Usual source of healthcare and use of sexual and reproductive health services by female sex workers in Brazil

The fact of having a usual source of care can improve access to health services and preventive activities. The article aimed to estimate the proportion of factors associated with usual source of care among female sex workers. This was a socio-behavioral survey with 4,328 female sex workers, ≥ 18 years, in 12 Brazilian cities. Female sex workers were recruited by the respondent-driven sampling method (RDS). A descriptive profile was performed of the female sex workers who had usual source of care and the association was analyzed between usual source of care and indicators of access to HIV prevention and reproductive health. As effect measure, the study used adjusted odds ratio (OR) in a logistic regression model. The data were weighted by the RDS-II estimator. 71.5% of the female sex workers reported having a usual source of care, and of these, 54.3% cited primary healthcare (PHC) as their main usual source of care. Among female sex workers 18-24 years of age, there was an association between usual source of care and having a Pap smear test (OR = 2.27; 95%CI: 1.66-3.12), seven or more prenatal visits (OR = 2.56; 95%CI: 1.30-5.03), and the use of a contraceptive method (OR = 1.64; 95%CI: 1.09-2.46). Among female sex workers ≥ 25 years, there was an association between usual source of care and attending talks on sexually transmitted infections (STIs) (OR = 1.45; 95%CI: 1.12-1.89), prior knowledge of post-exposure prophylaxis (OR = 1.32; 95%CI: 1.02-1.71), and history of Pap smear test (OR = 1.92; 95%CI: 1.54-2.40). The study’s results showed that female sex workers have PHC as their main usual source of care. usual source of care can also positively impact care and activities in reproductive health and prevention of HIV and STIs in this population group. The study showed a high proportion of female sex workers with a usual source of healthcare in Brazil, predominantly in PHC services, the latter consistent with findings in the general population in the last PNS 10 . This phenomenon can be understood as due to the expansion of PHC in previous years as the preferred portal of entry into the Brazilian Unified National Health System (SUS), which is present in the most remote sites in Brazil’s territory 27 , with community health workers in hard-to-reach com-munities 28 and healthcare workers trained in basic women’s healthcare, such as application of rapid tests and counseling for HIV and other STIs, low-risk prenatal care, and Pap smear test 29


Introduction
Usual source of care can be defined as relying on the same health service or healthcare worker in case of a health problem or need 1 . Studies show that usual source of care can reduce costs with hospitalization, increase coverage of preventive and health promotion activities, improve timely access to healthcare, and allow the supply of more equitable care for individuals and populations 2,3 . usual source of care is associated with longitudinal and continuing healthcare, facilitating comprehensive care 4,5 .
In the individual context, studies have reported important results among persons with a usual source of care, such as better self-rated health and experiences with the care provided by services 6 , better quality in terms of accessibility, use of services, and continuity of care, including by primary healthcare services 7 , annual health visits by adolescents 8 , and higher odds of mammogram tests for women 9 .
In Brazil, data from the National Health Survey (PNS) of 2013 estimated that 74.5% of the population had a usual source of care, which was associated with better self-rated health and less use of urgent and emergency services 10 . However, there are no data on usual source of care among more socially vulnerable populations such as female sex workers.
The female sex workers population in Brazil is estimated at 0.8% of the female population 15 to 49 years of age 11 . Compared to women in the general population, female sex workers have greater need of healthcare and disproportionately high rates of specific diseases such as HIV and other sexually transmissible infections (STIs) 12 . The profile of female sex workers is associated with low schooling, worse socioeconomic and work conditions, high-risk sexual practices for HIV infection, and illicit drug use 13,14 .
The most frequent health problems in female sex workers are related to sexual and reproductive health, such as abortion, unplanned pregnancy, situations related to maternal health, and STIs, besides frequent mental health issues resulting from exposure to situations of physical and sexual violence 15,16 . There are also associations with factors that increase structural vulnerability, such as the criminalization of sex trade establishments, housing instability, violence, discrimination, and stigmatization of sex work and in health services 17,18 . This same context involves barriers in access to services for HIV prevention and adequate healthcare 19,20 .
The main barriers in access to health services for female sex workers and reported in the literature are related to sexual and reproductive health, lack of financial resources, shame in seeking health services, discrimination by healthcare staff, and distance from their homes to the services 21 .
Given female sex workers' need for healthcare and the importance of having a usual source of care to decrease the barriers to access, this study aims to estimate the proportion of usual source of care and associated factors among female sex workers in Brazil.

Study design and population
The study's data are from a survey conducted in 4,328 female sex workers in 12 cities of Brazil from July to November 2016. The survey aimed to estimate the prevalence of HIV, syphilis, and hepatitis B and C and to monitor risk practices for these infections. The 12 cities were selected by then-Department of Surveillance, Prevention, and Control of STIs, AIDS, and Viral Hepatitides (DIAHV) Brazilian Ministry of Health, based on geographic criteria and the epidemiological relevance of the HIV/ AIDS epidemic in these cities 22,23 . The study was approved by the Institutional Review Board of the Oswaldo Cruz Foundation (Fiocruz; approval number 1.338.989). All participants signed informed consent, guaranteeing anonymity, voluntary participation, and the possibility of leaving the study at any moment.

Sampling process
The minimum sample size was set at 350 female sex workers for each city, and the women were recruited by the respondent-driven sampling (RDS) method, in which participants themselves recruit their peers 24 .
The first participants, called "seeds", were selected by the researchers to represent the heterogeneity of the female sex workers population, according to demographic and socioeconomic conditions. The seeds were selected after conducting a qualitative formative survey through in-depth interviews and focus groups with female sex workers to establish the study's logistic organization in each city. From 5 to 10 seeds began the recruitment in the study´s cities, and each seed received 3 invitations to invite female sex workers from their contact networks (recruitment chains) 25 . Inclusion criteria were women ≥ 18 years of age who reported at least one sexual relation in exchange for money in the previous 4 months and presented a valid invitation for participation in the study.
For the recruitment's success, RDS requires a system of primary and secondary incentives. The primary incentive consisted of a beauty kit (a toiletry bag containing a mirror, nail polish, lipstick, male and female condoms, and lubricant gel), a snack, and BRL 30.00 (USD 5.00) for transportation. The secondary incentive was BRL 15.00 (USD 2.50) as a form of reimbursement for each person recruited (maximum of three).
Data collection was done in public healthcare units located in the respective cities. Participants answered a questionnaire applied by previously trained interviewers, containing sociodemographic variables, knowledge on STIs, access to social support, access to educational and preventive material on STIs, frequency of testing for HIV, syphilis, and hepatitis B and C, access to and use of healthcare, sexual practices, alcohol and drug use, and history of discrimination.

Variables
Usual source of care was defined according to the following question: "Do you usually go to the same place, the same doctor, or the same health service when you need healthcare?" (yes, no). The health services cited as usual source of care were categorized as: (i) primary healthcare (PHC) (health post or family health unit); (ii) specialized public services (specialized clinics, specialized healthcare centers, polyclinics, and public hospital outpatient departments); (iii) private (private medical offices and physicians, or healthcare workers at the commercial establishment where they work as female sex workers); (iv) public and private emergency departments (urgent care units or other public emergency care units or private hospital emergency departments); and (v) none (has no usual source of care or cites a pharmacy as the source).
The analysis was performed in two stages. The first aimed to know the profile of female sex workers that have a usual source of care, and next those that had a usual source of care C were analyzed according to the characteristics of the female sex workers, considering three dimensions: (i) sociodemographic and sex work characteristics: age (18-24 and ≥ 25 years), skin color (white, black, brown, and other), schooling (primary or incomplete secondary versus complete secondary or more), income (BRL 1.00-500.00, BRL 501.00-1,000.00, and > BRL 1,000.00), main sex work location (commercial establishments and places on the street); (ii) participation in nongovernmental organizations (NGOs) for sex workers' human rights (no, yes); and (iii) violation of human rights: report of discrimination in the healthcare service due to being a sex worker (no, yes) and changes in care from health services when the individual discloses that she is a female sex workers (no, yes).
The second stage analyzed whether usual source of care was associated with greater access to health, considering HIV prevention and reproductive health. The target indicators for HIV prevention were attendance at talks on STIs/HIV in the last 6 months (no, yes), received male condoms in the last 6 months (no, yes), received female condoms in the last 6 months (no, yes), HIV test any time in life (no, yes); knowledge of PEP -post-exposure prophylaxis (no, yes). Indicators of use of reproductive health services were Pap smear test (never/more than 2 years before versus less than two 2 years before), number of prenatal visits (< 7 and ≥ 7) for pregnancy in the last two years, and contraceptive method (except condoms) at the time of the survey (birth control pill, diaphragm, IUD, and injectable contraceptive) (no, yes).

Statistical analysis
The data analysis considered the complex sampling design for recruitment by RDS, that is, dependence between observations resulting from the recruitment chains and the unequal probabilities of selection. The question in the questionnaire that measured the size of each sex worker´s contact network was: "How many female sex workers that work here in the city do you know personally, that is, you know them, and they know you?" The sample was weighted by the RDS-II estimator 26 . The analysis was performed in Stata, version 14 (https://www.stata.com). Details of the weighting procedure were described by Damacena et al. 25 .
To know the profile of the female sex workers that have a usual source of care, we performed a descriptive analysis of the weighted frequency with an estimate of the respective 95% confidence intervals (95%CI). The analysis of association between usual sourcer of care and the target variables used as measure of effect the odds ratios (OR) and respective 95%CI. The final model included variables with p-value ≤ 0.20 in the bivariate analysis. Two multivariate models were performed, stratified by age 18-24 years versus ≥ 25 years, since a difference was observed in relation to usual source of care and age bracket: (i) model with indicators of measures of STI and HIV prevention, and (ii) model with indicators of use of reproductive health services.

Results
Of the 4,328 recruited female sex workers, we excluded 83 seeds from the RDS, since they were picked intentionally by the researchers, and 4,245 female sex workers (98.1%) thus entered the analysis. Most of the recruited female sex workers were over 25 years of age (66.3%), with brown self-reported race/ skin color (52.3%), with complete secondary schooling or more (52.2%), and with a monthly income of up to BRL 1,000.00 (56.2%) (data not shown).
A usual sources of care was reported by 71.5% (95%CI: 69.59-73.29) of the female sex workers, and more than half (54.31%) reported PHC as the principal type of usual source of care. As for indicators of STI and HIV prevention, the majority received male condoms (81.9%) and had undergone an HIV test sometime in life (77.5%), while a minority had attended talks (24.4%), received female condoms (33.6%), and were aware of PEP (31.1%). As for indicators of use of reproductive health services, 51.5% of the female sex workers had undergone a Pap smear test in the previous 2 years, 59.3% reported 7 or more prenatal visits in a pregnancy in the last 2 years, and 53.3% used contraceptive methods (not counting condoms) ( Table 1).
The proportion of individuals with a usual source of care was higher in female sex workers 25 years or older (73.1%), with brown skin color (72.8%), and with monthly income from BRL 501.00 to BRL 1,000.00 (75.1%). As for indicators of STI and HIV prevention, having a usual source of care was more frequent among female sex workers that had attended talks on HIV/STIs in the previous 6 months (77.8%), received male condoms in the previous 6 months (72.6%) or female condoms in the previous 6 months (75.6%), were familiar with PEP (75.9%), and had undergone an HIV test sometime in life (73.1%). In relation to use of reproductive health services, usual source of care was more frequent in female sex workers who had undergone a Pap smear test in the previous 2 years (78.6%), who had 7 or more prenatal visits during a pregnancy in the two years prior to the interview (80%), and who used contraceptive methods (70.7%) ( Table 2).
In the multivariate analysis, among female sex workers 18 to 24 years of age, there was an association between usual source of care and Pap smear test (OR = 2.27; 95%CI: 1.66-3.12), seven or more prenatal visits in a pregnancy in the previous two years (OR = 2.56; 95%CI: 1.

Discussion
The study showed a high proportion of female sex workers with a usual source of healthcare in Brazil, predominantly in PHC services, the latter consistent with findings in the general population in the last PNS 10 . This phenomenon can be understood as due to the expansion of PHC in previous years as the preferred portal of entry into the Brazilian Unified National Health System (SUS), which is present in the most remote sites in Brazil's territory 27 , with community health workers in hard-to-reach communities 28   According to Starfield et al. 3 , the use of PHC as usual source of care translates as better health outcomes. In Brazil, PHC greatly expanded the coverage of basic health services for the population, following the process of re-democratization with the 1988 Federal Constitution, especially for the poorer population and in women's health 30 . These data demonstrate the importance of the mode of healthcare based on PHC in Brazil, the principal provider of preventive services for the Brazilian population 31 . Despite the predominance of PHC as usual source of care among female sex workers, its implementation is not homogeneous across Brazil 32,33 , which poses a challenge in large cities, where home care, community activities, and coverage of this service are still limited 29,34 .
Use of health services depends on individual needs determined by demographic and social characteristics, in addition to health status and the services' accessibility, acceptability, and adequacy 35 . One of the most important demographic characteristics in the determination of health services use is age, a factor that may be heavily associated with overall health status, but other factors include social gender markers, race/color, schooling, income, and social class, which can aggravate individuals' health status by subjecting them to limited conditions for access to health services, as is the situation of many black women on the outskirts of large Brazilian cities, with little access to education, work, and income 36,37 .
The study found that indicators of STI and HIV prevention and use of reproductive health services were better among female sex workers with usual source of care. Besides, indicators of use of reproductive health services were better among younger female sex workers with usual source of care. The majority of famele sex workers in this study had a usual source of care related to PHC. We can thus raise the hypothesis that PHC has a positive effect on these indicators, since one of the reproductive health services offered by PHC is family planning, including the prescription of contraceptive methods, among which hormonal contraceptives. Couto et al. 38 , in a study of famale sex workers, reported that younger women (18 to 34 years) used more hormonal contraceptives when compared to their older peers, probably due to greater access and receptiveness in health services.
These results may also be related to the expansion and intensification of activities and services in HIV/AIDS prevention in PHC since 2011, strategically integrating the network of prevention, diagnosis, and treatment 39,40 . Many of these services are also offered by specialized HIV/AIDS services. Thus, the role of usual source of care related to these specific services in indicators of STI and HIV prevention can be further explored in future studies. In other countries, access by female sex workers to STI and HIV prevention services is more difficult. In Nepal, for example, free distribution of con-Cad. Saúde Pública 2021; 37(10):e00188120 doms is only allowed for married persons, since sex work is illegal, and it is culturally unacceptable for unmarried women to purchase condoms 41 , despite evidence in favor of condom distribution and reduction of HIV incidence among female sex workers and their clients 20 .
The study has some limitations. This is a cross-sectional study that cannot assess the chronological sequence of the events in relation to predictive factors. In addition, although RDS is the best method for accessing hard-to-reach populations such as female sex workers, it is a non-probabilistic sampling process, which can affect the population estimates, especially due to the dependence between observations. To overcome these limitations in the RDS, the study used a robust analysis that considered the complex study design and sampling weight to correct the estimates.

Conclusion
The study's findings have important implications for the healthcare model in Brazil and in other countries, especially those that aim to adopt PHC as the health system's backbone. The expansion and consolidation of PHC can increase this population's access to services and thus allow female sex workers to obtain better follow-up and monitoring of their general health needs, since these women had PHC as the principal usual source of care in this study. In this sense, the services comprising PHC are potential spaces for investment in healthcare activities targeted to female sex workers, aimed at decreasing the vulnerabilities to which they are often exposed.
Female sex workers are one of the key populations for dealing with issues related to HIV. Thus, knowing which health services are used by this group and providing a usual source of care may impact care and activities in reproductive health and HIV and STI prevention in this population. It is thus necessary to invest in strategies to link this population to health services and combat stigma and discrimination related to sex work.   03). Los resultados de este estudio muestran que las trabajadoras sexuales tienen la APS como su principal fuente habitual de atención. Además, la fuente habitual de atención puede impactar la atención y las acciones en relación con la salud reproductiva y la prevención del VIH y las ITS en esta población.