Cross-sectional study of HIV prevalence and the characteristics of injecting drug users in Colombia

Abstract

Objective: To determine socio-demographic characteristics, prevalence and risk behaviors of HIV in injecting drug users in three cities of Colombia.
Method: A cross-sectional study was conducted in three cities of Colombia; information was obtained from 796 participants over 18 years of age, with prior signed informed consent. Statistical analysis of the data and generated output tables were conducted in RDSAT and SPSS.
Results: There are extensive networks of injection drug users (IDUs). The population of IDUs was characterized as mostly men between 18 and 34. The data suggest a recent introduction of HIV into networks and a high degree of risk behavior for HIV spread in networks and used syringes. People who reported sharing syringes, were at greater risk of not using a condom when having sex with casual partners, this factor is increased when controlling for other variables consulted (OR = 4.10, 95% CI 1.23 to 16.05; p
Conclusions: The data in this report indicate a high risk for a possible expansion of HIV among injecting networks, supported the introduction of HIV into networks of injectors. For Colombia, this research constitutes a first step in the search for strategies to prevent the further spreading of the infection.

Full text

http://informahealthcare.com/jsu
ISSN: 1465-9891 (print), 1475-9942 (electronic)
J Subst Use, Early Online: 1–4
!2013 Informa UK Ltd. DOI: 10.3109/14659891.2013.824037
ORIGINAL ARTICLE
Cross-sectional study of HIV prevalence and the characteristics of
injecting drug users in Colombia
Dedsy Berbesi, Angela Segura, and Liliana Montoya
School of Medicine, CES University, Medellin, Colombia
Abstract
Objective: To determine socio-demographic characteristics, prevalence and risk behaviors of HIV
in injecting drug users in three cities of Colombia.
Method: A cross-sectional study was conducted in three cities of Colombia; information was
obtained from 796 participants over 18 years of age, with prior signed informed consent.
Statistical analysis of the data and generated output tables were conducted in RDSAT and SPSS.
Results: There are extensive networks of injection drug users (IDUs). The population of IDUs was
characterized as mostly men between 18 and 34. The data suggest a recent introduction of HIV
into networks and a high degree of risk behavior for HIV spread in networks and used syringes.
People who reported sharing syringes, were at greater risk of not using a condom when having
sex with casual partners, this factor is increased when controlling for other variables consulted
(OR ¼4.10, 95% CI 1.23 to 16.05; p50.00).
Conclusions: The data in this report indicate a high risk for a possible expansion of HIV among
injecting networks, supported the introduction of HIV into networks of injectors. For Colombia,
this research constitutes a first step in the search for strategies to prevent the further spreading
of the infection.
Keywords
Addiction dependence, HIV, injecting
History
Received 29 April 2013
Revised 17 June 2013
Accepted 3 July 2013
Published online 14 August 2013
Introduction
HIV transmission remains a major public health challenge.
Worldwide, 34 million people were living with HIV at the end
of 2011 and an estimated 0.8% of adults aged 15–49 years
have been diagnosed with HIV (World Health Organization,
2013). In order to prevent further contagion, behavioral
surveillance of people at risk of HIV infection remains an
important component of an HIV surveillance program
(Mathers et al., 2010).
Intravenous drug use has been a very important risk factor
from the beginning of the AIDS epidemic owing to the fact
that the prevalence of HIV infection is higher among people
who inject drugs than among the general population. In
countries with available data (UNAIDS, 2012), the prevalence
of HIV infection is at least 22 times higher among people
who inject drugs than for the whole population. In 2007, it
was estimated that about 16 million people inject drugs
around the world, from which 3 million are living with HIV.
HIV epidemics among injection drug users (IDUs) differ
from those from other risk groups, because they spread
rapidly within this community and from there to the general
population (OMS, UNODC, & UNAIDS, 2012). Due to their
sexual practices, IDUs are at greater risk of acquiring sexually
transmitted diseases, especially HIV. Women also have a
higher tendency to be immersed in risky sexual behavior such
as exchanging sex for money or drugs, having multiple sexual
partners, sharing needles during sex or practicing unprotected
sex with an IDU partner (Ross et al., 2008).
Among IDUs, mainly young males, the main risk factor that
increases the likelihood of transmission is sharing syringes and
needles and there are reports of infection by this route up to
43.5%. Women are also affected by this risk factor up to 31%,
as they are more likely to be involved as recipients after
injecting their sexual partners. This is particularly common in
developing countries (Ross et al., 2008).
Further transmission of HIV in this population is
influenced by socio-cultural factors such as a low capacity
to make informed decisions, limited access to education,
family and social problems that generate psychotic behavior,
unemployment and economic difficulties (Deiss et al., 2008;
Des Jarlais & Friedman, 1987).
During the last two decades, a steady and progressive
increase in the availability of heroin in Colombia has
established the need for containment of a potential HIV
epidemic due to intravenous drug abuse. Therefore, identifying
risk behaviors for HIV transmission among IDUs in Colombia
constitutes a first step in the search for strategies to prevent
further spreading of the infection. The objective of this
research was to identify social characteristics, demographics
and risk behaviors of injecting drug users in three cities in
Colombia.
Correspondence: Professor Dedsy Berbesi, School of Medicine, CES
University, Medicine, Calle 10 A Nro 22-04, Medellin, Colombia.
E-mail: dberbesi@ces.edu.co
J Subst Use Downloaded from informahealthcare.com by HINARI on 08/21/13
For personal use only.

Methods
Survey design and population
A cross-sectional study was conducted in three cities in
Colombia. Information was obtained from 796 participants
over 18 years old, with prior signed informed consents. These
three main cities were selected because of their high number
of injecting drugs users and their known status as micro-
traffic epicenters.
The ‘‘Respondent Driven Sampling’’ (RDS) method was
used to evaluate risk practices and HIV prevalence among
injecting drug users given the fact that RDS has emerged as
an alternative to sample hidden populations (McCreesh et al.,
2012; Paquette et al., 2011). Sample size for each of the
selected cities was calculated with the formula for the
proportion with a 3% ratio taking into account the prevalence
of HIV in Colombian population. We had a 1 a¼95%
confidence level, a 3.5% error, and a design effect of four with
a total of 240 IDUs from each city.
Data collection
Recruitment began with a group of ‘‘seeds’’ or non-
randomized participants from the target population following
the previously described methodology. Briefly, in this study
three initial seeds were determined for each city. From these
seeds, the recruitment chains began, ending when the final
sample of IDUs was obtained. The sample size was achieved
within 8 weeks. Two types of incentives were given to each
participant: a primary incentive for taking part in the study
(US$ 4) and a secondary one, linked to the successful
recruitment of three new participants (US$ 5).
Participant data were collected through a survey technique
with a designed form that included questions about demo-
graphic characteristics, risk behaviors, HIV knowledge, and a
test for HIV. When an HIV test result was positive, both the
participant and the local health authority were immediately
informed. Ethical considerations and informed consent were
approved by the ethics committee of CES University.
Statistical analysis
A description of IDUs characteristics and HIV prevalence
determination with 95% CI were initially performed.
Contingency tables were designed to establish an association
between independent risk factors and dependent variable
condom use. This analysis was based on statistical tests (Chi
square and Fisher’s exact test), OR and 95% CI.
Multivariate analysis was performed by calculating a
binary logistic regression model (explaining the use of
shared needles) calculating crude OR and adjusted OR after
variables with p50.25. All tests were two-tailed and used
p¼0.05 as the cutoff for statistical significance. All statistics
were performed using SPSS software (version 21.0; SPSS
Inc., Chicago, IL).
Results
A total of 796 IDUs were interviewed, from which 92% were
men, 79.4% were single, 78.6% were under 30 years old, and
65.9% had complete primary school education. The age range
was from 18 to 60 years (mean 26.6SD 7.9). By sex, the
average age for men was 26.3 years and for women 29.3 years.
In terms of socio-economic status, 62.5% had a low income
status, 32.9% middle tier, and 4.6% belong to a higher
stratum. Estimated HIV prevalence for the studied population
was 2.6%, for men and 3.1% for women. There was no
statistically significant association between sex and HIV
prevalence (OR 2.57; 95% CI 0.87–7.58). The predominant
marital status was single (88.3%), followed by having a
partner (11.7%). No statistically significant association was
found between HIV prevalence and marital status (OR 1.20
95% CI 0.34–4.17).
Most study participants were recent drug users, with an
average of 4.5 5.0 years of use. Currently used drugs
included cocaine base (56.9%), weed (85.4%) and cocaine
(58.4%). Alcohol abuse was reported by 68.1% of the study
population and all participants were intravenous heroin users.
The current number of sexual partners varied between groups,
where 56.9% of the participants reported having had sex with
casual partners in the last 6 months, 27% reported having had
between 2 and 10 sexual partners, and 4% reported having had
more than 10 sexual partners in the past 6 months.
Consistent use of condom was indicated by 19% of
participants and those with multiple partners had a similar use
of condoms during vaginal sex (15%). Overall 33% of the
study population related they had never used a condom. Also
47% of IDUs had used syringes received from others during
the past 6 months. Cleaning of such syringes was usually
made with water (63.7%), followed by alcohol (23.0%). Basic
demographic and sexual behavior characteristics of partici-
pants discriminated by the risk behavior of sharing needles
are displayed on Table 1.
Participants who shared needles, had twice the risk of not
using condoms with a regular partner (OR ¼1.77, 95% CI
1.13–2.79, p50.00). In all, 27% of IDUs who shared needles
had sex with other IDUs. People who reported sharing
syringes, were at greater risk of not using a condom when
having sex with casual partners. This factor is increased when
controlling for other consulted variables (OR ¼4.10, 95% CI
1.23–16.05, p50.00). After running the logistic regression
model, adjusted ORs were found for the risk factors
associated with syringe sharing in IDUs (Table 2).
Discussion
In Colombia, IDUs are not prioritized as a population at risk
for HIV transmission. This is one of the few studies conducted
on IDUs practices in the country and the only work performed
in three cities with the same methodology. This study
identified a significant number of IDUs who have risk
practices such as sharing needles and improper use of
condom, with both being known risk factors for HIV. For
several decades, in other countries, these factors have already
been widely documented and discussed (Bravo et al., 2004),
but in Colombia, this research constitutes a first step in the
search for strategies to prevent further spreading of the
infection.
In the present study, the IDUs demographic profile was
very similar to the ones from previous studies (Cobos Calleja,
et al., 2003; Mendez et al., 2009). It consisted of young
people under 34 years old, single males, whose homes were
2D. Berbesi et al. J Subst Use, Early Online: 1–4
J Subst Use Downloaded from informahealthcare.com by HINARI on 08/21/13
For personal use only.

located in low socio-economic status. These drug users,
according to some authors (Bravo et al., 2004), start their drug
use by consuming other not injected substances and then
become IDUs. Essentially, this population is predisposed to
use all kinds of substances (Castan
˜oPe
´rez & Calderon
Vallejo, 2010).
Currently, in many countries due to the availability of
resources, drug users have adopted several preventive con-
ducts in order to reduce the harm associated with direct
intravenous drug use (Wodak & McLeod, 2008). However,
conversely, HIV transmission through sex has increased.
Colombia has not initiated any significant change in risky
IDU practices, and there are no harm reduction programs,
compared with other countries that have been working on this
issue for the last 20 years. Using scientific knowledge and
experience about epidemics in injecting drug users, which
were collected during recent decades, constitute a unique
opportunity to prevent the rapid spread of HIV among the
IDU networks (OMS, UNODC, & UNAIDS, 2012).
Individual sexual behavior and condom use are key
determinants for being at risk of contracting sexually
transmitted infections. When this knowledge is applied to
the population, programs for HIV/AIDS prevention can be
designed and implemented (USAIDS, 2010). This research
allows for the orientation of the educational actions that must
be reinforced, and for strategies that allow its application in
difficult situations, such as that of IDUs.
Similar to the findings in the present study, in a research
conducted in Sao Paulo (Brazil, 2012), the increased risk
profile for HIV was associated with inconsistent condom use
and the failure to adopt safe sex practices and unfavorable
conditions in the social and economic environment
(Grangeiro et al., 2012). In the three cities of Colombia,
IDU population behaved similarly to what was reported in
Spain (March et al., 2007). In Colombia, the proportion of
consistent condom use with occasional partners was 21.5%
and 14.6% with regular partners, and in Spain, it was 22% and
15%, respectively.
This study and others around the world about IDUs have
observed that women have a higher prevalence of HIV
infection (Oliveira & Paiva, 2007). Given the fact that females
are prone to work in prostitution, to have sex under the
influence of substances, and to be in a submissive position
that does not allow them for safer sex negotiation, the problem
of HIV transmission increases and becomes an even more
complicated public health problem (Mendez et al., 2009).
The trend of IDUs of having more than one high-risk
behavior, such as having multiple casual partners, increases
the likelihood of HIV infection (Amirkhanian, 2012).
Evidence indicates that sexual transmission between injectors
and non-injectors plays an important role in the global spread
of HIV (Khan et al., 2009). Thus, drug users work as a bridge
which facilitates the spread of blood borne diseases from IDU
population to the general population (March et al., 2007).
It has been considered that IDUs not infected with HIV
have implemented prevention strategies in circumstances
where other IDUs have risk practices, such as sharing of
syringes (World Health Organization, 2013). This shows that
IDUs need prevention efforts, as the condition of not being
infected is not a coincident situation. It is worthy of notice
that in the present study, risk practices related with sharing
needles were observed, but with a low prevalence of HIV.
This suggests that there are strategies and tactics that can help
prevent HIV in this specific population. Nevertheless, the
mechanism by which IDUs remain unaffected by HIV is still
unknown.
Table 1. Socio-demographic and sexual behavior characteristics of
participants discriminated by the risky behavior of sharing needles.
Shares
needles
Does not
share needles
Variable N%N%
2
Significance
Sex
Women 26 7.1% 36 8.7% 0.68 0.40
Men 339 91.3% 376 92.0%
Lives with parents
No 251 68.8% 243 59.0% 7.96 0.00
a,b
Yes 114 31.2% 169 41.0%
Lives alone
No 225 61.6% 301 73.1% 11.44 0.00
a,b
Yes 140 38.4% 111 26.9%
Age
530 290 79.5% 322 78.2% 0.19 0.65
430 75 20.5% 90 21.8%
Marital status
Without a partner 322 89.9% 352 86.5% 2.15 0.14
b
With a partner 36 10.1% 55 13.5%
Socio-economic status
Low 222 61.3% 260 64.0% 0.60 0.43
Medium-high 140 38.7% 146 36.0%
Condom use with regular partner
No use 210 85.4% 210 76.6% 6.25 0.01
a,b
Use 36 14.6% 64 23.4%
Condom use with casual partner
No use 179 78.5% 123 61.8% 14.01 0.00
a,b
Use 49 21.5% 76 38.2%
Condom use with sex worker
No use 132 74.6% 93 67.4% 1.95 0.16
b
Use 45 25.4% 45 32.6%
a
Significant association (p50.05).
b
Risk factors included in the logistic regression analysis (p50.25).
Table 2. Risk factors associated with syringe sharing in injection drug
users.
Variable Crude OR 95% CI Adjusted OR 95% CI
Lives with parents
No 1.53 1.13 2.05 1.16 0.53 2.55
Yes
Lives alone
No 0.59 0.43 0.80 1.03 0.49 2.16
Yes
Marital status
No partner 1.39 0.89 2.18 1.16 0.49 2.16
With a partner
Condom use with regular partner
No use 1.77 1.13 2.79 0.52 0.16 1.70
Use
Condom use with non-regular partner
No use 2.25 1.47 3.45 4.46
a
1.23 16.05
Use
Condom use with sex worker
No use 1.14 0.86 2.31 0.73 0.29 1.84
Use
ORc, uncorrected OR; ORaj, adjusted OR.
a
Significant association.
DOI: 10.3109/14659891.2013.824037 Injecting drug users 3
J Subst Use Downloaded from informahealthcare.com by HINARI on 08/21/13
For personal use only.

The fact that the present study used a respondent driving
sample, which is considered as a non-random sample
selection that must meet a number of assumptions to be
considered as a dependent sample of the social network, could
be considered as a limitation. Similarly, higher strata popu-
lation was not reached in this study because they showed no
interest in receiving an incentive for participating. Finally,
data were derived from collected self-reports of sexual
practices and condom use and, as such, lack of honesty or
inaccuracies in responses could have affected the outcomes.
In several Latin American countries, there are groups like
IDUs, where HIV is concentrated, but in countries like
Colombia, there are no harm reduction strategies for reducing
its transmission. It is very important for the country to
prioritize public health policies and HIV prevention strategies
for IDUs, for example, opium replacement therapy, counsel-
ing and needle exchange programs, and safe elimination of
contaminated syringes and needles, among others.
The presence of risk behaviors, such as sharing syringes,
inconsistent use of condoms, and failure to adopt safer sex
practices, as well as the unfavorable conditions in their social
environment and the absence of polydrug use policies, require
an intervention in this population (Hagan et al., 2007). A
public health approach, using strategies across systems and
mobilizing multiple sectors, can enhance IDU access to HIV
prevention and care (Pe
´rez-Saleme & Herna
´ndez-Tepichin,
2004). An approach with integrated services is needed, as
well as leadership, collaboration and resources being essential
(Birkhead et al., 2007).
Declaration of interest
The authors report no conflicts of interest. The authors alone
are responsible for the content and writing of this article.
References
Amirkhanian, Y. A. (2012). Review of HIV vulnerability and condom
use in central and eastern Europe. Sexual Health, 9, 34–43.
Birkhead, G. S., Klein, S. J., Candelas, A. R., O’Connell, D. A.,
Rothman, J. R., Feldman, I. S., Tsui D. S., Cotroneo R. A., &
Flanigan, C. A. (2007). Integrating multiple programme and policy
approaches to hepatitis C prevention and care for injection drug users:
A comprehensive approach. The International Journal on Drug
Policy, 18, 417–425.
Bravo, M. J., Royuela, L., Barrio, G., Rodrı
´guez-Arenas, M. A., & de la
Fuente, L. (2004). Prevalence of indirect sharing of drug-injecting
paraphernalia in Galicia, Madrid, Seville and Valencia [Spain].
Gaceta sanitaria/S.E.S.P.A.S, 18, 472–478.
Castan
˜oPe
´rez, G. A., & Caldero
´n Vallejo, G. A. (2010). Consumo de
heroı
´na en Colombia, pra
´cticas relacionadas e incidencia en la salud
pu
´blica. Revista Cubana de Salud Pu
´blica, 36, 311–322.
Cobos Calleja, T. de los, Casanueva Gutie
´rrez, M., & Jove Gonza
´lez, C.
(2003). Perfil de los usuarios de drogas ingresados en un hospital.
Anales de Medicina Interna, 20, 8–17.
Deiss, R. G., Brouwer, K. C., Loza, O., Lozada, R. M., Ramos, R., Cruz,
M. A., Patterson T., Heckarthorn D., Frost S. D., & Strathdee, S. A.
(2008). High-risk sexual and drug using behaviors among male
injection drug users who have sex with men in 2 Mexico-US border
cities. Sexually Transmitted Diseases, 35, 243–249.
Des Jarlais, D. C., & Friedman, S. R. (1987). HIV infection among
intravenous drug users: Epidemiology and risk reduction. AIDS
(London, England), 1, 67–76.
Grangeiro, A., Holcman, M. M., Onaga, E. T., Alencar, H. D. R. de,
Placco, A. L. N., & Teixeira, P. R. (2012). Prevale
ˆncia e
vulnerabilidade a
`infecc¸a
˜o pelo HIV de moradores de rua em Sa
˜o
Paulo, SP. Revista de Sau
´de Pu
´blica, 46, 674–684.
Hagan, H., Campbell, J. V., Thiede, H., Strathdee, S. A., Ouellet, L.,
Latka, M., Hudson S., & Garfein, R. S. (2007). Injecting alone among
young adult IDUs in five US cities: Evidence of low rates of injection
risk behavior. Drug and Alcohol Dependence, 91, S48–S55.
Khan, M. R., Bolyard, M., Sandoval, M., Mateu-Gelabert, P., Krauss, B.,
Aral, S. O., & Friedman, S. R. (2009). Social and behavioral correlates
of sexually transmitted infection- and HIV-discordant sexual partner-
ships in Bushwick, Brooklyn, New York. Journal of Acquired Immune
Deficiency Syndromes (1999), 51, 470–485.
March, J. C., Oviedo-Joekes, E., & Romero, M. (2007). Inconsistent
condom use among socially excluded heroin users. Gaceta Sanitaria/
S.E.S.P.A.S, 21, 321–328.
Mathers, B. M., Degenhardt, L., Ali, H., Wiessing, L., Hickman, M.,
Mattick, R. P., Myers B., Ambekar A., & Strathdee, S. A. (2010). HIV
prevention, treatment, and care services for people who inject drugs:
A systematic review of global, regional, and national coverage. The
Lancet, 375, 1014–1028.
McCreesh, N., Frost, S. D. W., Seeley, J., Katongole, J., Tarsh, M. N.,
Ndunguse, R., Jichi, F., Lunel, N. L., Maher, D., Johnston, L. G.,
Sonnenberg, P., Copas, A. J., Hayes, R. J., & White, R. G. (2012).
Evaluation of respondent-driven sampling. Epidemiology, 23,
138–147.
Mendez, D., Ana, M., Schwarz, P., & Camarotti, A. C. (2009). Pra
´cticas
sexuales en usuarios de drogas y riesgo de transmisio
´n del VIH/sida.
Revista Argentina de Sociologı
´a, 7, 150–163.
Oliveira, J. F. de, & Paiva, M. S. (2007). Vulnerabilidade de mulheres
usua
´rias de drogas ao HIV/AIDS em uma perspectiva de ge
ˆnero.
Escola Anna Nery, 11, 625–631.
OMS, UNODC, & UNAIDS. (2012). Guides for countries to set targets
for universal access to HIV prevention, treatment and care for
injecting drug users. Recuperado a partir de http://apps.who.int/iris/
bitstream/10665/77969/1/9789241504379_eng.pdf
Paquette, D., Bryant, J., & Wit, J. (2011). Respondent-driven sampling
and the recruitment of people with small injecting networks. AIDS and
Behavior, 16, 890–899.
Pe
´rez-Saleme, L. M., & Herna
´ndez-Tepichin, G. (2004). Epidemiology
of HIV infection: Twenty-years of experience. Revista de
Investigacio
´n Clı
´nica; Organo del Hospital de Enfermedades de la
Nutricio
´n, 56, 134–142.
Ross, M. W., McCurdy, S. A., Kilonzo, G. P., Williams, M. L., &
Leshabari, M. T. (2008). Drug use careers and blood-borne pathogen
risk behavior in male and female Tanzanian heroin injectors. The
American Journal of Tropical Medicine and Hygiene, 79, 338–343.
UNAIDS. (2012). Global Report. UNAIDS report on the global AIDS
epidemic. Recuperado a partir de http://www.unaids.org/
USAIDS. (2010). Conocimiento de VIH/SIDA, Otras ETS y Formas de
Prevencio
´n. Recuperado a partir de http://www.measuredhs.com/
Wodak, A., & McLeod, L. (2008). The role of harm reduction in
controlling HIV among injecting drug users. AIDS, 22, S67–S79.
World Health Organization. (2013). Universal access to HIV/AIDS
prevention, treatment and care. Recuperado a partir de http://
www.who.int/hiv/topics/universalaccess/en/index.html.
Notice of Correction:
A change has been made to the name of the lead author since this article’s original online publication date of
August 14, 2013.
4D. Berbesi et al. J Subst Use, Early Online: 1–4
J Subst Use Downloaded from informahealthcare.com by HINARI on 08/21/13
For personal use only.