Prevalence of High-Risk HIV Transmission Behaviors in Illinois Prisons: A Cross-Sectional
A Report Prepared by the HIV/AIDS Research and Policy Institute at
Chicago State University by
Adedeji S. Adefuye MD, MPH, FRIPH;
Titilayo C. Abiona, MD, FMCPH;
Joseph A. Balogun, PT, PhD, FACSM;
Patricia E. Sloan, EdD, RN, FAAN
This study was funded, in part, by the Illinois General Assembly and the Illinois
Department of Public Health. No portion of this report can be adapted or duplicated
without the written permission of the authors.
Under the African American HIV/AIDS Response Act, P.L. 94-0629 signed into law by
Governor Rod Blagojevich in October of 2005, Chicago State University was mandated
to conduct a study to investigate the correlation between HIV infection and incarceration.
The legislative mandate was based on the beliefs that:
1) inmates engage in high risk HIV transmission behaviors while incarcerated;
2) inmates get infected with HIV while incarcerated;
3) infected inmates transmit HIV to their non-incarcerated sexual/injecting drug
use partners upon their release from prison;
4) high HIV prevalence rates in prison increases transmissibility among inmates and
subsequently their partners once they are released;
5) incarceration increases risky sexual and social behaviors of former inmates compared
to those that have never been incarcerated; and
6) incarceration disrupts sexual relationships in a way that increases risky sexual
behaviors among the incarcerated and non-incarcerated.
Given the multiple factors driving HIV infection within and outside the prison environment,
it is apparent that a single study cannot provide the answers to fulfill the legislative
mandate. To provide the synthesis of information needed to answer the legislative
mandate, three studies were planned.
Phase I Study Objective
A cross-sectional study was designed to determine the prevalence of HIV risk behaviors
among inmates of Illinois prisons.
A cross-sectional design was employed in which a multistage sampling method was used
to select inmates for the study. The correctional facilities were stratified based
on level of security, i.e., maximum, medium and minimum. Subsequently, seventeen (15
male and 2 female) correctional centers were randomly selected. In all, 2,000 (1,500
males and 500 females) inmates were randomly selected to participate based on the
relative population of the correctional facilities. Over-sampling was done in order
to adjust for non-participation because of inmate transfers, participation refusal,
and potentially unusable surveys.
The research questionnaire designed for study Phase 1 sought HIV risk behaviors that
have been reported in the literature (unprotected sexual activity between male inmates,
unprotected heterosexual sex, homosexual rape, sharing injection equipment, and tattooing
using unsterile instruments among prison inmates. The overall average response rate
to questions was 96%
The psychometric properties of the research questionnaire were ascertained prior to
the Phase I study. The questionnaire was administered on two occasions to 89 ex-offenders
within a 2 week interval. The sample was 76.4% African American, 10.1% Caucasian,
1.2% Hispanic/Latinos and 2.3% “other” racial groups. A test-retest reliability of
the four parts of the research questionnaire was established with the Pearson’s product
moment correlation coefficient. The Pearson’s reliability coefficient (r) was 0.71
(p=0.001) for questions related to drug use, 0.78 (p=0.000) for questions related
to tattooing, 0.79 (p=0.03) for questions on body piercing, and 0.74 for questions
related to sexual behavior at 95% confidence level. The Cronbach’s alpha (KR-20) for
internal consistency of items on the questionnaire was 0.803.
Of the 2,000 estimated sample size, 1,819 (1,293 males and 526 females) completed
the surveys, giving a response rate of 91%. The study participants were 30.3% were
Caucasian American, 55.4% African American, 9.1% Hispanic/Latinos, and 5.2% belonged
to “other” groups.
Salient Findings from the Phase 1 Study: Overall, the data suggest that inmates engage
in HIV risk behaviors prior to and during incarceration. The most common reported
behavior of inmates in prison that may constitute HIV risk is tattooing. About 66%
of inmates have tattoos. While a quarter of inmates had tattoos done in prison, most
of them came to the prisons with tattoos. Since instruments for tattooing and cleaning
agents are not allowed in prisons, inmates getting tattoos in prison may contract
HIV through the use of dirty instruments which are not sterilized and are often shared.
Significantly more females than males have tattoos; however males were 15 times more
likely to get tattoos in prison compared to females. Getting a tattoo in prison was
significantly related to how inmates received tattoos prior to incarceration. Inmates
who got their tattoos done by both professional artists and amateur artists were 13
times more likely to get tattoos in prison compared to inmates whose tattoos were
done by professionals only. This finding suggests that individuals who exhibit certain
high risk behaviors in the community continue such behavior in prison thus supporting
the “importation” model that behaviors are brought into prison form the outside world.
Another high risk behavior investigated during Phase I was body piercing. Body piercing
occurred in prison; however, this was an uncommon practice reported by about 3% inmates.
There were no significant differences in the proportion of male and females who got
their bodies pierced in prison. Being 50 years old or older, engaging in other risk
behaviors prior to incarceration (getting a body piercing done by an amateur artist,
and having tattoos) and while incarcerated (injecting drugs, sharing needles and getting
tattoos) were associated with getting a body piercing in prison. This finding may
suggest that inmates who exhibit high risk HIV behaviors represent unique groups who
have multiple risk behaviors. No independent risk factors for incarceration were identified.
The small number of inmates who responded getting their bodies pierced in prison may
account for this finding. Similar to tattooing, body piercing constitutes HIV risk
because instruments for making them and agents for cleaning these instruments are
not readily available in prison.
The third high risk behavior surveyed was drug use. The proportion of inmates reporting
injection drug use while in prison was very small, less than 1%, although 14% of inmates
reported a lifetime use of injecting drugs. Inmates who had been incarcerated for
four years or more, who were Hispanic/Latinos or Native Americans, and those who engaged
in other risk behaviors in prison such as tattooing and body piercing were more likely
to inject drugs in prison. Inmates who were incarcerated in medium or minimum security
facilities were less likely to inject drugs compared with inmates in maximum security
facilities. This finding coupled with the higher risk associated with longer stay
in prison may reflect that inmates with longer sentences and those convicted of more
serious offences are more likely to inject drugs in prison.
Among inmates who had ever injected drugs, more than half had shared needles, while
among inmates who had injected drugs in prison, two thirds had shared needles. In
addition, almost half of the inmates who had ever shared needles did so in the six
months prior to incarceration. Although the proportion of inmates reporting that they
shared needles in prison is very small, this could be an important mode of HIV transmission
if these individuals have other risk factors such as unprotected sex, getting tattoos
and having their bodies pierced. It is possible that the same instruments are used
for tattooing, body piercing and injection drug use thus increasing the risk of HIV
transmission among these people and their sexual partners if they also engage in sexual
intercourse in prison.
Inmates who had been incarcerated more than three times were less likely to report
engaging in sexual intercourse in prison. Most of the inmates in this study identified
themselves to be heterosexual. More females than males responded to the question on
sexual orientation. The reason for this is not clear. Study inmates reported high
risk sexual behaviors prior to incarceration with significant male/female differences.
While more than half of both males and females reported having two or more vaginal
(penile-vaginal sexual intercourse) sexual partners, females were more likely to report
having one sexual partner while males were more likely to report having more than
3 partners. Males also reported more anal sexual partners compared to females. Male
inmates reported having more female sexual partners prior to incarceration compared
with females reporting of male sexual partners. Also females were more likely to report
being homosexual or bisexual compared with men prior to incarceration. There may be
some under-reporting of homosexual or bisexual relationships among the males because
of the lack of acceptance and stigma associated with homosexual behavior especially
among African Americans.
Condom use prior to incarceration was also very low. Over 70% of women and two thirds
of men had never or rarely used a condom. Overall, less than 10% of inmates consistently
used a condom prior to incarceration. About 20% of inmates used condoms during their
last vaginal or anal sexual intercourse. Few differences existed in condom use among
males and females. This low condom use among our study participants is a matter of
grave concern considering the high risk sexual behavior they exhibit. It also suggests
that inmates whether in prison or in the community exhibit HIV high risk behaviors.
Less than 10% of inmates reported having had sexual intercourse in prison. Females
significantly reported more sexual intercourse than men. While females were more likely
to have been involved in oral sex, males were more likely to engage in anal and vaginal
sexual intercourse. Only 7% of inmates who reported having sex said the sex was forced;
males significantly reported being forced compared with females.
Inmates who have spent one year or more in correctional facilities were more likely
to report that they had engaged in sexual intercourse in prison. The odds increased
as the number of years of incarceration increased. While inmates who had been incarcerated
for one to three years were three times more likely to report engaging in sexual intercourse,
the risk was forty times more for those who had been incarcerated for seven years
or more. In addition, inmates who are bisexual were four times more likely to report
engaging in sexual intercourse compared with inmates who identified as heterosexual.
Being incarcerated more times (four times or more) and having ever been married or
currently married were associated with less sexual intercourse in prison. Inmates
who reported being bisexual, lesbian or getting tattoos in prison were more likely
to have sexual intercourse in prison.
This study has highlighted the risk factors for HIV reported by inmates in Illinois
prisons both before and during incarceration. More risk factors occur outside in prison
because these behaviors are not allowed within the prison environment. Both the importation
and deprivation models have been used to describe the behavior of inmates and these
models can explain inmates’ high risk HIV behaviors.
Based on the results of the Phase 1 study, the following recommendations are made:
1. Our result showed that the study participants engaged in substantial HIV risk behaviors
prior to incarceration. Consequently, HIV prevention education and other HIV related
services need to be intensified in the communities outside prison.
2. We found that HIV risk behaviors (tattooing, body piercing, injection drug use
and sexual intercourse) occur in Illinois. From public health perspectives, effective
prevention efforts are warranted. In absence of decriminalizing condom distribution
and needle exchange programs, stricter surveillance needs to be enforced to reduce
these high risk behaviors. Tighter controls are required in the following areas:
A. Illicit drugs getting into the prisons.
B. Inmates acquiring instruments and materials used for tattooing, body piercing
and injection drug use.
C. Sexual intercourse among inmates.
In addition, HIV prevention education should be expanded in the prisons. This has
the potential to reduce HIV risk behaviors among inmates.
3. The proposed Phase II study is urgently required to explore the context within
which HIV risk behaviors occur in prisons. The findings from the focus groups will
provide in-depth information needed to make specific recommendations for HIV prevention
in Illinois prisons.
4. To provide a full understanding of the link between incarceration and HIV infection
in Illinois, Phase III of the proposed research studies will focus on HIV and STD
testing and also monitor HIV risk behaviors of inmates at the point of entry to prison,
discharge, six months and one year in the community. The study will also assess the
HIV risk behaviors of partners of inmates.
5. To determine the impact of community education and HIV prevention efforts, a multi-site
comparative longitudinal study would predict HIV and incarceration for different matched
areas and measure long-term effects of social marketing and “best practices’ educational