Adherence to antiretroviral therapy among women living with HIV/AIDS in the interior of the Brazilian state of Pará: cross-sectional study

ABSTRACT BACKGROUND: High prevalence of human immunodeficiency virus (HIV) infection and occurrence of drug-resistant strains have been recorded in northern Brazil. Abandonment of treatment and insufficient and inadequate adherence to antiretroviral therapy (ART) among people living with HIV/AIDS (PLWHA) have been recorded in the metropolitan area of Belém, the capital of the state of Pará. OBJECTIVES: To identify the sociodemographic profile and level of adherence to ART among women seen at a referral unit in the interior of Pará, northern Brazil. DESIGN AND SETTING: Cross-sectional study at a referral unit for care for PLWHA. METHODS: We included 86 women living with HIV/AIDS (WLWHA) in the Rio Caeté integrated region, northeastern Pará. Social, demographic and behavioral information, as well as the ART level, were obtained using forms that have been described in the scientific literature. Logistic regression models were used to assess associations of variables with ART. RESULTS: Most WLWHA were single (52.4%), young (47.7%) and heterosexual (97.7%), had low levels of education (63.0%), were unemployed (69.8%), had one sexual partner (75.7%), used condoms (46.7%) and were not using either licit drugs (68.7%) or illicit drugs (89.6%). Their adherence level was classified as insufficient , and only their viral load showed an association with ART. CONCLUSIONS: The participants’ low level of education and poor socioeconomic conditions may have been interfering with their adherence to ART. Such influences can be minimized through multiprofessional interventions that take the individuality of women served by the healthcare service into consideration.


INTRODUCTION
In Brazil, the epidemic scenario of infection by the human immunodeficiency virus (HIV) and the acquired immunodeficiency syndrome (AIDS) has undergone several changes over time. 1 Currently, involvement of socially more vulnerable populations, non-homogeneous distribution of the disease among Brazilian regions, especially with increased numbers of notifications in small and medium-sized municipalities, and the growing number of HIV-infected women are hallmarks of this epidemic. [2][3][4] These characteristics indicate that the Brazilian healthcare system presents deficiencies with regard to prevention and treatment of HIV infection, especially in municipalities and population groups located in more distant and difficult-to-access areas, as occurs in northern Brazil. 1,2,[5][6][7] Over the last ten years, the northern region of Brazil has shown an upward trend in the rate of HIV/AIDS detection: 16.4 cases per 100,000 inhabitants were registered in 2007 and 23.6 cases per 100,000 inhabitants were registered in 2017 (an increase of 44.2%), with the state of Pará contributing an increase of 55%. 2 In this Brazilian state, high prevalences of HIV infection and occurrences of drug-resistant strains have been recorded among people living with HIV/AIDS (PLWHA) in the cities of Belém and Bragança. 7-10 Abandonment of treatment and insufficient and inadequate adherence to antiretroviral therapy among PLWHA has also been recorded in the city of Belém. 11 In northern Brazil, there are few studies on adherence to antiretroviral therapy (ART) among correlated with abandonment of treatment and with insufficient or inadequate adherence to ART, such as: complexity of therapeutic methods and their respective effects, socioeconomic status, low level of education, family beliefs and values, affective social support, PLWHA's relationships with doctors and other professionals in healthcare services, use of psychotropic drugs and mental disorders. 1,12,14,15 Non-adherence or low adherence to treatment and incorrect use of ART are considered to be strong threats to the effectiveness of treatment among PLWHA. These situations have been directly correlated with therapeutic failure. They facilitate proliferation of HIV strains that are resistant to existing drugs, which gives rise to a need for combined use of other drugs. 14.16 Accurate assessment of adherence and other aspects of this process is essential for proper planning of care for PLWHA and for development of effective strategies for adherence to ART.

OBJECTIVES
The objectives of the present study were to identify the sociodemographic profile of women attended at a referral unit for specialized HIV/AIDS care in the Rio Caeté integrated region, Pará, northern Brazil, and their level of adherence to ART.

Study area
This study was conducted among women assisted at a specialized care service (SCS) in the city of Bragança. This SCS formed a reference unit for specialized care relating to HIV/AIDS in the Rio Caeté integrated region, which is located in the northeast of  (Figure 1). Fishing, agriculture and extraction of natural resources, such as crabs, shrimps, wood and minerals, are the main economic activities developed in this region. Most of these municipalities have low human development indexes (HDI) and a variety of socioeconomic problems, such as high illiteracy rates, informal work and crack use, and more than half of the population is below the poverty line. 7,17,18

Study design and sampling
This cross-sectional study consisted of a convenience sample (non-probabilistic). Hence, participants living with HIV/AIDS were selected when they attended the SCS for medical consultations or medication withdrawal. Specifically, all women aged 18 years or over who had previously been diagnosed with HIV/ AIDS, had been receiving ART for more than three months and were being attended at the SCS of the Rio Caeté integrated region were invited to participate in this study. Women who refused to complete the data-gathering questionnaires, pregnant women  Caeté integrated region, Pará. All of these women were invited to participate in this study. However, 24 women were not included: nine were just starting ART and 15 did not agree to answer the questions of the data-gathering instruments. Therefore, the sample was composed of 86 WLWHA, i.e. 78.2% of the WLWHA attended at the SCS of the Rio Caeté integrated region during the study period.

Data-gathering
Data on the participants were gathered by asking them to answer two questionnaires. The first of these had been adapted from a previous study 19 and was used to obtain sociodemographic, clinical and behavioral information. It contained questions relating to the were obtained from the participants' medical records.
The second instrument was the online version in Portuguese of a multidimensional questionnaire assessing adherence to antiretroviral treatment ("Cuestionario para la Evaluación de la Adhesión al Tratamiento Antiretroviral", iCEAT-VIH). 13 The main issues

Statistical analysis
All the study data were entered into an Excel database  (Table S1).
A general profile of the responses relating to ART that the participants provided in the iCEAT-VIH questionnaire can be seen in only one pill and only 15 participants (17.4%) had had any opportunistic infections in the last 12 months.
The bivariate analysis indicated that only the variable "plasma viral load" was associated with low adherence to ART ( Table 3).
In addition, a multivariate analysis was performed using age, educational level, having a sexual partner, plasma viral load, CD 4 + T-lymphocyte count, time since diagnosis of HIV infection and length of ART use. Again, only the variable "plasma viral load" was associated with low adherence to ART among these women in the Rio Caeté integrated region (supplementary material; Table S2). At the beginning of the HIV/AIDS pandemic, transmission of the retrovirus was marked by unprotected sex in homosexual relationships and among people with multiple partners. 22 According to Bertagnoli and Figueiredo,23 HIV transmission now occurs more    This study had limitations that should be considered. Firstly, the sample size was small and restricted to a single healthcare service, although this data-gathering site is a reference point in assistance for PLWHA in the interior of Pará. In addition, the data-gathering was limited to use of questionnaires. The responses to the questions were statements that were not further investigated, which may have given rise to bias in the information collected. Lastly, scenarios and questions that were misinterpreted by some of the potential participants may have led to their refusal to participate in the study.

CONCLUSIONS
This study was unique and is very important for the epidemio-