Sociodemographic and Nutritional Factors Associated with Adherence to Antiretroviral Therapy in PLWHA in Benin

Adherence to antiretroviral treatment (ART) is critical for suppression of viral replication, repletion of CD4 cells, avoiding viral resistance, improving immune response and slowing HIV infection progression. This study sought to measure adherence to ART and to determine its associated sociodemographic, lifestyle and nutritional factors in HIV-infected adults taking antiretroviral (ARV). Using a cross-sectional study design, medical records of 991 HIV positive patients who started antiretroviral therapy from July, 1 2011 to June, 3


Introduction
Acquired immune deficiency syndrome (AIDS) is now a manageable chronic illness. To achieve expected maximal results from antiretroviral therapy (ART), patient adherence to ART is critical [1]. The benefits of highly active antiretroviral therapy in the treatment of HIV infection have been well described including viral suppression, CD4 lymphocyte repletion, and reductions in AIDS related opportunistic diseases and death [2]. According to literature review, main factors that affects adherence to ART were social capital [3], stigmatization and discrimination [4; 5], barriers to accessibility to care (distantly health care facilities, costs of care, travel costs) [6], depression condition [7], educational level [8], overburden work load [7]. Factors that limit the success of ART include poor therapy adherence, regimen complexity, viral resistance, drug tolerability and toxicity, therapy costs, and presence of comorbid conditions such as substance abuse and addiction [1].
In Benin, ART began since February 2002 in three sites. In scaling up according to 3 by 5 initiative, Benin national AIDS control program (PNLS) opted for decentralized health centers for HIV/AIDS care [9]. In 2014, 1718, people living with HIV/AIDS (PLWHA) with highly active ART (HART) (6.3%) aged 15 years and over were reported drop out in 2014. Poor adherence severely impacted negatively on effectiveness of HART and was linked with the likelihood of the drug resistance [10]. To improve program performance and quality of services to PLWHA, we need to understand, in view of actual implementing experiences, the factors that influence patient adherence to ART.

Setting
Patients were selected in 46 HIV/AIDS care sites in all parts of Benin.

Study Design and Population
The study was a cross-sectional one. The study population was PLWHA who started ART from July, 1 st 2011 to June, 30 th 2012. The present study is part of the study on PLWHA's one year survival.

Selection of ART Centers and Participants
Forty two (42) ART centers with active queue of more than 100 patients were randomly selected out of all 86 centers available in Benin in 2014. In order to take into account representativeness of sites, four complementary sites with fewer than 100 patients, were randomly selected. PLWHA whose medical records included complete information on adherence to ART participated in the study.
Inclusion criteria were: being HIV positive, start antiretroviral therapy from July, 1st 2011 to June, 30th 2012. Exclusion criterion was being pregnant woman taking ART for preventing mother to child HIV infection.

Study Variables
The independent variables were sociodemographic, lifestyles and nutritional factors. The main dependent variable was the adherence to ART. Adherence is defined as a patient's ability to follow a treatment plan, take medications at prescribed concerning times and frequencies. Adherence measurement in this study was based on the number of missed medication. Good adherence referred to compliance with medical recommendations with optimal regularity (≤ 1 missed medication in a month). Poor adherence meant patient not consistent with medical recommendations (>1 missed medication in a month). Patients whose medical reports pointed out one or more missed doses (forgetting of skipping medication) within a month were considered showing poor adherence [11].

Data Collection
Data were collected from patient medical record and regular patients' data base in each site. Investigators were divided in duo: each had to cover one or many HIV/AIDS care sites according to patients' active file in the site and distance between sites. Investigators were enrolled for two days training to be familiar with study procedure. National supervision of data collection was also set.

Statistical Analysis
Data are collected from regular patients' medical records in each site and were analyzed using SPSS version 20.0. Chi square test was used to determine the association between dependent and independent variables

Regulatory Approvals
The study was approved by Benin National Health Research Ethics Committee before the operational stage beginning. The interviewer explained the purpose of the study and procedures and gained written informed consent before commencing the interview. The participants were also informed that their participation was voluntary and that they could withdraw from the interview/discussion at any time without consequences. The participants were assured that their responses would be treated in confidence and they were assured anonymity through the use of strict coding measures.

Results
A total 991 patient's medical records out of 3295 eligible had complete data regarding main dependent and independent variables. Table 1 summarizes socio-demographic characteristics of participants. The majority of the patients were female (69.6%) and 59.9% aged between 30 and 49 years. The mean age of participants was 36.4 ± 10.6 years. Most patients (66.4%) were married, 48.2% were with illiteracy and 28.0% were housewives. Participants were less involved in social network as 89.0% of them were not member of any association and 242% were living with their partner and infants. Table 1 shows lifestyles and health risk factors in PLWHA in HART including currently smoking and drinking alcohol. Most of participants didn't smoke (98.2%) while almost 10% consumed alcohol and 23.8% were underweight (Body mass index < 18.5). Table 2 shows that participants in stage 2 and 3 of WHO AIDS classification represented 26.8% and 46.6% respectively and 99.4% of participants were taking first line ART.

Medical History and ART Treatment
The mean duration of receiving ART was 14.4 ± 7.6 months. The number of patients who reported being admitted in hospital after being infected with HIV during last 12 months due to poor health or opportunistic infection was 8%. The most common reasons for hospital admission were prolonged fever and chronic diarrhea. Adherence to Antiretroviral Therapy in PLWHA in Benin

Discussion
This study focused on determining the relation of sociodemographic, lifestyle and nutritional factors with adherence to ART among adult people living with HIV taking antiretroviral therapy in Benin. Based on the review of patients' medical records, results indicated that almost one third of the PLWHA showed poor adherence to ART. Adherence to ART in this study was associated with sex, social networking and nutritional status.

Prevalence of Adherence to ART
The result of 69.2% ART adherence among PLWHA in the study is in convergence to those found in India and in Zambia [12;13], but lower than finding reported in China [14], and Vietnam [15]. Proportion of adherence to ART is also lower than those reported in developed countries [1; 16]. Some caution is needed in comparing adherence rates across studies since the methods of measuring adherence (medical records review, pill counts and blood analysis) can affect findings. Assessing ART adherence is difficult as patients may over or underestimate their adherence due to recall bias and the desire to avoid criticism [16].

Factors Associated with Adherence to ART
In the study, proportion of non-adherent in women was lower than proportion of non-adherent in men. In contrast, a study among Canadian cohort of HIV-positive, authors reported that female gender was independently associated with a lower likelihood of being adherent to ART (OR = 0.70; 95% Cl: 0.53-0.93), after adjusting for clinical characteristics as well as drug use patterns measured longitudinally throughout follow-up [17].
In a gender-stratified multivariate analysis study in Montefiore Medical Center's Substance Abuse Treatment Program in the Bronx, New York, authors reported that gender difference in adherence to ART was explained by others factors. Worse adherence in women was associated with problem of alcohol use and active heroin use. In contrast, for men, not belonging to an HIV support group and active crack or cocaine use were associated with worse adherence [18].
In the present study, proportions of being member of an association were not significantly different between men and women. In contrast BMI status distribution differed significantly between men and women (p<0.001). This could partly explain gender difference in adherence to ART.
Being active member of an association was associated to adherence to ART. Social capital effect on therapeutic adherence has been shown by others [3]. In three countries (Nigeria, Uganda, Tanzania), social capital (link with neighbour and carers) and patients social responsibility or accomplishment lead the latter to adequate adherence of ART [3]. In our study, a survey with ARV site leaders has shown that near third (28.6%) of them hasn't take any action of networking PLWHA, nor organizing adherence training for patients, family or community.
In the study, no effect of educational level on adherence was pointed out. In literature, educational level effect on therapeutic adherence was controversial. Hansana et al. [8] reported positive effect of the first on the latter while Getachew et al. [19] have shown that educational status of PLWHA taking ARV was negatively associated with adherence to ART.
Malnutrition which referred to BMI<18.5 Kg/m 2 in the study was significantly associated with non-adherence to ART. Malnutrition was more frequent in the non-adherent (38%) participants than in the adherent (28%) counterparts. This is in line with a quasi-experimental study in Zambia [20], in Uganda [21] which reported that better nutritional status improves adherence to ART. However, this is not in line with the cross sectional study in Ethiopia which reported that nutritional status showed no impact on adherence to ART [22]. This study reported that people with good nutritional status miss ARV doses more than those with poor nutritional status [22]. Difference in studies design may explain these results.
AIDS control programs managers and heath careers can design and implement a set of feasible interventions basing on findings to maximize adherence. Indeed, according to World Health Organization, [23], some feasible interventions have proven their efficacy on therapeutic 74 Sociodemographic and Nutritional Factors Associated with Adherence to Antiretroviral Therapy in PLWHA in Benin adherence: networking PLWHA with voluntary or community workers, pairs support and adherence support groups, ensuring nutritional support to PLWHA taking antiretroviral therapy. At individual level, feasible interventions are SMS reminders, nutritional support, and adherence training for patients, their families and the whole community.
The study has some limitations. It hasn't examined the social representations that patients receive from antiretroviral treatment that can affect adherence to therapy. Also, the study didn't examine ARV stocks shortage and the quality of the relation between health workers and patient which were not included in the present study's objectives.

Conclusions
Adherence to antiretroviral therapy in Benin was suboptimal. Sociodemographic and nutritional factors were associated to adherence to ART. Intensive adherence counseling should be provided to all patients before initiation and during antiretroviral therapy. Health care providers should base their intervention on associated factors identified in the study.

Conflict of Interest
Authors declare no conflict of interest.