August 2019 and March 2020. The questionnaire covered aspects relating to the prevalence of stroke and its respective variables and was applied only in the third stage of the NHS sample. Specific publications provide more details on the methodology [6]. The 2019 NHS was approved by the National Research Ethics Commission (CONEP) in August 2019 (number 3,529,376). Additionally, it agreed with Resolution 196/96 of the National Health Council, assuring research subjects of their voluntariness, anonymity, and possibility of withdrawing from the study at any time. All the interviewees signed a free and informed consent form. Study area Brazil is located in South America and is the fifth largest country in the world, with approximately 211 million inhabitants, making it the 5th most populous country. The country is politically and administratively divided into 27 federative units (26 states and one Federal District) and 5,570 municipalities. The states are grouped into five regions (North, Northeast, Southeast, South, and Midwest) with distinct geographic, economic, and cultural characteristics for political and operational purposes [9]. Variables We selected 51 clinical and sociodemographic variables for analysis in this study. The supplementary methods describes the baseline questions for constructing the indicators employed. Statistical and spatial analyses The analyses were performed using SPSS software (version 20.0; Chicago, IL, United States) at a significance level of 5%. Continuous variables are summarized as the means and standard deviations, and categorical variables are presented as absolute and relative frequencies with their respective 95% confidence intervals (95% CIs). To calculate the stroke incidence rate, the number of cases was divided by each federal unit’s total population, and the result was multiplied by 100,000 inhabitants. For this purpose, population data were obtained from the IBGE based on population estimates for the year 2019 [10]. For the spatial analyses, choropleth maps were constructed using the epidemiological and clinical indicators of stroke in each federative unit of Brazil. The maps were generated with QGIS 3.28.7 software. Additionally, the digital cartographic mesh of Brazil (divided by federative units/states) was extracted from the Geographic Projection System in shapefile format (Geodesic Reference System, SIRGAS/2000) from the IBGE website [10]. Based on the response to the item “Degree to which stroke limits the respondent’s routine activities (e.g., working, housework, etc.)”, participants in the study were classified as functionally dependent or functionally independent for the logistic regression analysis. Those who reported that stroke moderately, intensely, or very intensely limited their daily activities were classified as functionally dependent, while those who reported no limitation or only slight limitation were classified as functionally independent. Moreover, logistic regression analysis was performed using a dichotomous dependent variable, functional dependence (FD- = 0 and FD + = 1) with the following admission variables to predict group associations: human development index (HDI), sex, age, rehabilitation in the last 12 months, health status, physical exercise, years of schooling, use of medication, and health professional status. All variables were classified as categorical, except HDI. These variables were entered into the logistic regression to confirm the final variables in the model. The Hosmer and Lemeshow test confirmed that all the regression models adequately fit the data (p value > 0.05). Results In the 2019 NHS, a total of 90,846 participants aged 15 years or older were interviewed. A total of 1975 individuals had received a stroke diagnosis. These data correspond to a prevalence of 2.1% of the Brazilian population or a total of 4,413,090 people living with a stroke diagnosis in 2019. Most of them were female (52.9%), brown (48.8%), and lived in urban areas (78.1%). The mean age was 63.77 ± 14.93 years, and for stroke diagnosis, the mean age was 53.13 ± 18.92 years. The Northwest Region of Brazil had the highest number of stroke cases, at 731 (37%). Fig. 1 illustrates each federal unit’s stroke incidence rate per 100,000 inhabitants. The highest incidence rates were observed in the states of Roraima (6.8/100,000 inhabitants), Acre (6.1), and Amapá (4.9). Stroke incidence rate per 100,000 inhabitants in each federative unit in Brazil. Data from the 2019 National Health Survey Additionally, the participants were asked whether the stroke had caused some limitation in usual activities (such as working and doing household chores); 962 individuals (48.7%) reported no limitation, 373 (18.9%) reported little limitation, 239 (12.1%) reported moderate limitation, 232 (11.7%) reported intense limitation, and 169 (8.6%) reported very intense limitation. For further comparisons, we classified the individuals into functional independence (no limitation and low limitation) and functional dependence (moderate, intense, and very intense limitation) groups. Table 1 shows the sample characteristics of all individuals with a stroke diagnosis, classified as functional dependence or independence. We observed that people with functional dependence were older, male, and had less than nine years of schooling. Considering the occurrence of functional dependence per federal unit, the highest percentages were observed in the states of Rondônia (47%), Alagoas (41%), Rio Grande do Norte (40%), Mato Grosso do Sul (40%), Espírito Santo (39%), and Sergipe (37%) (Fig. 2). Table 1Sociodemographic characteristics of people with stroke and classified as presenting functional dependence or functional independenceAll Stroke group(n = 1975)Functional dependence(n = 640)Functional independence(n = 1335)n% (95%CI)n% (95%CI)n% (95%CI)Age (years)63.77±14.93-65.33±14.07-62.5±15.16-Age of stroke diagnosis (years)53.13±18.92-56.69±18.58-51.53±18.85-Sex (%) Female Male104493152.9 (50.6–55.05.6.05)47.1 (44.9–49.3)31132948.59 (44.7–52.4)51.4 (47.5–55.2)73360254.9 (52.2–57.5)45.09 (42.4–47.7)Ethnicity Brown White Black Indigenous Asian963709267201648.8 (46.5–50.9)35.9 (33.8–38.04.8.0413.5 (12.08–15.1)1 (0.6–1.5)0.8 (0.5–1.3)321222867450.2 (46.6–54.02.6.02)34.7 (31.1–38.4)13.4 (11.01–16.3)1.1 (0.5–2.24)0.6 (0.02–1.6)642487181121248.1 (45.4–50.7)36.5 (33.9–39.1)13.6 (11.8–15.5)0.9 (0.5–1.5)0.9 (0.5–1.5)Schooling years < 9 years ≥ 9 years Not declared109450737455.39 (53.1–57.5)25.67 (23.7–27.6)18.93 (17.2–20.7)36312814956.7 (52.8–60.5)20 (17.08–23.2)23.3 (20.1–26.7)73137922554.8 (52.08–57.4)28.4 (26.04–30.8)16.9 (14.9–18.9)Geographic region Northwest Southwest North South Midwest73143035924620937 (34.9–39.1)21.8 (20.01–23.6)18.2 (16.5–19.9)12.5 (11.07–13.9)10.6 (9.3–13.6)258146102706440.3 (36.5–44.1)22.8 (19.7–26.2)15.9 (13.3–18.9)10.9 (8.75–13.5)10 (7.91–12.5)47328425717614535.4 (32.9–38.03.9.03)21.2 (19.1–23.5)19.3 (17.2–21.4)13.2 (11.4–15.1)10.9 (9.3–12.6)Residence Urban area Rural area154243378.1 (76.2–79.8)21.9 (20.1–23.8)49015076.6 (73.1–79.6)23.4 (20.3–26.8)105228378.8 (76.5–80.9)21.2 (19.09–23.4)Associated factors Hypertension (yes) Diabetes (yes) Actual smoker (yes) Past smoker (yes) Hypercholesterolemia (yes) Sedentary lifestyle (yes)1371497263837656150869.4 (67.3–71.4)25.2 (23.3–27.1)13.4 (11.8–14.8)42.4 (40.2–44.5)33.2 (31.1–35.3)76.4 (74.4–78.1)4581737427420755271.6 (67.9–74.9)27 (23.7–30.6)11.5 (9.31–14.2)42.8 (39.03–46.6)32.3 (28.8–36.06.8.06)86.3 (83.3–88.7)91332418956344995668.4 (65.8–70.8)24.3 (19.09–30.6)14.1 (12.3–16.1)42.2 (39.5–44.8)33.6 (31.1–36.2)71.6 (69.1–73.9)95% CI: 95% Confidence Interval Sociodemographic characteristics of people with stroke and classified as presenting functional dependence or functional independence Sex (%) Female Male 1044 931 52.9 (50.6–55.05.6.05) 47.1 (44.9–49.3) 311 329 48.59 (44.7–52.4) 51.4 (47.5–55.2) 733 602 54.9 (52.2–57.5) 45.09 (42.4–47.7) Ethnicity Brown White Black Indigenous Asian 963 709 267 20 16 48.8 (46.5–50.9) 35.9 (33.8–38.04.8.04 13.5 (12.08–15.1) 1 (0.6–1.5) 0.8 (0.5–1.3) 321 222 86 7 4 50.2 (46.6–54.02.6.02) 34.7 (31.1–38.4) 13.4 (11.01–16.3) 1.1 (0.5–2.24) 0.6 (0.02–1.6) 642 487 181 12 12 48.1 (45.4–50.7) 36.5 (33.9–39.1) 13.6 (11.8–15.5) 0.9 (0.5–1.5) 0.9 (0.5–1.5) Schooling years < 9 years ≥ 9 years Not declared 1094 507 374 55.39 (53.1–57.5) 25.67 (23.7–27.6) 18.93 (17.2–20.7) 363 128 149 56.7 (52.8–60.5) 20 (17.08–23.2) 23.3 (20.1–26.7) 731 379 225 54.8 (52.08–57.4) 28.4 (26.04–30.8) 16.9 (14.9–18.9) Geographic region Northwest Southwest North South Midwest 731 430 359 246 209 37 (34.9–39.1) 21.8 (20.01–23.6) 18.2 (16.5–19.9) 12.5 (11.07–13.9) 10.6 (9.3–13.6) 258 146 102 70 64 40.3 (36.5–44.1) 22.8 (19.7–26.2) 15.9 (13.3–18.9) 10.9 (8.75–13.5) 10 (7.91–12.5) 473 284 257 176 145 35.4 (32.9–38.03.9.03) 21.2 (19.1–23.5) 19.3 (17.2–21.4) 13.2 (11.4–15.1) 10.9 (9.3–12.6) Residence Urban area Rural area 1542 433 78.1 (76.2–79.8) 21.9 (20.1–23.8) 490 150 76.6 (73.1–79.6) 23.4 (20.3–26.8) 1052 283 78.8 (76.5–80.9) 21.2 (19.09–23.4) Associated factors Hypertension (yes) Diabetes (yes) Actual smoker (yes) Past smoker (yes) Hypercholesterolemia (yes) Sedentary lifestyle (yes) 1371 497 263 837 656 1508 69.4 (67.3–71.4) 25.2 (23.3–27.1) 13.4 (11.8–14.8) 42.4 (40.2–44.5) 33.2 (31.1–35.3) 76.4 (74.4–78.1) 458 173 74 274 207 552 71.6 (67.9–74.9) 27 (23.7–30.6) 11.5 (9.31–14.2) 42.8 (39.03–46.6) 32.3 (28.8–36.06.8.06) 86.3 (83.3–88.7) 913 324 189 563 449 956 68.4 (65.8–70.8) 24.3 (19.09–30.6) 14.1 (12.3–16.1) 42.2 (39.5–44.8) 33.6 (31.1–36.2) 71.6 (69.1–73.9) 95% CI: 95% Confidence Interval Percentage of functional dependence due to stroke complications per federative unit in Brazil. Data from the 2019 National Health Survey Regarding stroke clinical complications and patient follow-up in Brazil, only 322 individuals (16.3%) reported using locomotor aid, 220 (68.3%) reported using a walking stick, crutch or walker, 124 (38.5%) used a wheelchair, 7 (2.1%) used an orthosis and 5 (1.5%) used other locomotor aid. The majority of the participants received locomotor aid outside of the Brazilian Unified Health System (SUS). In addition, most of them (n = 1082; 54.8%) had not received regular rehabilitation in the last 12 months. Among those who received any rehabilitation in the previous 12 months, 252 (61.2%) were in public health services or affiliated with the SUS, 140 (34%) were in private services or affiliated with health insurance, 17 (4.1%) other type of health service and 3 (0.7%) did not know the kind of service. Additionally, 337 (17.1%) patients received physiotherapy, and 156 (7.9%) patients received another kind of rehabilitation. In the 2019 Brazilian NHS, 1599 (81%) individuals did not have any health insurance, but 1840 (93.2%) had a medical appointment in the last year, 453 (22.9%) used aspirin, and 1050 (53.2%) used other types of medication for stroke. In addition, 1091 (55.2%) did not change their diet because of stroke. When asked about their self-perception of health status, 65 (3.3%) answered “very good”, 449 (22.7%) “good”, 917 (46.7%) “regular”, 405 (20.5%) “poor”, and 139 (7%) “very poor”. The logistic regression model was significant, categorizing 71.6% correctly. Male sex (exp(b) = 1.836 [95% CI: 1.40–2.40]), rehabilitation in the last 12 months (exp(b) = 1.993 [95% CI: 1.48–2.66]), use of medication for stroke (exp(b) = 3.1013 [95% CI: 2.19–4.14]) and being accompanied by health professionals (exp(b) = 1.469 [95% CI: 1.06–2.02]) were predictors of functional dependence. Additionally, having good health status (exp(b) = 0.199 [95% CI: 0.112–0.356]), having regular health status (exp(b) = 0.314 [95% CI: 0.191–0.516]), and practicing physical exercise (exp(b) = 0.314 [95% CI: 0.304–0.610]) were protective factors for functional dependence (Table 2). Table 2Results from logistic regression analysis (dependent variable functional independence = 0 and functional dependence = 1)Independent variablesßSEWalddf P Exp(B)95% CIHDI−0.961.4470.4410.500.3830.22–6.53Male0.600.13819.5110.00011.8361.40–2.40Age−0.0130.0072.9810.080.9870.97–1.002Rehabilitation in the last 12 months (yes)0.690.14921.48310.00011.9931.489–2.668Health status (good)−1.610.29619.6210.00010.1990.112–0.356Health status (regular)−1.1590.25420.8510.00010.3140.191–0.516Physical exercise (yes)−0.8430.17822.36910.00010.4310.304–0.610Years of school (less than 9 years)0.1660.161.07810.2991.1810.863–1.616Use of medication (yes)1.1030.16246.3310.00013.0132.193–4.140Health professional monitoring0.3850.1645.50810.0191.4691.065–2.026Constant0.0821.1520.00510.9431.085Model X 2236.29, df = 14, p = 0.000Pseudo R 20.249Hosmer and Lemeshow test (χ2)12.721, df = 8, p = 0.122HDI Human Development Index, SE standard error, Wald Wald statistics, Pseudo R2 Nagelkerke R square, df degrees of freedom, CI confidence interval Results from logistic regression analysis (dependent variable functional independence = 0 and functional dependence = 1) HDI Human Development Index, SE standard error, Wald Wald statistics, Pseudo R2 Nagelkerke R square, df degrees of freedom, CI confidence interval Discussion This study aimed to analyze data on stroke and its associated factors reported in the 2019 NHS. Our analyses revealed a self-reported stroke incidence of 2.1%, which demonstrates the significant impact of this disease as a public health issue on a national scale. Moreover, the prevalence was greater among women, brown individuals, those with lower education levels, older age groups, and individuals in the Northeast region. Most patients do not receive adequate health support or specialized follow-up after a stroke episode. Taken together, these data corroborate the critical situation of the disease in Brazil and reinforce the need for investment in public policies to improve treatment and rehabilitation services for stroke victims. Compared to the previous edition of the 2013 NHS, two studies reported a 0.6% increase in stroke prevalence [7]. The 2019 GBD study compared the rates of incident stroke cases in Brazil between 1990 and 2019 and found a 43.3% reduction in stroke incidence and a 57.8% decrease in stroke-related deaths [11]. Another study also demonstrated a significant 37% reduction (95% CI 32–42) in stroke incidence between 1995 and 2013 in Joinville, located in the southern region of Brazil [12]. These findings highlight improvements in socioeconomic conditions and advancements in acute stroke treatment in the country during this period [13]. Despite the downward trend shown by these studies, the 2019 NHS identified a slight increase in self-reported stroke prevalence. This may be related to the demographic transition experienced in Brazil in recent years [14], along with a significant rise in noncommunicable diseases (NCDs). Estimates indicated a 76% increase in mortality rates from NCDs in Brazil in 2019 [15]. Another factor that helps explain the increased prevalence of stroke is the rising trend in risk factors. The prevalence of hypertension and diabetes mellitus increased in the 2019 NHS compared to the 2013 edition, rising from 21.4% to 23.9% for hypertension and from 6.2% to 7.7% for diabetes mellitus, respectively [16, 17]. This pattern aligns with global trends. The number of individuals aged 30 to 79 diagnosed with hypertension doubled between 1990 and 2019, while the incidence of diabetes increased by 102.9% from 1990 to 2017 [18, 19]. Notably, the 2019 NHS also showed that the Northeast region had the highest stroke prevalence. Brazil has long been marked by significant social disparities, with the Northeast region having the lowest Human Development Index (HDI) in the country [9]. In 2021, the state of Alagoas, located in this region, recorded the highest premature mortality rates attributable to major stroke risk factors: high blood pressure (90.96 per 100,000), high BMI (65.67 per 100,000), and high LDL cholesterol (47.51 per 100,000) [15]. Several studies have shown that regions with lower socioeconomic conditions experience a higher prevalence of stroke and greater mortality rates [20–22]. These concerns are further underscored by the findings of the SAMBA study, which compared stroke incidence, lethality, and functional outcomes across Brazil’s macroregions. The study found that Sobral, a city in the Northeast, had the highest stroke lethality and the worst functional outcomes 90 days post-stroke [4]. Together, these findings illustrate how stroke outcomes vary across regions in Brazil and reinforce the urgent need for increased investment in stroke prevention, acute care, and rehabilitation services, particularly in the North and Northeast of the country. Furthermore, the 2013 NHS revealed that 25% of individuals with stroke experienced severe or very severe limitations as a result of the condition [7]. Our data from the 2019 NHS indicated a decrease, with 20.3% of individuals experiencing this limitation. This decrease may be attributed to advancements in public policies related to stroke treatment and the publication of ordinances 665 and 664 in 2012. Following this, there was an expansion in the number of stroke units in the country, which may have impacted the reduction in severe functional dependence [23, 24]. However, employing a different classification, we found that 32.4% of individuals were classified as functionally dependent. These data are noteworthy, particularly considering that participants had been diagnosed for an average of approximately ten years. Among the states, Rondônia had the highest prevalence of functional dependence (47%), followed by Alagoas (41%), Mato Grosso do Sul (40%), Rio Grande do Norte (39.7%), Espírito Santo (39.5%), and Sergipe (37.1%). Despite the high level of functional dependence, the majority (54.8%) had not received regular rehabilitation in the last 12 months. Access to stroke rehabilitation is a global issue, particularly in low- and middle-income countries (LMICs), where rehabilitation services are less available [25]. There are few specialized outpatient services, which are primarily concentrated in large urban centers [26]. Additionally, distance, transportation, and workforce shortages pose additional barriers to treatment continuation [26]. In Brazil, there are 278,708 physiotherapists and 52,528 speech therapists, most of whom are Southeast region [27, 28]. These data may account for the concentration of individuals with functional dependence in states localized in the country’s North, Northeast, and Midwest regions. Likewise, it is worth noting that having access to rehabilitation and being accompanied by health professionals are predictors of functional dependence. One possible explanation is that more severely ill patients may have access to rehabilitation for extended periods, or the quality of the service provided may not meet their needs. Recently, Europe developed a Stroke Action Plan to define efficient and cost-effective long-term rehabilitation strategies [29]. All countries should follow this example to prioritize stroke rehabilitation. According to Bernhardt et al., the next challenge in advancing stroke recovery is ensuring that the rehabilitation services offered to patients are timely, of adequate intensity and duration, high quality, and effective in meeting the patient’s needs [30]. This study has some limitations. The results are based on self-reported data, which depend on participants’ interpretation and understanding and may lead to underestimation. A study conducted in a poor neighborhood of São Paulo reported an age-adjusted stroke prevalence of 4.6% in men and 6.5% in women [31]. Another study using the same methodology in the Northern region of Brazil found a prevalence of 6.3% in rural areas and 3.7% in urban areas [32]. These variations may reflect the selection of specific regions and populations and cannot be generalized to the national level. The same applies to studies based on specialist-confirmed cases focused on specific geographic areas [12, 33]. At the national level, public data from the Ministry of Health on stroke-related deaths and hospitalizations also help illustrate the impact of stroke in the country, although they are subject to interpretation limitations [34, 35]. Therefore, surveillance studies such as the NHS are essential to better understand the country’s health status and support the development of effective public policies. Conclusion According to the 2019 NHS, the prevalence of self-reported stroke was 2.1% in the Brazilian population, or a total of 4,413,090 people living after stroke diagnosis in 2019 in Brazil. The most affected groups were females, nonwhite individuals, those with lower education levels, elderly individuals, and those living in the Northeast Region. Compared to the 2013 NHS, there was a reduction in the number of individuals with severe and very severe functional limitations. However, we observed a high prevalence of individuals with functional dependence and a heterogeneous distribution across Brazilian states. Additionally, most participants did not receive rehabilitation therapies in the last 12 months, and we identified predictors for functional dependence to be undergoing rehabilitation in the previous 12 months and being followed by a healthcare professional. Supplementary Information Supplementary Material 1 Acknowledgements Not applicable. Abbreviations Global Burden of Disease Brazilian Mortality Information System Brazilian Civil Registry Center National Health Survey Brazilian Institute of Geography and Statistics National Research Ethics Commission 95% confidence intervals Functional dependence Human development index Brazilian Unified Health System Cardiovascular diseases Low- and middle-income countries Authors’ contributions JAOB, BRLN, AFS, LAA, MBS, LJAR: conceptualization, data curation, formal analysis, investigation, methodology; writing-original draft, writing-review and editing. Funding No funding to declare. Data availability All data of 2019 NHS are available on the website https://www.pns.icict.fiocruz.br/. Declarations Ethics approval and consent to participate The 2019 NHS was approved by the National Research Ethics Commission (CONEP) in August 2019 (number 3,529,376). Additionally, it agreed with Resolution 196/96 of the National Health Council, assuring research subjects of their voluntariness, anonymity, and possibility of withdrawing from the study at any time. All the interviewees signed a free and informed consent form (ICF). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement All data of 2019 NHS are available on the website https://www.pns.icict.fiocruz.br/.