Methods
A cross-sectional study was conducted at the Champ-Dollon prison during two varicella outbreaks, which occurred in the male detention units at two distinct times (2009 and 2011).
Study population and setting
Capitalizing on the blood samples taken to test for varicella immunity in all varicella-exposed detainees, we offered these same consenting participants the opportunity to test for HIV, HCV, and syphilis (HSV-2 was only offered to the 2011 cohort) and to take a structured socio-demographic survey. We offered consenting participants of entire affected prison units the opportunity to participate. Women, representing 5% of the prison population in Champ-Dollon, are detained in a separate unit. Since the varicella outbreaks did not affect this unit, all participants were men. We made sure that detainees were not included twice in the study (i.e., in case of re-arrest). There were no exclusion criteria. Of the 281 detainees, 273 agreed to participate in the study (n = 116 in 2009, and n = 157 in 2011; participation rate = 97.2%).
Ethics statement
The Ethics Research Committee of the Geneva University Hospitals approved the study (EC: 09–137).
Data collection
Each participant underwent serological testing for HIV, HCV, and syphilis. One participant refused the HIV serology. Serum was insufficient to process HSV-2 in one participant and syphilis and HIV in two participants. HSV-2 was tested for only among participants recruited in 2011 (n = 157). Serum samples were tested by commercial immunoenzymatic assays for HIV (HIV, Ag/Ab Combo, Architect, Abbott) and HCV (anti-HCV, Architect, Abbott). Anti-HIV and anti-HCV reactivity were always confirmed via INNO-LIA™ HIV I/II Score and INNO-LIA™ HCV score line immune assays (Innogenetics, Belgium), respectively. HSV-2 serostatus was determined using an HSV-2 IgG enzyme-linked immunosorbent assay (ELISA) (Immunowell, HSV type 2, IgG test, GenBio). A positive HSV-2 serology indicates present or past infection. Syphilis was screened using a two-step Chemiluminescent Microparticle ImmunoAssay (CMIA) (Syphilis TP, Architect, Abbot). Positive cases were confirmed via Treponema Pallidum Particle Agglutination (TPPA) (Serodia®) and the serum Rapid Plasma Reagin (RPR) test (Human). Those CMIA+/RPR+/TPHA+ and without a history of syphilis treatment were assumed to have syphilis. Participants answered a standardized questionnaire administered by a physician. Questions (provided in six different languages: English, French, German, Italian, Spanish, and Russian) explored socio-demographic characteristics (age, sex, country of birth, self-reported socioeconomic status, and education level) and exposure to factors potentially related to STIs and/or BBDs (sexual behavior and past and current injecting drug use). The recall period for sexual behaviors and history of injecting drug use was lifetime.
Clinical management of participants
All participants with confirmed HIV, HCV, HSV-2, and syphilis underwent a clinical evaluation. In cases presenting with HIV and HCV, patients were referred to the Geneva University Hospitals for specialized care. HCV-positive patients whose length of incarceration allowed doctors to consider possible treatment (interferon at the time of the study) benefited from a viremia and a liver biopsy. In accordance with the principle of equivalence of care, the indications for treatment were the same as those used in the general Swiss population. Participants with latent syphilis were treated with intramuscular penicillin (three times once a week). Participants with HSV-2 and recurrent genital ulcers received a prescription for antiviral medication to reduce symptoms during the next recurrent infection or antiviral prophylactic treatment (according to the frequency of relapses). All positive participants were also educated to limit the risk of transmission.
Statistical analysis
Associations between infections (HCV, HSV-2) and other categorical variables (age, gender, region of origin, being an injecting drug user [IDU], and sexual health characteristics) were tested using Chi-square tests. We also computed Cramer’s V to provide an overview of the importance of the relationships for significant associations. The Cramer’s V ranges from 0 (no association) to 1 (perfect association), with .10 being a small association, .30 a medium association, and .50 a high association. Two multivariate logistic regressions that used HVC and HSV-2 as dependent variables and categorical factors as independent variables were performed. Stepwise regressions including variables (i.e., age, gender, region of origin, being an IDU, and sexual health characteristics) used in the bivariate analyses were performed. Odd ratios are reported. For HIV and syphilis, only descriptive statistics were computed because of the low prevalence rate of infected participants (n = 1 and n = 4 respectively). Descriptive and bivariate analyses were performed for the whole sample and then separately for 2009 and 2011 to test whether there were significant differences between the two time points. All analyses were performed using SPSS (version 24).
Results
Table
1 summarizes the socio-demographic characteristics of the all-male participants. The mean age was 29.8 ± 9.0 years. The age range was 18–64 years, with 25% of the participants being younger than 23, 50% younger than 28, and 75% younger than 34. A total of 90.9% were migrants originating from outside Western Europe, and 63.7% were undocumented (no Swiss or European Union passport or residence permit in Switzerland). Overall, 72.1% attended secondary school (among all participants, 50.6% completed secondary school).
Table 1
Socio-demographic characteristics
Sex (male) | 273 (100) | 116 (100) | 157 (100) | – |
Region of originb |
| Central and Eastern Europe | 104 (38.1) | 41 (35.3) | 63 (40.1) |
.001
|
| Sub-Saharan Africa | 77 (28.2) | 37 (31.9) | 40 (25.5) |
| North Africa | 39 (14.3) | 26 (22.4) | 13 (8.3) |
| Latin America | 26 (9.5) | 3 (2.6) | 23 (14.6) |
| Western Europe | 25 (9.2) | 8 (6.9) | 17 (10.8) |
| Asia | 2 (0.7) | 1 (0.9) | 1 (0.6) |
Age |
| < 28 years | 131 (48.0) | 63 (54.3) | 68 (43.3) | .072 |
| ≥ 28 years | 142 (52.0) | 53 (45.7) | 89 (56.7) |
Education level (8 missing values) |
| Secondary school not completed | 74 (27.9) | 78 (69.0) | 113 (74.3) | .340 |
| Secondary school completed | 191 (72.1) | 35 (31.0) | 39 (25.7) |
Self-reported socioeconomic status (8 missing values) |
| Low | 43 (16.2) | 101 (87.1) | 121 (79.6) |
.033
|
| Intermediate or high | 222 (81.3) | 12 (10.3) | 31 (20.4) |
In terms of sexual health, 2.3% of participants reported homosexual or bisexual behavior. More than half (52.8%) reported having had sexual activities with sex workers. Two-thirds (67.0%) reported occasional or no condom use. Almost three-quarters (73.0%) reported having had more than five sexual partners in their lifetime, and 5.9% of the participants declared a history of injecting drug use.
There were some differences between the two years of data collection: participants were more likely to come from North Africa in 2009 and Latin America in 2011, and the participants’ socio-economic level was lower in 2011. There were no differences for other risk factors included in the analyses presented below (history of sexual intercourse with sex workers, being an IDU, and history of genital ulcers).
Table
2 shows the prevalence of infections among participants. Antibody prevalence of HIV, HCV, syphilis, and HSV-2 were 0.4%, 6.2%, 1.1%, and 22.4%, respectively. There were no significant differences between 2009 and 2011. Only HSV-2 was comorbid with other infectious diseases: one participant also had HIV, two had syphilis, and one had HCV. There was no case of co-infection between HIV, syphilis, and HCV.
Table 2
Serological prevalence of HIV, HCV, syphilis and HSV-2
HIV (Ag/Ab Combo+ & Inno-Lia+) | 1/270 | 0.4% (0.1–2.1) | 0/115 (0%) | 1/155 (0.7%) | 1 |
Syphilis (ELISA+ & TPHA+ & RPR+) | 3/271 | 1.1% (0.4–3.2) | 2/114 (1.8%) | 2/157 (1.3%) | 1 |
HSV-2 (ELISA HSV-2+) | 35/156 | 22.4% (16.6–29.6) | – | 35/156 (22.4%) | – |
HCV (EIA+ & Inno-Lia+) | 17/273 | 6.2% (3.9–9.7) | 8/116 (6.9%) | 9/157 (5.7%) | .694 |
A total of 76.5% of participants with HCV were aware of their infection before screening. Among HCV-positive participants who were aware of their status before screening, 92.5% reported being IDUs, whereas 25% of participants who were HCV-positive and unaware of their status before screening reported being IDUs, representing a statistically significant difference (p = 0.04). The HIV positive participant was aware of his infection before screening.
According to bivariate analysis, associations of HCV positivity with a history of being an IDU, originating from European countries, having sexual intercourse with sex workers, and having a low education level were statistically significant. We found an important association between HCV and injecting drug use. The other associations were small (max. = .181) (Table
3). Results showed no association between HCV-positivity and age, self-reported socioeconomic level, non-use of condoms, same-sex sexual activities, and the number of sexual partners in a participant’s lifetime. In the multivariate model, only IDUs were significantly associated with HCV. Participants who were IDUs were 227 times more likely to have HCV compared with participants who were not IDUs (
p < .001).
Table 3
Bivariate and multivariate analyses of HCV and HSV-2 according to sociodemographic and sexual health factors
Region of origin |
| European countries | 14/129 (10.9%) |
.003
a
| | 12/79 (15.2%) |
.028
a
| |
| Other regions | 3/144 (2.0%) | 0.181 | | 23/77 (29.8%) | 0.176 | |
Age |
| < 28 years | 5/131 (3.8%) | .113a | | 10/67 (14.9%) | .051a |
.019
|
| ≥ 28 years | 12/142 (8.5%) | – | | 25/89 (28.1%) | 0.156 | 2.63 |
Education level |
| Secondary school not completed | 1/74 (1.4%) |
.047
b
| | 15/39 (38.5%) |
.009
a
|
.034
|
| Secondary school completed | 16/191 (8.4%) | 0.129 | | 20,112 (17.9%) | 0.214 | 3.21 |
Condom use for sexual protection |
| Always | 3/87 (3.5%) | .165a | | 16/63 (25.4%) | .585a | |
| Sometimes or never | 14/177 (7.9%) | – | | 19/88 (21.6%) | – | |
Sexual intercourse with sex workers |
| Never | 4/125 (3.2%) |
.044
a
| | 13/65 (20%) | .421a | |
| At least once | 13/140 (9.3%) | 0.124 | | 22/86 (25.6%) | – | |
Sexual orientation |
| Homosexual/bisexual | 1/6 (16.7%) | .662b | | 0/3 (0%) | .999b | |
| Heterosexual | 16/259 (6.2%) | – | | 35/148 (23.6%) | – | |
Number of sexual partners |
| 0 to 5 | 3/71 (4.2%) | .369a | | 8/44 (18.2%) | .450a | |
| 6 or more | 14/192 (7.3%) | – | | 25/105 (23.8%) | – | |
Injection drug use |
| Yes | 13/16 (81.2%) |
<.001
b
|
<.001
| 0/7 (0%) | .138a | |
| No | 4/253 (1.6%) | 0.774 | 227.00 | 35/145 (24.1%) | – | |
History of genital ulcers |
| Yes | 2/28 (7.1%) | .695b | | 7/15 (46.7%) |
.045
b
| |
| No | 15/239 (6.3%) | – | | 28/138 (20.3%) | 0.187 | |
Bivariate analyses showed that HSV-2 was significantly associated with education level, region of origin, history of genital ulcers, and marginally associated with age. As for region of origin, 29.8% of participants from Africa and Latin America were seropositive against 15.2% of those originating from European countries. All effect sizes were small (max. = .214), with the highest being between HSV-2 and education level (Table
3). Results showed no association between HSV-2 infection and self-reported socioeconomic level, non-use of condoms, same-sex sexual activities, number of sexual partners in a participant’s lifetime, and sexual intercourse with sex workers. In the multivariate model, level of education and age were significantly associated with HSV-2. Participants with a low educational level (respectively, younger) were 3.21 (respectively 2.63) times more likely to have HSV-2 compared to participants with a higher level of education (respectively, older).
Discussion
This study aimed to estimate the prevalence rate of HIV, syphilis, HCV, and HSV-2 in a sample of male detainees and to investigate associated risk factors. Overall, we found a high prevalence of HIV, HCV, syphilis, and HSV-2. Findings on hepatitis B are reported in a previous study (prevalence rate = 5.9%) [
21].
Prevalence rates of infectious diseases
The prevalence of HIV (0.4%) was two times higher in comparison with the general Swiss population (0.2%) [
22]. However, this prevalence of 0.4% may be seen as low when considering the profile of the detainees (large proportion of migrants and multiple drug use) and in comparison with the prevalence found in other European prisons. For example, a previous study [
23] established that the HIV prevalence in detainees in developed countries ranges from 0.2% in Australia to more than 10% in some European nations. The countries with the highest prevalence rates were in sub-Saharan Africa and Eastern Europe [
12].
The prevalence of HCV (6.2%) was nine times higher in comparison withthe general Swiss population (0.71%) [
24], but 2.5-fold lower than the average (15.5%) reported in prisons in Western European countries [
12].
HSV-2 prevalence (22.4%) was more than two times higher in comparison with the Swiss general population of men of similar age (10.6% among 25–34-year-old participants) [
25]. Studies investigating HSV-2 seroprevalence in prison are scarce. Among male detainees in Italy and Portugal, two studies reported an HSV-2 seroprevalence (21.0% and 19.9%) close to that reported in our study population [
26]. In one prison in New South Wales, Australia, the prevalence of HSV-2 was estimated to be 21% among males [
27].
Finally, the prevalence of syphilis among participants (1.1%) was sevenfold higher in comparison with the general population in European countries (0.16%) [
28]. This high prevalence rate suggested that future studies should explore the cost-effectiveness of systematic screening. Indeed, it is estimated that a third of people infected with latent syphilis will present with significant complications if left untreated [
29]. Syphilis remains a frequent infectious disease in low- and middle-income countries, and its prevalence rate has recently increased in Western countries [
30]. Therefore, screening and treatment of such high-risk populations as detainees would help to reduce this burden of disease.
Risk factors for HCV and HSV-2
We investigated risk factors for the two infectious diseases with a sufficient sample size for analysis. The factor most strongly associated with HCV was the history of drug injection (Cramer’s V = .774). This was also the only factor associated with HCV in the multivariate analysis. This observation was consistent with earlier findings [
31], which showed that the HCV prevalence among detainees was higher compared to the general population due to the high proportion of IDU. The prevalence of HCV was lower among detainees originating from Africa, Latin America, and Asia than among those from Europe, a higher proportion of whom are IDUs. The proportion of migrants from Africa and Latin America, who are rarely IDUs (only 1.2%), is higher among detainees in Switzerland than in other Western European countries; this fact may explain the lower prevalence of HCV in Swiss prisons than in other Western European countries.
HCV-positive IDUs were more aware of their status (92.5%) than HCV-positive participants who did not inject drugs (25%). Therefore, the subgroup of detainees who did not inject drugs should not be neglected. Complementary studies targeting this subgroup are necessary to investigate HCV epidemiology, particularly among participants from countries with high HCV prevalence and where transmission is not predominantly due to injection drug use. Prisons should be a place that offers screenings and treatment for HCV to this vulnerable population, as members of this population are often hard to reach outside the prison. A study has shown that HCV treatment of detainees is cost-effective [
32].
A low education level and increased age were the main risk factors associated with HSV-2 (as well as in the multivariate analysis). These factors corresponded to those identified in the literature [
33]. Region of origin and a history of genital ulcers were significant only in the bivariate analyses. Participants from Africa and Latin America with HSV-2 positivity were twice as prevalent as European participants. According to Looker et al., the global burden of HSV-2 varies by region. The HSV-2 prevalence among 15–49-year-old men is the highest in Africa (25%), followed by the Americas (10%), Southeast Asia (7%), and European countries (4%) [
34]. However, the region of origin was no longer significant in the multivariate analysis. This was probably due to a multicollinearity with socio-economic level, because migrants originating from low-income countries were likely to have a low educational level.
Factors related to sexual risk history did not efficiently discriminate HSV-2-infected detainees, but history of genital ulcers was associated with HSV-2 in bivariate analyses. Nevertheless, 79% of HSV-2-positive participants did not report a history of genital ulcers. This proportion is consistent with data reported in the literature, where only 9–25% of people who are HSV-2 positive report a history of symptoms suggestive of genital herpes [
35]. Among our HSV-2-positive participants, 15% reported recurrent genital ulcers. Four out of five participants with recurrent genital ulcers did not seek medical care because they thought that no treatment was available. These people could be treated with antiviral medication to help reduce symptoms during recurrent infections, while antiviral prophylactic treatment could be prescribed to limit contagiousness and the frequency of relapses [
8]. Moreover, these patients need to be educated regarding their contagiousness during and outside symptomatic episodes [
36].
HSV-2 was also the only sexual infection associated with other diseases. The participant infected with HIV also had HSV-2, as did the two participants who had syphilis. The number of cases of HIV and syphilis was too small to conclude regarding the comorbidity between sexual infections, and future studies should focus on this important research question.
Strengths and limitations
The strengths of our study included the high participation rate and the use of reliable indicators necessary for planning preventive measures. However, this study had some limitations. First, it included a relatively low number of participants and a limited number of positive cases; thus, it is difficult to make a definite statement about the risk factors, particularly for HIV and syphilis. Nevertheless, the sample size was sufficient to identify the burden of BBDs and STIs in the study population, as well as factors associated with HCV and HSV-2. Second, this study focused on men, so data among women are needed, even if women represent only 5% of the total prison population in Switzerland. Third, because participants were asked to answer sensitive questions (e.g., sexual practices), this information may be less reliable due to reporting bias, social desirability [
37], and because the questionnaire was completed face-to-face with a physician. Fourth, for HCV and HSV-2, only serological markers were tested, indicating previous contact with the virus, but not necessarily current infection. Fifth, we used a self-reported measure of socioeconomic status. Future studies should use more reliable ways to assess this important factor. Finally, our findings may not necessarily be generalizable to detention centers in other countries. We hypothesize, however, that the profiles of infections described here would be comparable to other detention centers, where the sociodemographic profiles correspond to those described here. The study population was similar to other pre-trial prisons in Switzerland, which are also characterized by high proportions of migrants and males [
15].
Recommendations
This study highlighted some gaps in the policies designed to fight infectious diseases in Swiss prisons. As the burden of HIV, HCV, and STIs was high in this Swiss prison and because risk factors contributed to a higher risk of BBD transmission among prisoners [
38], effective measures should be improved to mitigate BBD and STI transmission. Effective and internationally recommended strategies, such as opioid substitution treatment, condom distribution, and needle and syringe exchange programs [
39], must be continued in the population study and introduced in other Swiss prisons where they are not yet enforced. Safer tattooing strategies should also be implemented, as tattooing is currently done in a clandestine and unsafe way by using inappropriate handcrafted equipment. Moreover, inmates share these tattoo devices, enhancing the risk of BBD transmission [
37,
40]. Identification of BBDs and STIs must be strengthened in detention facilities in a way that makes it possible to ethically screen as many people as manageable, ensuring individual autonomy and access to treatment [
39,
41].
Overall, we recommend strengthening preventive strategies in correctional settings, especially for detainees with a low education level and from countries with high HSV-2 endemicity [
34]. Region of origin, even if non-significant in the multivariate model, is an important factor when screening for disease at entry to a prison. The socio-economic level is not assessed, whereas the region of origin is. Therefore, this factor might be used as an indicator of the likelihood for the presence of HSV-2. For this subgroup, we also recommend integrating genital ulcers into the participants’ medical history and strengthening preventive measures, such as encouraging people to use condoms.
Acknowledgements
The authors would like to thank the medical and nurses’ team of the prison Champ-Dollon, as well as Agnès Lehmann and Giuseppe Togni (Laboratory Unilabs, Geneva) for their active collaboration.