HIV Infection in Senegalese Prisons: Prevalence and Associated Factors

HIV/AIDS is a serious health problem among prisoners and constitutes a 
big challenge for prison administration services, public health services and governments. Aims: Determine the prevalence of HIV infection among prisoners and 
describe associated factors. Population and Methods: This was a descriptive 
and analytical cross-sectional study conducted from December 2015 to October 
2016. A sample of 600 prisoners was randomly interviewed nationwide. They were 
selected directly from the sample frame from across the country and 
proportionally from all sites. Pre-established questionnaires providing 
information on the sociodemographic and biological characteristics of prisoners 
were used for the collection of data. Data entry and analysis were performed using EXCEL and SPSS 18 software. Results: The survey was based on a randomised sample of 600 prisoners, 593 of 
whom accepted blood extraction for the HIV testing. The series was mainly 
composed of males (89%). The median age was 33 years (18 - 69). The level of 
education among respondents was 60%. Regarding marital status, the proportion 
of married prisoners was 51.8%. Almost three-quarters (71.5%) were at their 
first imprisonment and 51.8% were on preventive detention. The reasons for 
imprisonment varied and were dominated by drug use (35.3%), while only 4.6% 
confessed that they continued using drugs while in prison. 97.8% of prisoners 
have had a sexual experience. Their median age was 19 years at their first 
sexual intercourse. 20.5% reported using a condom during their first 
intercourse. Multi partnership was estimated 
at 17.8%. The prevalence of HIV infection was high among prisoners (2%), the 
quadruple of the national rate. This prevalence was associated with female sex 
(4.5%) (p = 0.012) and housewives (12.5%) (p = 0.002). Conclusion: Prisoners are very vulnerable to HIV infection with 
a high prevalence compared to the national rate, hence the need for the enforcement 
of effective HIV prevention and care measures in prisons.


Introduction
In many countries, HIV/AIDS is a serious health problem among prisoners. As such, it is a serious challenge for prison administration services, public health services and governments [1]. The vulnerability of groups of prisoners is linked to detention conditions such as the lack of treatment against sexual transmissible infection (STIs), prohibited sexual relations but also the fact that the prisons are places of concentration of people considered as second rate citizens by society such as drug users, migrants and sex workers [2] [3].
Various studies show that the proportion of people infected with HIV in prisons is higher than in the general population [2] [3] [4]. In sub-Saharan Africa, the HIV prevalence was two to fifty times higher among prisoners than in the general population [2] [3] [4]. As evidence, higher prevalence was reported in Côte d'Ivoire (28%), Burkina Faso (11%) and Nigeria (9%) [5]. Jürgens and al reported high prevalence rates of 27% and 41.4% for Zambia and South Africa respectively [2].
In Senegal, the HIV epidemic is of the concentrated type with a high prevalence among key populations. In addition of these populations, the mostly exposed populations to HIV related risks, namely the most vulnerable ones, have been defined. They are composed of young people, prisoners, artisanal gold miners, people with disabilities, truckers, fishermen, soldiers and police [6]. Data on prisoners are scarce and mostly result from combined national surveillance surveys. These surveys have contributed to a better understanding of the evolution of the epidemic in Senegal and to the identification of related challenges and constraints [7]. Our study was conducted as part of the second combined national surveillance survey targeting prisoners. The objectives were to determine the prevalence of HIV infection in prison and to describe associated factors.

Population and Methods
This was a descriptive and analytical cross-sectional study conducted from December 2015 to October 2016. The inclusion criteria were 18 years of age and/or over, detained for more than 6 months and consenting to participate in the investigation. Non-inclusion was whether they have participated in the 2010 com-bined surveillance survey or refused to participate.
A randomised sample was drawn directly from the survey database nationwide. A full list of prisons was used to provide reliable samples for the drawing of primary units from the sample. Secondary units were drawn at the time of collection based on of the lists of prisoners provided by the administrators of the prisons selected in the sample. A total of 15 sites out of 28 were selected in the sample for a total of 600 prisoners. Data were collected based on the questionnaires validated by the National AIDS Control Program (CNLS). These data were collected and entered by a trained and tested personnel for a standardized understanding.
These questionnaires providing information on the socio-demographic characteristics (age, sex, marital status, level of education, profession, nationality, use of condoms, reason for incarceration, sexuality during the prison stay) and biological (retroviral serology results) of prisoners guided the collection of data.
The retroviral serology was performed using immunoenzymatic techniques (ELISA) and confirmed by Western Blot at the reference Laboratory of Bacteriology and Virology of Aristide le Dantec Hospital. Data entry and analysis was done using EXCEL and SPSS 18 software. The proportions were compared by the chi 2 and the Fisher tests with a value of p < 0.05, considered to be significant.

Epidemiological Aspects
A total of six hundred (600) prisoners were interviewed. 89% were male (n = 534) with a sex ratio M/F of 8.09. The median age of the series was 33 years with range of 18 and 69 years. The mostly represented age group was 30 -34 years (20%), followed by 25 -29 years (19%). 18% were between 35 and 39 years old, 9% were between 40 -44 and 45 -49 years old respectively, while those under 25 were 14% and 11% were over 50. Almost all were of Senegalese nationality (91.8%). Nationals from the sub-region were poorly represented. They were mainly nationals from Guinea (2%), Mali (2%), and Guinea Bissau (1.3%). 0.3% of the targeted population was from Gambia and 0.2% from Mauritania.
The proportion of other nationalities was 2.3% of the prisoners. More than half of the detainees were surveyed in the regions of Dakar, Thiès, Louga and Diourbel with 39.7%, 12.8%, 10% and 9% of the respondents respectively.
Over three-quarters of prisoners were poorly educated with 40% without school education and 31.7% had stopped at primary school level. 24% reached a secondary level, and 4.5% reached higher education level. Half of the targeted population was married, most of them under monogamy (41.2%). Singles represented 36.3% against 11.5% of divorced or widowed. The mostly represented profession was that of traders (20%) followed by workers (15%) and liberal profession (12%). Housewives represented 5% of the targeted population but this concerned 47% of female prisoners (Table 1). Advances in Infectious Diseases

Characteristics Related to the Stay in Prison
The reasons for imprisonment were dominated by drugs-related issues (35.30%) or theft (13.5%). Imprisonment for murder was 10.7% of cases. The other rea-sons for imprisonment were related to assault and battery (8%), infanticide (5%), frauds (5.8%) (Figure 1).
The mostly used drug was hashish/Indian hemp. 0.5% of inmates were familiar with heroin injection. During their prison stay, 4.6% of prisoners recognized that they continued using drugs, which was supplied with the complicity of co-detainees in 50% of the cases, and with prison guards in 26% of the cases.
24% of the prisoners preferred to keep the secret when it comes to the drug supply chain in prison. More than three-quarters (71.5%) of the study population were at their first detention and 18.7% at their second detention. Half of the respondents were in preventive detention. The median length of detention was 20 months with range of 7 and 236 months. 26% stayed longer than 24 months, 17% between 19 and 24 months and 5% between 6 and 12 months. More than half of the respondents (55%) said that they momentarily left prison for the following reasons: respond to court (68%), perform professional duties (14%), do shopping (11%) or perform city cleaning duties (7%) ( Table 2).

Sexuality and Risk Practices
Almost all the prisoners were sexually active. The median age at first intercourse was 19 years old (12 -30). 98.5% claimed to know the usefulness of condom but only 20% claimed to have used it during the first sexual intercourse. Multi sexual partnership was found in 17.8% of cases. During their prison stay, only one prisoner recognized having sex with another male prisoner.

Prevalence of HIV Infection
Of the 600 detainees, 593 (527 men and 66 women) agreed to have their blood sample taken, thus a refusal rate of 1%. Retroviral serology was positive with a prevalence of 2%, 1.7% in men (9/527) and 4.5% in women (3/66).

Discussion
Our study involved a randomised sample of 600 prisoners, selected nationwide.
An obvious male predominance was found in our series (89%) with a sex ratio M/F at 8.1. This observation supports the results described in several studies, especially those carried out in the sub-region where the predominance of male population was respectively 95% in Burkina Faso [8] and 75% in Côte d'Ivoire [9]. The median age of our study was 33 years, which testifies that the prison population is mainly composed of the youth. This was also described by Dembele in Mali [10] and by Ouedraogo in Burkina Faso [11].
Almost two-thirds (60%) of prisoners were educated in our series, which is relatively close to the enrollment rate found in a study carried out in the sub-region [12]. Higher education rates were found in prisons in developed countries [13]. The notion of first imprisonment described in several studies in the sub-region [9] [14] was also found in our series with a median imprisonment duration of 20 months. However, due to the slowness of judicial procedures in resource-limited countries, half of our series was in preventive detention, like what was described in Mali [10]. More than half of respondents said that they were temporarily released from prison, which confirms the fact that prisoners do not live in a vacuum. Therefore, they can be a vector for HIV transmission inside and outside prisons [5] [13]. Thus, adequate control of HIV infection in prisons must be considered a public health priority, not only for prisoners, but also for the entire community.
The main reasons for imprisonment within our study population were dominated by drug-related issues but with a lower proportion in comparison to those described in Northern countries [4]. In addition, drug injection was not documented in our series, unlike in the sub-region [9] and elsewhere [15] [16]. This non-documentation may be explained by the fact that the use of drugs is prohibited in our countries and severely repressed by law. That is why a group of prisoners claiming to use drugs prefer to keep the secret when it comes to the drug supply chain.
Though nearly all our target was sexually active and claimed to know the usefulness of condom, its use remained low. The low use of this means of protection was documented in several African studies [9] [11] and testifies the vulnerability of this target with regards to HIV infection. On the other hand, slightly higher rates are observed in Quebec [17] in terms of condom use.
The found prevalence is higher than the one described in Burkina Faso [12] but much lower than the one in Mali [10] and Nigeria [2]. We also notice an increase in this prevalence compared to the one revealed by the last national combined surveillance survey, which was 1.5% [7].
This growing dynamic of HIV prevalence in prisons had undoubtedly motivated the National HIV/AIDS control Program (CNLS) to reconsider the target prisoners by classifying them in the Key populations group after men who have sex with men, sex workers and injecting drug users [6]. This therefore suggests the need to develop awareness-raising actions on HIV among this population but also the necessity to coordinate the management of HIV-positive prisoners care. This is supported by the fact that prisons are suitable sites for HIV testing and treatment, since prisoners are confined in a restricted space and easily accessible for diagnosis, treatment, and health education.
Prevalence was associated with female sex and the profession of housewives. This finding corroborates the feminization of HIV due to several factors including biological, economic, and socio-cultural vulnerability of women [6] [22]. The use of drugs and homosexuality have been described in several studies [3] [23] [24]. This association was not found in our study and can be explained by the prison environment where prisoners are very apprehensive to these kinds of practices considered as criminal offenses in our country.
Our study faced some limitations due to information biais due to the fact that a portion of information collected as homosexuality and use of drugs is undernotified. Another limitation is the tiny sample size of female prisoners, which may pose the problem of generalization. However, despite these limitations, we believe that the results of our study have provided an important overview of the prevalence of HIV in Senegalese prisons and have helped to inform programmatic decisions regarding the targeted detainees.

Conclusion
The prevalence of HIV infection in prisons is high in Senegal, four times the national average. This suggests the need to enforce preventive actions to manage HIV infection in prisons. Moreover, additional studies would be necessary for a better understanding of factors associated with the vulnerability of prisoners to HIV in Senegal.