Prevalence and Factors Associated With HIV in Homeless People

Background Infection with the human immunode�ciency virus (HIV) compromises the immune system of infected people. Homeless people are more vulnerable to HIV infection due to behavioral and sexual factors. The main objective of this study was to estimate the prevalence of infection by the Human Immunode�ciency Virus (HIV) in homeless people in a capital of Northeastern Brazil. Methods Cross-sectional study carried out at the Reference Center for the homeless population (Centro Pop), in a capital of Northeastern Brazil, from September/2017 to September/2018. Data collected through a semi-structured form including socio-demographic data, risk and sexual behaviors and anti-HIV serology. Data were analyzed using Person's Chi-square test. The study was approved by the Research Ethics Committee (CEP) of the Federal University of Piauí.


Introduction
Infection with the human immunode ciency virus (HIV) compromises the immune system of infected people and is responsible for the development of the Acquired Immunode ciency Syndrome (AIDS), representing a great challenge, a matter of concern for public health worldwide 1 .According to the Joint United Nations Program on HIV (UNAIDS), in 2018, 37.9 million people worldwide were living with HIV, with 1.7 million new infections 2 .
Given the magnitude of the problem, homeless people are more vulnerable to HIV infection, considering sexual risk behaviors and substance use 3 .Worldwide, the prevalence of HIV is different in relation to the population and risk behavior.
An international study carried out in Jamaica found an HIV prevalence of 13.8%, with a gender difference (men 11.6%, women 26.7%, P = 0.007) 4 .Another study carried out with a homeless child in Iran showed an HIV prevalence of 4.5% 5 .In homeless people, the prevalence was 4.1% for HIV in Kenya, with factors associated with both anal insertion sex and receptive anal sex 6 .
In Brazil, 43,941 new cases of HIV and 37,161 cases of AIDS were diagnosed in 2018.Between the years 2007 to June 2019, with regard to age groups, it was observed that the majority of cases of HIV infection were found in people with 20 to 34 years old, with a percentage of 52.7% of cases 7 .In the Brazilian scenario, there are also differences in prevalence in different regions in the general population and in homeless people.
Still, Brazilian regions in 2018, presented variable rates of HIV infection with 5,084 (11.6%) in the North region, 10,808 (24.6%) cases in the Northeast region, 16,586 (37.7%) in the region Southeast, 7,838 (17.8%) in the South and 3,625 (8.2%) in the Midwest 7 .In the central-western region of Brazil, an HIV prevalence of 1.24% was detected, with a signi cant association in relation to previous HIV tests and years of schooling 3 .In the southeastern region, in a similar population, the prevalence was 4.9%, with the majority being male (85.6%), non-white (71.5%), with a mean age of 40.9 years and low education (72.0%) 8 .
In this sense, authors emphasize that homelessness is a substantial barrier to consistent and recommended care for HIV, access to and adherence to antiretroviral therapy (ART) and sustained viral suppression, thus increasing the risk of morbidity and transmission 9 .
Thus, homeless people have multiple factors and conditions of risk of HIV infection.There is still a shortage of data in the literature regarding the prevalence of this infection in the national and local scenario, and it is necessary to obtain more information about this population in order to implement interventions, aiming at preventing HIV and reducing diseases.
Given this, the study aimed to estimate the prevalence of infection by the Human Immunode ciency Virus (HIV) in homeless people, in a capital of Northeastern Brazil.

Type of study
Cross-sectional, analytical research carried out with homeless people attended at the Specialized Reference Center for the Homeless Population (POP Center), in the municipality of Teresina, Piauí, between September 2017 and September 2018.

Population and sample calculation
Convenience sampling refers to the selection of participants available at the time and place of the survey 10 .In this process, the following inclusion criteria were adopted: being 18 years old or older; being registered at the Pop Center and with at least one service in the Pop Center, being homeless people, or receiving support from hostels, and not showing aggressive behavior at the time of the interview.Thus, participants who did not accept rapid HIV testing were excluded.According to information from the Pop Center, an average of 500 homeless people sought some type of assistance at the Pop Center.In the year prior to the data collection, a sample calculation was performed and an estimate of 200 participants was obtained.

Data collection and study variables
Data were collected using a semi-structured form, previously validated in terms of form and content, by researchers specialized in the subject.The independent variables were: sociodemographic (age, time as homeless people, sex, years of schooling, self-reported color, religion, if he/she has a partner, region of birth, profession, income, if he/she was in a support home, if he/she suffered discrimination, experience in prison, reason why you are homeless people, where you were homeless people, type of homeless experience), risky behavior (use of alcohol, use of cigarettes, use of marijuana, use of crack, time using drugs, sharing injection syringe) and sexual behavior (sexual orientation, already had sexual intercourse, age at rst sexual intercourse, condom use at rst sexual intercourse, xed partner, eventual partner, relationship with people of the same sex, coerced sexual intercourse, sexual intercourse with people who use drugs, sexual intercourse with a sex worker, he/she paid in exchange for sex, received money in exchange for sex, had previous sexually transmitted infection -STI), as a dependent variable, the anti-HIV reagent.
In this study, a case of HIV was considered by means of a rapid reagent test, one in which two lines/bars of purple/pink color, one in the control area (C) and another in the test area (T), appear on the device after 15 minutes of the test.It should be noted that all the guidelines of the Ministry of Health were followed for rapid testing by the method of immunochromatography, with a view to the detection of antibodies to HIV 11 .
Data collection took place in a speci c room at the POP Center to ensure the con dentiality and anonymity of the information obtained.It is also noteworthy that it occurred in two stages, with the rst application of the form and, subsequently, rapid HIV testing.It is noteworthy that participants were properly explained about each stage and, after their consent, proceeded with data collection.
In this context, it is important to note that all participants were informed about the results of the exams.
During the delivery of results, guidance on harm reduction measures was carried out.The HIV-positive serological results were sent to the Specialized Assistance Service (SAS), of the municipality, previously agreed with the local health management.

Statistical data processing and analysis
Data were entered twice in a Microsoft O ce Excel for Windows 2007 spreadsheet and after validation they were exported to the IBM® Statistical Package for the Social Sciences (SPSS) 21.0 software.To check relationships between variables, Person's chi-square test was used, in which p < 0.005 was considered statistically signi cant.

Ethical and legal aspects
The study was carried out following the rules that regulate research with human beings, contained in Resolution 466/12 of the National Health Council and Operational rule 001/2013.Thus, the research was conducted with individuals who agreed to participate voluntarily after reading, explaining and signing the Informed Consent Form.
The present study was submitted to the Research Ethics Committee of the Federal University of Piauí (UFPI), with authorization: 1,755,893.

Discussion
The prevalence of HIV in homeless people was 3.9%, considered high when compared to other studies at an international level.A survey of homeless people in the United States of America found an HIV prevalence of 1.52% 12 .In this context, the HIV prevalence found in homeless people reveals high rates, varying from 9.8-61.1% in different regions of the world: USA (North America), Greece (Europe), Ukraine (Eastern Europe) 13,14,15 .Other studies are in agreement with these ndings, with a high prevalence of HIV in a similar population 5,4,6 .
In Brazil (South America), a study carried out with 564 individuals showed that 47.1% were homeless and with an HIV prevalence of 6.8% 16 .A survey carried out in central Brazil found an HIV prevalence of 1.24%, in homeless people.
It is possible to verify, therefore, that this is a worldwide problem, with high magnitude and data related to prevalence, with great variation in accordance with each region and there are still few studies related to the follow-up of homeless people with HIV.
In this study, there was a signi cant association with the prevalence of HIV in homeless people and sexual practice with people of the same sex p ≤ 0.012 and previous STI p ≤ 0.002.In this scenario, a survey carried out in southeastern Brazil, with 1405 homeless people, also concluded that history of STI, the fact that the person is younger and has a homosexual practice were associated with HIV infection 8 .Some authors in the United States of America have shown that intimate partner violence, the use of injectable and non-injectable substances, a history of incarceration and the severity of homelessness are associated with a higher prevalence of STIs in homeless people 17 .Another national study points out that previous HIV tests and years of schooling were factors associated with HIV infection 3 .
Thus, a survey in Jamaica showed that sex work, multiple partners, incarceration, the use of noninjectable drugs and female rape were associated with HIV among homeless people 4 .Thus, certain authors point out that homelessness was associated with less retention of HIV assistance, adherence to the dose of ART and sustained viral suppression 9 .
In this perspective, it is essential to expand mass testing for STIs, especially in populations at risk.In addition, a contextualized approach of homeless people is necessary in order to detect, treat and decrease the chain of STI transmissibility, as well as providing better health care for this population in need of care networks.
In this context, homeless people are young adults, with little education, mostly male and have no income.
In this regard, a survey conducted in Los Angeles, United States, with homeless people, corroborates these ndings 19 .In the Brazilian scenario, a study carried out in Ribeirão Preto, southeastern Brazil, with homeless people, showed that the subjects reported being literate, however, all had low education, with the fundamental level of regular education being the period in which occurs school dropout 19 .
Men are generally more vulnerable, as they are more likely to suffer injuries related to violence and accidents when compared to women, and are less likely to seek health services.These results suggest that the male population is more exposed to various injuries, considering the street situation, as well as the imperative need for educational quali cation and social programs for income-generating activity.
Homeless people have a long period of stay in this situation.A survey carried out in Tehran corroborates these data by nding that the average number of homeless people was 24 months 21 .Another study in Kenya corroborates these ndings 6 .A survey conducted in six Brazilian capitals showed that the subjects surveyed lived on the street, during their lives, at least once for less than six months, with less time of 30 days.In Greece, the minimum time was 12 months 14,16 .
Authors point out that the high risk of homelessness among HIV-positive homeless people is largely attributed to multi-morbid risk factors common to HIV and homelessness, rather than to an HIVindependent effect.In addition, it requires multidimensional preventive psychosocial interventions 22 .
The risk of HIV diagnosis has decreased with the placement of longer-term support homes.Supportive housing can help with primary HIV prevention 23 .Thus, there is an urgent need for the creation and expansion of social houses, shelters and social support networks to assist this population, a victim of social exclusion and homelessness.
The consumption of alcohol and other illicit drugs, as well as the sharing of syringes for drug use, were common practices among homeless people.Research in Kenya with homeless youth showed that 49% consumed alcohol at least weekly and 32% consumed marijuana 6 .Another study pointed out that of the homeless people in Tehran who abused intravenous drugs, 48.25% had a history of needle sharing 21 .
In Greece, the probability of infection was found to be 2.3 times higher in homeless people who inject drugs, and among users who injected drugs 2 years ago or less, the estimated incidence was 23.4 new cases of HIV 14 .Thus, it is necessary to mitigate public policies for harm reduction in this population and the strengthening of support networks, such as the psychosocial care centers for alcohol and drugs, with an emphasis on the reception, screening, diagnosis and treatment of this public.
In this context, homelessness signi cantly affects health and well-being.Homeless adults generally experience co-occurring and debilitating physical, psychological and social conditions.These determinants are associated with disproportionate rates of infectious diseases among homeless adults, including HIV, viral hepatitis, as well as tuberculosis 17 .
As a limitation of this study, it is highlighted that for the recruitment of participants, the method adopted was the selection of a non-probabilistic sample for convenience, except the aspect that the results obtained had restrictions, since they were collected in a single capital of the northeast and, therefore, do not show a broad generalization.

Conclusion
Homeless people have a high prevalence of HIV infection, with a signi cant association regarding sexual intercourse with people of the same sex and previous STI history.They are young adults, with low education, without xed income, long time on the street, who generally experience risky behaviors, such as high consumption of alcohol and other drugs, low frequency of condom use, in addition to an increased number of partners sexual.
These ndings suggest better investment in prevention activities, early detection of HIV and health promotion, with a view to improving access for monitoring detected cases and reducing a late diagnosis.

Table 1
Prevalence and sociodemographic factors associated with HIV in homeless people.
*p-value of Pearson's chi-square test.

Table 2
Prevalence and behavioral factors associated with HIV in homeless people.

Table 2 .
In this sense, an HIV prevalence of 3.9% (95% CI: 1.93-1.99)wasdetected in a population of 204 homeless people, with statistical signi cance in relation to sexual practice with a person of the same gender p ≤ 0.012 and previous STI p ≤ 0.002, according to Table2.