Study Design
A large cross-sectional research study was conducted, combining both quantitative and qualitative data collection approaches. The quantitative study included face-to-face interviews with migrant Venezuelan women and the results are presented here. The qualitative results from the focus group discussions and the quantitative and qualitative findings on gender-based violence will be reported separately.
Study Tools
The Minimum Initial Service Package (MISP) readiness assessment tools from the Inter-Agency Working Group (IAWG) on Reproductive Health, adapted for use in Brazil,25 were used in the data analysis(Supplementary material). The Ethics Committee of the University of Campinas, Campinas, Brazil approved the study protocol and all participants signed an informed consent form before being interviewed.
Data collection
The study was conducted in two cities in Roraima, where the principal border crossing points between Venezuela and Brazil are located: Boa Vista, the state capital, and Pacaraima, located at the main land crossing points from Venezuela. Roraima is a small state with a population of 600,000 inhabitants, located in north-western Brazil and sharing borders with Venezuela and Guiana. Boa Vista has 400,000 inhabitants and Pacaraima 17,000. According to the Brazilian constitution, healthcare is considered an obligation of the state and the right of all individuals (nationals, residents and migrants, including non-legally documented persons). Healthcare is provided by the National Health Service (Sistema Unificado de Saúde, SUS) and this is the principal source of healthcare for 75% of the entire population,26. The SUS offers full medical and surgical care within the public healthcare network at no cost to patients, including prescribed medication in accordance with a list of essential medicines, and contraceptives. However, the only long-acting reversible contraceptive (LARC) method available is the TCu380A intrauterine device (IUD).
In Boa Vista, there is only one public maternity hospital providing services in Obstetrics and Gynaecology. Due to the increasing number of migrant Venezuelan women the number of deliveries has increased over recent years. The proportion of deliveries performed for Venezuelan migrants at that hospital increased from 3.4% of all deliveries in Boa Vista in 2016 to 26.1% in 201927,28.
Study participants
This paper is based on analysis of the quantitative data collected from non-indigenous and indigenous women. The indigenous participants were from the Warão and Ñ E Pa tribes of the Orinoco Delta region; however, these participants were all fluent Spanish speakers. Venezuelan women aged 18-49 years, living in the cities of Boa Vista and Pacaraima in Roraima, Brazil, were included in the study. The participants were living in five of the eleven UNHCR shelters based in Boa Vista and in the two shelters based in Pacaraima; however, women living in informal non-UN settlements in Boa Vista who attended the St. Agostinho church to receive free food and goods from a programme supported by UNICEF and Caritas International were also included. Indigenous women not fluent in Spanish and adolescents under 18 years of age were excluded. In Boa Vista, a sample was purposively selected from the five largest UN shelters. These included the shelters with the highest proportion of women aged 18-59 years, and a shelter designated for indigenous persons. In Pacaraima, two additional shelters (including one exclusively for indigenous persons) established by the UNHCR were selected.
A total of 405 women were interviewed, representing 21.1% of the entire female population of the 13 UNHCR shelters in the two cities, with 343 women (84.7%) being recruited from UNHCR shelters (including 47 indigenous women) and 62 (15.3%) from the St Agostinho church. All the interviews were conducted in Spanish between 24 and 30 of November 2019. A team of five trained interviewers (three women and two men) performed the data collection. A pre-tested, semi-structured, electronic questionnaire was used that included both open-ended and closed-ended questions on sociodemographic characteristics (age, race, cohabitation status, education, employment, income, place of residence and migration information), pregnancy and childbirth (births and pregnancies, current pregnancy status, antenatal (ANC) or postnatal care and complications), other SRH issues (family planning preferences, other gynaecological issues), and availability of and access to SRH services, including user satisfaction. Data were entered directly into tablets and saved in a secure online database at the University of Campinas. The data entry system included validation rules to minimise data entry errors during the survey. A unique pre-defined identification number was attributed to each woman.
Migrant women living at UNHCR shelters or elsewhere have access to SRH services free of charge exclusively within the public healthcare network, including primary healthcare units and hospitals. At the UNHCR settlement, a mobile health unit visits periodically to check vaccination and ANC cards and to encourage pregnant women to attend ANC. All women give birth at the only public maternity hospital in Boa Vista. Normally, women in labour are transported to the hospital using transport services paid by the UN, the Brazilian army or other organisations. Reversible contraceptives are available free of charge only within the healthcare system and it is very difficult for migrant women to obtain postpartum or interval tubal ligation. At the UN shelters, a small room is set aside for emergencies and in some cases doctors or nurses affiliated to UNICEF or humanitarian organisations provide some healthcare. In addition, the Brazilian army has health teams and mobile units that periodically visit the shelters, providing primary medical care, including vaccination.
Study sampling
Sample size was calculated to determine whether the SRH needs of migrant Venezuelan women are being adequately met in Roraima. It was assumed that around 50% of the women would not be receiving adequate contraceptive care. Calculation of the population size was based on information provided by the UNHCR, showing a total of 7,233 persons living in the 13 UNHCR-administered shelters, including 1,917 women aged 18-59 years, in the two towns during the data collection period. Consequently, it was estimated that, for a 95% confidence interval (95%CI) (Z=1.96), a margin error of 5% (0.05) and an estimated response rate of 80%, at least 358 women would be required24. This number was increased to 405 women, including 47 from the indigenous population. Since the results were similar for the indigenous and non-indigenous women, the data were pooled together.
Statistical Analysis
Data analysis consisted of simple frequency distribution, using means and standard deviations (SD), as well as bivariate analyses using the χ2 or Fisher’s exact test and the calculation of 95% confidence intervals (CI). When the distribution of the data was not normal, the Mann-Whitney test for two independent groups was used. Univariate and multivariate analysis was performed (with Stepwise criteria of variables selection) in order to identify variables associated to lack of procurement for healthcare services. Significance level was set at p<0.05.