Completeness of the COVID-19 vaccination schedule and associated factors among persons experiencing homelessness in Brazil: a national cross-sectional study Authors: Juliana Soares Tenorio de Araújo, Thaís Zamboni Berra, Ariela Fehr Tártaro, Rander Junior Rosa, Reginaldo Bazon Vaz Tavares, Yan Mathias Alves, Maria Eduarda Pagano Pelodan, Heriederson Sávio Dias Moura, Letícia Perticarrara Ferezin, Antônio Carlos Vieira Ramos, Murilo César do Nascimento, Giselle Lima de Freitas, Ricardo Alexandre Arcêncio Published: 2026 ABSTRACT AbstractObjectivesTo analyse the completeness of the COVID-19 vaccination schedule and identify factors associated with vaccine uptake.DesignCross-sectional study.SettingData were collected through face-to-face interviews conducted in all 26 Brazilian state capitals and the Federal District between 2022 and 2023, using a sequential sampling approach.ParticipantsA total of 1392 individuals aged 18 years or older experiencing homelessness for at least 6 months were included.Primary and secondary outcome measuresThe primary outcome was the completeness of the COVID-19 vaccination schedule (complete vs incomplete), based on self-reported vaccination status. Secondary analyses examined sociodemographic, institutional and behavioural factors associated with vaccine uptake using binary logistic regression.ResultsCompletion of the vaccination schedule was positively associated with receiving government aid (OR: 1.58; 95% CI 1.09 to 2.30), visits from street clinic health agents (OR: 3.19; 95% CI 1.95 to 5.36), prior COVID-19 diagnosis (OR: 5.77; 95% CI 3.17 to 11.15), support for mandatory vaccination (OR: 3.76; 95% CI 2.48 to 5.76), trust in vaccine efficacy (OR: 3.92; 95% CI 2.63 to 5.89), seeking information from community sources (OR: 1.91; 95% CI 1.01 to 3.88) and trust in federal authorities (OR: 1.57; 95% CI 1.06 to 2.31).ConclusionsThis study identified structural, social and individual factors associated with complete COVID-19 vaccination among people experiencing homelessness in Brazil. Although overall coverage was substantial, gaps in vaccination completeness persisted. Social support, healthcare outreach and trust in vaccines were associated with higher uptake, highlighting important barriers and facilitators to vaccination in socially vulnerable populations. STRENGTHS AND LIMITATIONS OF THIS STUDY The study included 1392 participants across all Brazilian capital cities, providing unprecedented nationwide data on COVID-19 vaccination barriers among people experiencing homelessness. The study included 1392 participants across all Brazilian capital cities, providing unprecedented nationwide data on COVID-19 vaccination barriers among people experiencing homelessness. The study included a wide range of sociodemographic, institutional and behavioural variables, allowing a comprehensive analysis of factors associated with vaccination uptake. The study included a wide range of sociodemographic, institutional and behavioural variables, allowing a comprehensive analysis of factors associated with vaccination uptake. The absence of a unified national registry for homelessness may limit the generalisability of the results. The absence of a unified national registry for homelessness may limit the generalisability of the results. The sample may not fully capture the heterogeneity of Brazil’s homeless population, particularly in rural areas or non-capital cities. The sample may not fully capture the heterogeneity of Brazil’s homeless population, particularly in rural areas or non-capital cities. Introduction The development of vaccines against COVID-19 represents a major advancement in science, technology and public health. The approval of safe and effective vaccines by regulatory agencies in record time was a remarkable achievement and marked a historic milestone for Brazil’s National Immunisation Programme, comparable to the eradication of smallpox in the previous century.1 However, ensuring equitable distribution, fair access and widespread acceptance of vaccines remains a significant challenge, particularly among people experiencing homelessness (PEH). Structural inequalities are a major barrier to achieving this goal.2 Disparities have emerged, especially among individuals with incomplete vaccination schedules. Evidence shows that the more advanced a person’s vaccination status, the lower their risk of death from COVID-19, with mortality potentially reduced by up to 16-fold depending on the age group.3 4 PEH is at greater risk of infection and progressing to severe forms of the disease⁵. Living in conditions of extreme poverty, they are particularly vulnerable to infectious diseases and face serious barriers to accessing healthcare services.5 6 Brazil operates under a universal healthcare system, known as Sistema Único de Saúde (SUS), a national public policy that provides free and universal healthcare to the population. It includes primary healthcare (PHC) services, such as basic health units and the Community Health Agents Programme, which play a key role in health promotion, monitoring vulnerable populations and facilitating referrals to specialised care.7 Despite this universal coverage, PEH face specific barriers that limit access to these services, including lack of identification documents, stigma and geographical mobility. Although some studies have addressed COVID-19 vaccination among PEH,8 9 10 11 most research has focused on high-income countries in the Global North, particularly developed economies, with an emphasis on vaccine access and acceptance.12 13 However, little is known about the situation of PEH in Latin America, including Brazil, where structural inequalities such as extreme poverty, stigma and social exclusion exacerbate barriers to healthcare and vaccination.2 7 A systematic review of vaccination coverage among PEH indicated that although adherence to vaccination was considerable, overall coverage remained substantially lower than in the general population, with single-dose coverage rates ranging from 35% to 47%.14 Moreover, existing studies have not sufficiently addressed the social determinants affecting vaccination in this group, including discrimination, lack of identification documents and limited access to healthcare services.5 15 The absence of specific data on Brazil and other Latin American countries represents a critical knowledge gap, as public health strategies must be tailored to local contexts to be effective. Focusing on Brazil, one of the most socially unequal countries,6 this study seeks to contribute to understanding the barriers faced by this population in accessing vaccination. The study aims to analyse the completeness of the COVID-19 vaccination schedule among PEH in Brazil and identify factors associated with vaccine uptake, providing evidence to inform targeted public health interventions that strengthen equity and effectiveness in vaccination campaigns. This study is grounded in a social determinants of health perspective, considering both structural determinants (eg, social protection, housing conditions and access to healthcare) and individual-level factors (eg, trust, attitudes and information-seeking behaviours) that may influence vaccination uptake. In this context, ‘associated factors’ are understood as encompassing these multiple dimensions. Materials and methods Study design This cross-sectional study is part of the broader ‘COVID-19 Social Thermometer in Brazil’ initiative and consists of a secondary analysis of data collected through this project. The study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.16 Study setting The study was conducted across Brazil’s 26 states and the Federal District. Brazil has a population of approximately 203 million people distributed across a vast territory of 8.5 million km². The country is marked by significant regional disparities in socio-economic development, access to healthcare and demographic profiles. The Human Development Index (HDI) varies widely, with the Federal District reporting the highest HDI, while states in the North and Northeast frequently exhibit the lowest indicators, particularly in schooling and income.17 Brazil experienced one of the most severe COVID-19 epidemics in the world, with over 37.5 million confirmed cases and 702 116 deaths reported by April 2025, resulting in an overall case fatality rate of 1.87%.18 19 The pandemic affected all federal units, with São Paulo recording the highest number of cases and deaths.18 19 The population experiencing homelessness in Brazil has increased sharply over the past decade, a situation further exacerbated by the COVID-19 pandemic. Between 2012 and 2022, the number of PEH grew by 211%, reaching approximately 281 400 individuals.18 20 Regarding the national vaccination campaign, which began in January 2021, Brazil achieved high coverage rates: 88.1% of the population received at least one dose, 81.8% completed the primary vaccination schedule and 58.7% received booster doses.21 PEH were included as a priority group for vaccination starting in late 2020.21 Immunisation efforts were facilitated by policies allowing vaccination without requiring personal identification documents.21 Study population and sampling strategy The study population consisted of individuals aged 18 years or older, either Brazilian citizens or naturalised, who were experiencing homelessness or living in public spaces for a minimum of 6 months at the time of the interview. Living in public spaces refers to individuals experiencing unsheltered homelessness, such as those residing on streets, sidewalks or other open public areas. A sequential sampling strategy was employed.22 23 Participants were approached consecutively at predefined recruitment sites, including healthcare facilities, social assistance centres and community spaces frequently attended by the target population. Recruitment was conducted continuously until the predefined sample size was reached, ensuring sufficient statistical power and representation across study locations. Procedures were implemented to minimise duplicate participation, including field coordination strategies and the use of unique identifiers, such as fingerprint registration, exclusively for the purpose of ensuring data quality. The minimum sample size (n=385) was calculated using Cochran’s formula,24 considering a 95% confidence level, 5% margin of error and a 50% estimated proportion. An additional 10% was included to account for potential sample losses.22 24 The final sample comprised 1392 participants. Data collection instrument Data were collected using the questionnaire ‘Social Thermometer COVID-19: Social Opinion’, originally developed by the National School of Public Health at NOVA University Lisbon and subsequently adapted and validated for the Brazilian context using the Delphi technique.25 26 The final instrument included 141 variables, comprising checklists, multiple-choice questions and Likert-type scales. The questionnaire was implemented using the REDCap (Research Electronic Data Capture) platform, hosted by the University of São Paulo (USP), Ribeirão Preto campus.27 28 This platform ensures data traceability, consistency and secure storage for observational and clinical research.29 The full questionnaire Social Thermometer COVID-19_Social Opinion is available in online supplemental file 1. Data collection Data collection was conducted between 2022 and 2023 through face-to-face interviews using mobile devices (smartphones or tablets). Interviewers were trained and standardised by the study coordination team following a common protocol applied across all study sites to ensure consistency in data collection procedures and minimise measurement bias. Each interview lasted approximately 20–30 min. Participants were recruited through a national network involving research institutions, universities, civil society organisations and social movements, which facilitated access to vulnerable populations through trust-building strategies. This network was previously established as part of the broader ‘COVID-19 Social Thermometer in Brazil’ initiative. No financial incentives were provided for participation. To improve data reliability, interviewers used probing questions regarding vaccination status and, when available, verified vaccination cards. However, self-reported information may be subject to recall and social desirability bias. Study variables The dependent variable was the completeness of the COVID-19 vaccination schedule (Yes=1; No=0), based on self-reported vaccination status. A complete vaccination schedule was defined according to Brazilian national guidelines at the time of data collection, corresponding to completion of the primary vaccination series, regardless of booster doses.3 Independent variables included: sex (male, female); race/colour (black/brown, white); marital status (single, divorced/widowed, married/stable union); occupation (unemployed, retired, student, formal employment, informal employment); education level (no schooling, elementary, high school, higher education or more); monthly income (no income, <1 minimum wage, 1–5, 5–10, >10); receipt of government aid (yes, no); use of SUS (yes, no); presence of a basic health unit in the area (yes, no); prior COVID-19 diagnosis (yes, no); agreement with mandatory vaccination (yes, no); perceived social pressure (yes, no); trust in vaccine effectiveness (yes, no); trust in government actions (yes, no); and sources of vaccination information. Questions regarding specific information sources (official sources, informal sources, health professionals and street clinics/community leaders) were asked to all participants, regardless of whether they reported actively seeking vaccine-related information. The variable ‘Did not seek information’ refers to the absence of active information-seeking behaviour. Institutional variables included Visit by a community health agent (CHA) (yes, no); Visit by a street clinic health agent (SCHA) (yes, no). CHAs are linked to PHC services, while SCHAs are part of mobile street clinic teams that provide outreach care specifically to PEH. Race/colour classification followed the Brazilian Institute of Geography and Statistics: white, brown (mixed race) and black.30 For analytical purposes, variables were dichotomised where appropriate. Data analysis Descriptive analyses were conducted to characterise the study population, including absolute and relative frequencies for categorical variables and measures of central tendency for age, using SPSS V.22.31 Binary logistic regression was used to assess factors associated with vaccination completeness. Multicollinearity was evaluated using the variance inflation factor (VIF), and variables with VIF >10 were excluded.32 Model selection was performed using the backward method based on the Akaike information criterion.32 Adjusted ORs with 95% CIs were estimated. Model fit was assessed using the Hosmer-Lemeshow test, likelihood ratio tests and pseudo-R² measures (Cox-Snell, Nagelkerke and McFadden). Predictive performance was evaluated using the area under the receiver operating characteristic (ROC) curve.32 Analyses were performed using RStudio V.4.3.133, adopting a significance level of 5%. Patient and public involvement Patients or members of the public were not involved in the design, conduct, reporting or dissemination plans of this research. Results Sociodemographic profile of the sample A total of 1392 PEH completed the questionnaire. Table 1 presents their sociodemographic profile. The majority were male (n=961; 69.0%), with a mean age of 38.3±15.2 years (range: 23–81). Most participants self-identified as black or brown (n=1097; 78.8%). A large proportion reported being single, divorced or widowed (n=1247; 89.6%) and had completed only elementary education (n=1060; 76.1%). Table 1Sociodemographic profile of people experiencing homelessness, Brazil, 2022–2023.VariablesFrequency(n=1392)%Gender Male96169.0 Female42930.8 Others20.2Race/colour Black and brown109778.8 White29521.2Marital status Single, divorced, widowed124789.6 Married/stable union14510.4Occupation Unemployed66948.0 Informal32823.6 Formal22916.5 Retired966.9 Student705.0Schooling Primary and lower secondary education106076.1 High school27319.6 Higher or more533.9 No schooling60.4Income in minimum wage Above 1020.1 From 1 to 564846.6 From 5 to 10211.5 Less than 151537 No income20614.8Received government aid Yes58341.9 No80958.1Uses the Unified Health System Yes98971.0 No40329.0Visited by the community health agents Yes47133.8 No92166.2There is a basic health unit Yes73853.0 No36726.4 I do not know28720.6 Age 8.3±15.2 years (minimum: 23, maximum: 81) Sociodemographic profile of people experiencing homelessness, Brazil, 2022–2023. Regarding occupation, 48.0% (n=669) were unemployed, and 23.6% (n=328) worked in the informal sector. In terms of income, 46.6% (n=648) earned between one and five minimum wages, while 37.0% (n=515) earned less than one minimum wage. Access to social policies was limited: 41.9% (n=583) received some form of government assistance, and 71.0% (n=989) reported using the Unified Health System. However, only 33.8% (n=471) received visits from CHAs. Additionally, 53.0% (n=738) stated that a basic health unit was available in their neighbourhood. Vaccination status and perceptions about the vaccine Table 2 shows that 28.5% (n=397) of participants reported having received a confirmed COVID-19 diagnosis, while 70.5% (n=981) had not and 1.0% (n=14) were unsure. Regarding vaccination status, 62.4% (n=869) had completed the primary COVID-19 vaccination schedule. Although some participants reported receiving at least one dose of a COVID-19 vaccine, the present analysis focused on completion of the primary vaccination schedule as the main outcome, as it better reflects adequate and sustained immunological protection. Table 2The COVID-19 diagnosis, vaccination schedule and people experiencing homelessness’ perceptions of measures to address the pandemic, Brazil, 2022–2023VariablesFrequency%(n=1392)Diagnosed with COVID-19 No98170.5 Yes39728.5 I do not know141Complete basic scheme against COVID-19 Yes86962.4 No52337.6Do you feel social pressure to take the COVID-19 vaccine? Yes41629.9 No88163.3 I do not know956.8Do you agree with mandatory vaccination against COVID-19? Yes81058.2 No48835.1 I do not know946.8Trust in the effectiveness of vaccines Yes97570 No32523.3 I do not know926.6Trusts the state government’s actions on vaccines Trust101873.1 Do not trust26719.2 I do not know1077.7Trust in the federal government’s actions on vaccines Trust95968.9 Does not trust32523.3 I do not know1087.8Trusts the actions of the municipal government on vaccines Trust101472.8 Does not trust27119.5 I do not know1077.7 The COVID-19 diagnosis, vaccination schedule and people experiencing homelessness’ perceptions of measures to address the pandemic, Brazil, 2022–2023 Moreover, 58.2% (n=810) supported mandatory vaccination, while 35.1% (n=488) did not and 6.8% (n=94) were undecided. A total of 70.0% (n=975) expressed confidence in the effectiveness of vaccines, although 63.3% (n=881) reported not feeling social pressure to be vaccinated. Trust in government actions varied across levels, with 73.1% (n=1018) reporting trust in state governments, 72.8% (n=1014) in municipal governments and 68.9% (n=959) in federal government actions. Sources of information Among those who sought information about the COVID-19 vaccine, 20.5% (n=285) obtained information from healthcare professionals, and 13.3% (n=185) sought guidance from street clinics or community leaders; these were the most commonly consulted sources of information. However, the majority did not consult official sources (94.8%; n=1320), and 93.8% (n=1305) did not rely on informal sources. Regarding general information exposure about COVID-19, 86.6% (n=1205) of participants reported receiving information through official press outlets, including television news, radio and newspapers. In contrast, when specifically asked about actively seeking information regarding COVID-19 vaccination, 87.3% (n=1215) reported not seeking vaccine-related information from any source. Despite this overall low level of active information-seeking, important differences emerged according to the type of source used. Individuals who sought information through street clinics or community leaders showed higher odds of completing the vaccination schedule, whereas reliance on media or informal sources was associated with lower vaccination uptake. Detailed data are presented in table 3. Table 3Sources of information about the COVID-19 vaccine among people experiencing homelessness, Brazil, 2022–2023VariablesN (1392)%Sources of official information No132094.8 Yes725.2Informal sources No130593.8 Yes876.3Health professionals No110779.5 Yes28520.5Street clinics and community leaders No120786.7 Yes18513.3Did not seek information No121587.3 Yes17712.7 Sources of information about the COVID-19 vaccine among people experiencing homelessness, Brazil, 2022–2023 Factors associated with complete vaccination: multivariate analysis Table 4 presents the results of the binary logistic regression analysis. Seven factors were identified as being associated with a higher likelihood of having a complete COVID-19 vaccination schedule among PEH. Individuals who received some form of government assistance were 1.58 times more likely to be fully vaccinated (OR: 1.58; 95% CI 1.09 to 2.30). Those who received visits from CHAs were 3.19 times more likely to have completed the vaccination schedule (OR: 3.19; 95% CI 1.95 to 5.36), and individuals with a history of a positive COVID-19 diagnosis were substantially more likely to be fully vaccinated, presenting the strongest association observed in the model (OR: 5.77; 95% CI 3.17 to 11.15). Table 4Factors associated with completeness of the COVID-19 vaccination schedule among people experiencing homelessness, Brazil, 2022–2023VariableCategoryAdjusted OR (95% CI)P valueSexMaleReference—Female0.70 (0.46 to 1.06)0.09Uses the Unified Health System (SUS)NoReference—Yes1.58 (0.99 to 2.50)0.06Received government assistanceNoReference—Yes1.58 (1.09 to 2.30)0.01*Visited by street clinic health agentNoReference—Yes3.19 (1.95 to 5.36)<0.001*Primary healthcare unit in the neighbourhoodNoReference—Yes0.60 (0.40 to 0.91)0.01*Prior COVID-19 diagnosisNoReference—Yes5.77 (3.17 to 11.15)<0.001*Agrees with mandatory vaccinationNoReference—Yes3.76 (2.48 to 5.76)<0.001*Trusts vaccine effectivenessNoReference—Yes3.92 (2.63 to 5.89)<0.001*Sought information from informal sourcesNoReference—Yes0.04 (0.01 to 0.25)<0.001*Sought information from street clinics or community leadersNoReference—Yes1.91 (1.01 to 3.88)0.04*Sought information from official sourcesNoReference—Yes0.04 (0.01 to 0.23)<0.001*Did not seek informationNoReference—Yes0.37 (0.24 to 0.59)<0.001*Trusts the federal government regarding vaccinesNoReference—Yes1.57 (1.06 to 2.31)0.02*Adjusted ORs obtained from multivariable logistic regression.The variable ‘Uses the Unified Health System (SUS)’ was retained in the final model despite a p value slightly above 0.05 because its exclusion resulted in an increased Akaike information criterion indicating poorer model fit. Given its theoretical relevance to healthcare access among people experiencing homelessness, it was maintained as an adjustment variable.*p<0.05 indicates statistical significance.SUS, Unified Health System (Sistema Único de Saúde). Factors associated with completeness of the COVID-19 vaccination schedule among people experiencing homelessness, Brazil, 2022–2023 Adjusted ORs obtained from multivariable logistic regression. The variable ‘Uses the Unified Health System (SUS)’ was retained in the final model despite a p value slightly above 0.05 because its exclusion resulted in an increased Akaike information criterion indicating poorer model fit. Given its theoretical relevance to healthcare access among people experiencing homelessness, it was maintained as an adjustment variable. p<0.05 indicates statistical significance. SUS, Unified Health System (Sistema Único de Saúde). A higher likelihood of complete vaccination was also observed among those who agreed with mandatory COVID-19 vaccination (OR: 3.76; 95% CI 2.48 to 5.76), trusted the effectiveness of the vaccines (OR: 3.92; 95% CI 2.63 to 5.89), sought information through street clinics or community leaders (OR: 1.91; 95% CI 1.01 to 3.88) and expressed trust in the federal government’s statements about vaccines (OR: 1.57; 95% CI 1.06 to 2.31). Conversely, individuals who relied on official press outlets for information (OR: 0.04; 95% CI 0.01 to 0.23), consulted informal sources (OR: 0.04; 95% CI 0.01 to 0.25) or did not seek any information about COVID-19 (OR: 0.37; 95% CI 0.24 to 0.59) were less likely to be fully vaccinated. Interestingly, reporting the presence of a PHC unit in the neighbourhood was associated with lower odds of complete vaccination. This finding may reflect that geographical proximity does not necessarily translate into effective access to utilisation of services among PEH, who often face structural, administrative and psychosocial barriers that persist regardless of service availability. Variables such as sex, marital status, schooling and other covariates showed no statistically significant association with complete vaccination (p>0.05). To validate the binary logistic regression model (table 4), comprehensive statistical tests were performed, including the Hosmer-Lemeshow test (p=0.14), likelihood ratio test (p<0.01) and fit indices such as Cox-Snell (0.24), Nagelkerke (0.42) and McFadden (0.32). Additionally, the model showed a good fit, with a ROC curve value of 0.86, indicating satisfactory predictive capacity. Discussion This study analysed the completeness of the COVID-19 vaccination schedule among PEH in Brazil and identified factors associated with vaccine uptake. The findings suggest that both structural factors, such as receiving government assistance and contact with SCHAs, and individual-level factors, including confidence in vaccine effectiveness and attitudes towards vaccination, were associated with higher odds of completing the primary vaccination schedule. At the same time, active information-seeking regarding COVID-19 vaccination was limited among participants, highlighting important challenges related to communication and access to reliable information. The sociodemographic profile of the participants was characterised predominantly by men, individuals self-identifying as black or brown, low educational attainment and unstable employment conditions. These characteristics are consistent with previous Brazilian studies on PEH and reflect broader social inequities that disproportionately affect historically marginalised populations.34 35 Similar patterns have also been described in studies conducted in North America and Europe, where structural racism, poverty, unemployment and housing instability are recognised as important determinants of homelessness and barriers to healthcare access.36 In the context of the COVID-19 pandemic, these vulnerabilities may also contribute to unequal access to vaccination and preventive measures.37 38 The association between receiving government assistance and greater vaccination coverage suggests that social protection policies may help reduce barriers to healthcare access among PEH. Participation in social programmes may facilitate institutional linkage, improve access to documentation and transportation and increase opportunities for interaction with public services, including healthcare.7 35 Previous studies conducted in socially vulnerable populations have similarly suggested that income support and social assistance may be associated with improved healthcare engagement and vaccination uptake.8 39 However, a considerable proportion of participants reported not receiving any government assistance, which may reflect potential barriers related to bureaucratic requirements, limited access to information and institutional distrust.7 34 39 Structural inequities that may have intensified during the pandemic could have further contributed to unequal access to preventive measures, healthcare services and vaccination opportunities among socially vulnerable populations, reinforcing pre-existing social exclusion processes.40 The association between visits from SCHAs and higher vaccination completeness reinforces the potential relevance of outreach-based healthcare strategies for PEH. These professionals may contribute to reducing logistical barriers, facilitating communication and strengthening links with PHC services.7 41 42 Community-based approaches, including mobile healthcare teams and shelter-based interventions, have been associated with improved vaccine uptake and increased trust in healthcare services among homeless populations.43 International evidence also suggests that interventions led by community health workers may be associated with greater engagement in vaccination campaigns through culturally sensitive communication and trust-building strategies.43 In this context, the Brazilian ‘Consultório na Rua’ model may represent an important mechanism for expanding healthcare access among socially vulnerable groups.41 42 Although more than half of participants reported completing the primary COVID-19 vaccination schedule, vaccination coverage remained lower than that observed in the general Brazilian population during the same period.4 44 Similar disparities have been reported in studies from other countries, where PEH consistently presented lower vaccination uptake despite the availability of vaccines.8 14 39 45 These differences may reflect persistent structural barriers affecting access to healthcare services and vaccination opportunities among socially marginalised populations.38 40 Previous studies have emphasised that inclusion health populations, such as PEH, frequently experience difficulties related to stigma, mobility, documentation requirements and fragmented access to health systems, which may limit the effectiveness of universal vaccination strategies46 . An important finding of this study relates to the role of information sources and trust. Although most participants reported not actively seeking information about COVID-19 vaccination, individuals who sought information through street clinics or community leaders showed higher odds of completing the vaccination schedule. In contrast, reliance on informal sources or official media outlets was associated with lower vaccination uptake. These findings should be interpreted cautiously, as the cross-sectional design does not allow causal inferences. Nevertheless, the results appear to suggest that proximity, trust and communication dynamics are relevant factors influencing vaccination behaviour among PEH. Previous studies have shown that institutional distrust, misinformation and communication barriers may contribute to lower vaccine uptake among socially marginalised populations.37 39 The WHO has also highlighted accessible, culturally sensitive and context-adapted communication strategies as potentially relevant for improving vaccine acceptance in vulnerable groups.47 Trust in vaccine effectiveness and agreement with mandatory vaccination were also associated with greater vaccination completeness. These findings are consistent with previous studies indicating that confidence in vaccines and trust in health authorities are important predictors of vaccine uptake among vulnerable populations.9 37 39 At the same time, low levels of active information-seeking behaviour observed in this study may be indicative of informational vulnerability, including difficulties accessing reliable information and limited engagement with institutional communication channels. Structural conditions such as poverty, social exclusion and previous experiences of discrimination may further influence these processes and contribute to reduced trust in health systems and public institutions.40 The strong association between prior COVID-19 diagnosis and vaccination completeness could reflect multiple mechanisms. Experiencing the disease may be associated with increased perceived vulnerability and awareness of the potential severity of infection, as well as greater contact with healthcare services and opportunities for vaccination. However, given the cross-sectional design of the study, the temporal direction of this association cannot be determined. Interestingly, reporting the presence of a PHC unit in the neighbourhood was associated with lower odds of complete vaccination. Although this finding should be interpreted with caution, it may indicate that geographical availability of services does not necessarily translate into effective access for PEH. Previous studies have demonstrated that administrative barriers, stigma, discrimination and unstable living conditions may continue to limit healthcare utilisation even when services are geographically available.6 7 48 This study has limitations that should be considered when interpreting the findings. First, the cross-sectional design does not allow causal inferences, and reverse causality cannot be excluded. Therefore, the identified associations should not be interpreted as evidence of causal relationships. Second, the use of self-reported data may have introduced recall bias and social desirability bias, particularly regarding vaccination status and perceptions related to COVID-19. Third, the absence of a unified national registry for PEH in Brazil may have limited representativeness and affected the external validity of the findings. In addition, recruitment was conducted primarily in capital cities and at service-access points, which may have underrepresented more socially isolated individuals. Despite these limitations, the study provides nationwide data on COVID-19 vaccination among PEH in Brazil and may contribute to addressing an important gap in the literature on vaccination inequities in socially vulnerable populations. Conclusion This study identified structural, social and individual factors associated with completeness of the COVID-19 vaccination schedule among PEH in Brazil. Although vaccination coverage was substantial, gaps in completeness were observed within this population. The findings highlight potential associations between social support, healthcare outreach, trust in vaccines and vaccination uptake among PEH. These results may contribute to the current evidence on vaccination among socially vulnerable populations in Brazil. Supplementary material