changes in high risk behavior

AIDS Behav (2006) 10:S47–S56
DOI 10.1007/s10461-006-9138-y

ORIGINAL PAPER

Changes in High-Risk Behaviors Over Time Among Young Drug
Users in South Vietnam: A Three-Province Study
Le Thuy Lan Thao Æ Christina P. Lindan Æ
Deborah B. Brickley Æ Le Truong Giang

Published online: 20 July 2006
Ó Springer Science+Business Media, Inc. 2006

Abstract Vietnam is in the midst of an expanding
HIV epidemic, primarily driven by an increase in
injection drug use in young people. This study was
conducted to understand the patterns and initiation of
drug use, and the sexual risk behavior among youth in
three provinces in southern Vietnam. A cross-sectional
survey was conducted among male and female drug
users under age 25 recruited from drug treatment
centers (N = 560) and the community (N = 240) in Ho
Chi Minh City, Dong Nai and Ba Ria-Vung Tau. The
majority of those surveyed (82%) began by smoking
heroin; after a year, 57% were injecting heroin and/or
opium. Initiation of drug use frequently occurred in
entertainment venues. Among injectors, 23% shared
needles; 71% of all users were sexually active of whom
77% had unprotected sex. More than half of those
recruited from treatment centers had previously been
in drug treatment. Public health programs to prevent

and treat the dual epidemics of HIV and drug abuse
must be able to access and respond to the needs of
youth, many of whom are unemployed and exposed to
drug traffic.

L. T. L. Thao Æ L. T. Giang
Ho Chi Minh City Provincial AIDS Committee, Ho Chi
Minh City, Vietnam
C. P. Lindan Æ D. B. Brickley (&)
Institute for Global Health, University of California,
50 Beale Street, Suite 1200, San Francisco,
CA 94105, USA
e-mail:
L. T. Giang
Ho Chi Minh City Health Department, Ho Chi Minh City,
Vietnam

Keywords IV Drug Users Æ Risk-taking Æ Sexual
Behavior Æ Heterosexual Transmission Æ Vietnam Æ
Asia

Introduction
Since the first AIDS case was reported in Ho Chi Minh
City (HCMC) in 1990, the HIV epidemic in Vietnam
has spread rapidly, primarily due to the expansion of
injection drug use throughout the country (Hien, Long,
&Huan, 2004). The estimated total number of HIV
infections in Vietnam is 263,000 (Ministry of Health
Vietnam 2005), 51% of which are attributed to injection drug use (Hien et al., 2004). The greatest number
of reported HIV infections occur in the north, where

drug use proliferates along trafficking routes and the
borders of China and Laos (Beyrer et al., 2000;
Hammett et al., 2005). However, all provinces in
Vietnam currently report persons with HIV infection
and an increase in the number of cases due to sexual
transmission. The southern city of Ho Chi Minh
accounts for a quarter of the total national reported
and estimated cases of HIV (UNAIDS and WHO,
2005). In 2004, the HIV prevalence among injection
drug users in HCMC was 53% and 16% among female
sex workers (Hien et al., 2004).
During the past 10–15 years, the epidemiology of
drug use in Vietnam has changed, providing new
challenges for the control of narcotic use, as well as
HIV transmission. Prior to 1996, narcotic drug abuse
occurred primarily among older men who smoked or
injected blackwater opium (Lindan et al., 1997;
Reid and Costigan, 2002). By 2002, approximately
80% of registered drug addicts were under age 35

123


S48

(‘‘Vietnam: Country profile’’, 2002). This increase in
drug use among youth largely accounts for the dramatic rise in the proportion of reported HIV infections
among young people: 16% of all HIV infections in
1995 were among those less than 30 years of age,
compared to 69% in 2004 (Hien et al., 2004).

Control of HIV infection is linked to control of drug
abuse in Vietnam; both are managed within the
Committee for AIDS and for Control of Drug Abuse
and Prostitution. In Ho Chi Minh City, a pilot program
was established in which 22 drug treatment centers
(also known as rehabilitation centers) housing more
than 30,000 drug users provided detoxification, education, vocational training and health care. Although
there are rehabilitation centers elsewhere in Vietnam,
the HCMC program had by far the greatest number.
Identified drug users remain in treatment for up to and
in some cases more than 2 years. Approximately twothirds of these clients are now being released as their
mandated rehabilitation time has been completed
(‘‘3,097 drug addicts in Ho Chi Minh City finish
rehab’’, 2005). HCMC is grappling with ways to integrate these young people into society.
Effective strategies to prevent initiation of drug use
among youth need to be developed and evaluated. The
public health approach to HIV and illicit drug use has
been based primarily on using peer outreach to provide
education, some needle exchange and counseling
(Khoat, West, Valdiserri, & Phan, 2003). Methadone
treatment is not yet available, but it is anticipated that
it will soon become legal (JVnet, 2005). High unemployment rates, injecting as well as smoking practices,
and sexual risk behavior with multiple partners are
issues that need to be addressed to successfully prevent
HIV infection in this population.
In order to learn more about patterns of drug use
and sexual and injecting behaviors among young drug
users, this study was conducted by the HCMC AIDS
Committee in three different provinces in the southern
region of the country.

Methods
Setting
Subjects were recruited from three cities in southern
Vietnam, chosen to represent different urban settings:
a metropolis, a suburb, and a smaller resort town. Ho
Chi Minh City is the largest city in Vietnam, with a
population of 7 million and the commercial and
industrial center of the country. Dong Nai is a rapidly

123

AIDS Behav (2006) 10:S47–S56

developing industrial zone in the suburbs of HCMC.
Ba Ria-Vung Tau is a beach resort about 120 km
from HCMC that attracts both domestic and foreign
tourists.
Study Subjects
Subjects were recruited both from drug rehabilitation
centers and community venues. A sample size of 400
participants from HCMC, and 200 from each of the
two other cities was based on feasibility and cost. It was
planned that approximately 75% of subjects would be
recruited from drug rehabilitation centers and 25%
from the community, based on the belief that the
majority of identified drug users would be in treatment.
In addition, the feasibility of recruitment from drug
rehabilitation centers was expected to be easier. In
HCMC, where large numbers of drug users in both the

community and rehabilitation centers could easily be
identified, this sampling approach was taken. In the
other two locations, however, the total number of drug
users was much smaller, and the ability to recruit
subjects was limited by the size of the rehabilitation
centers and/or the availability of identified users in the
community. Thus, 88% of study participants from
Dong Nai and 43% from Ba Ria-Vung Tau were recruited from treatment centers.
At the time of the study, there were eight government rehabilitation centers in HCMC, and participants
were recruited from each, proportional to the percentage of total drug admissions received by that site.
For example, a total of 300 subjects were to be recruited from rehabilitation centers in HCMC, and 100
from the community. If center A received 20% of all
drug rehabilitation admissions in the city, then 20% of
the 300 subjects were enrolled from that site. Both
Dong-Nai and Vung-Tau had only one drug center
each, and all eligible residents at the time of the study
were asked to participate.
Recruitment was conducted by experienced social
workers and counselors who were employed and
trained by the HCMC AIDS Committee. In the drug
treatment centers, health and social workers who were
familiar with young drug users conducted the interviews. Recruitment and interviewing of community
based drug users were carried out by peer educators
currently working in those communities. All study
personnel had experience in counseling drug users;
they also received additional training in interviewing
and in research methods as part of this study.
Recruitment occurred from October 1999 to March
2000.

AIDS Behav (2006) 10:S47–S56

Subjects were self-identified users of opiate drugs
and < 25 years in age. Because all users in treatment
could not be enrolled, participants were selected by
systematic sampling. Counselors described the study
protocol and procedures and received verbal consent
from the subjects. Participation was voluntary and
study staff made clear that refusal to participate would
not influence ability to receive treatment or services.
Subjects received a small gift (value less than US$1) to
compensate for their participation. All questionnaires
were anonymous and no identifying information was
collected; names of the subjects were not recorded or
linked with the survey in any way.

S49

using v2-test for differences in categories, or student’s
t-test for differences in mean values. Data are presented stratified by recruitment site (drug rehabilitation
centers versus the community).
Ethical Review
The study was reviewed and approved by the Provincial AIDS Committee of Ho Chi Minh City before an
Institutional Review Board (IRB) was established. An
IRB and NIH Federal Wide Assurance are currently in
place. The Committee on Human Research at UCSF
provided approval for UCSF co-authors to participate
in data analysis and manuscript writing.

Measures
Questionnaire items and methods of approaching drug
users were explored during focus group discussions
with peer educators. The questionnaire was administered by trained study staff in Vietnamese using lay
language and common terms. Areas addressed included basic demographic characteristics of respondents, drug use behaviors (both at the beginning of
drug use and currently), sexual behavior, condom use,
and knowledge and attitudes related to HIV. There
were four questions about HIV transmission regarding
whether HIV could be transmitted from mother to
child, by sharing needles or injection equipment,
through sex, or via mosquitoes or insects. There were
three questions asking whether it is possible to prevent
HIV transmission by using condoms during sex, using
clean needles/injection equipment, and by not touching
or eating with persons who have AIDS. There were
five additional questions regarding the utility of condoms. Discriminatory attitudes towards persons with
HIV/AIDS (PLWHA) were identified by asking five
questions: whether PLWHA should be fired from jobs
and isolated; whether their identity should be provided
to the public; whether they should be allowed to work
and live as usual; whether their identity should be
hidden to avoid discrimination; and whether their
health and psychological well-being should be supported.
Analysis
Data were entered onsite into EpiInfo 6 and analyzed
using Intercooled Stata 7.0. Distribution of responses
were evaluated using proportions, mean and median
values. Data was initially stratified by city, gender,
recruitment venue (community or treatment center),
age, employment status, and financial status to identify

relationships of interest. Differences were evaluated

Results
Demographics
Table 1 shows the demographic profile of subjects by
type of recruitment site (community versus treatment
center), which varied by city. Approximately 19% of
participants were women. The median age was 20
(range 13–24), and 19 subjects were between 13 and
15 years of age. The majority had only elementary or
primary education, and 5% were illiterate. Of those
older than 18 years, only 34% had achieved a high
school education (data not shown). About one-half
were employed, 17% were students, and the remainder
had no job. Three quarters of all subjects described
their families as poor or struggling with just enough to
live on. The vast majority (87%) of subjects lived with
their families and 65% were completely financially
dependent upon them; 55% reported that their family
was a main source of drug money. Of the 560 recruited
from treatment sites, 37% had entered voluntarily, 20%
had been sent there by their families, and 43% had been
arrested and were undergoing mandatory treatment.
There were some notable differences between those
interviewed from the community compared to treatment centers. Those in treatment tended to be younger
(less than 22 years old) and to have higher levels of
education and be students. They were more likely to be
financially dependent on families, and those families
were better off financially—only 6% were described as
poor compared to 21% of families of community

participants. In addition, a greater proportion in
treatment (62% vs. 40%) relied on the family as the
source of drug money.
Across provinces, subjects recruited from the treatment centers were similar demographically, whereas
there were some differences among those recruited

123


S50

AIDS Behav (2006) 10:S47–S56

Table 1 Demographic characteristics of 800 young drug users by recruitment site
Recruitment site
Total

All
City***
Ho Chi Minh City
Dong Nai
Ba Ria-Vung Tau
Gender, male
Age, years***
13–15
16–18
19–21
22–24
Educational level***
Illiterate

Primary/elementary
Some high school or more
Employment***
Job
Student
Unemployed
Ever married
Living with family***
Financially dependent on family***
Completely
Partially
Independent
Family economic status***
Poor
Enough to live on
Wealthy/middle class
Sources of drug moneya
Work
Loans
Family
Other
Reason for being in treatment
Voluntary
Government mandated
Request of family

Community

Treatment centers

N

%

N

%

N

%

800

100

240

30

560

70

400
200
200
650

50

25
25
81

100
25
115
195

25
13
58
81

300
175
85
455

75
88
43
81

19
177
312
292

2

22
40
37

14
43
80
103

6
18
33
43

5
134
232
189

1
24
41
34

39
503
258

5
63

32

22
174
44

9
73
18

17
329
214

3
59
38

372
133
295
91
699

47
17
37
11
87

115
21
104
33
184

48
9
43
14
77

257
112
191
58
515

46
20
34
10
92

519
149
132

65
19

17

123
52
65

51
22
27

396
97
67

71
17
12

85
519
196

11
65
25

51
143
46

21
60
19

34
376
150

6
67
27

319
46
443
80

40
6
55
10

126
6
97
35

53
3
40

15

193
40
346
45

34
7
62
8

–
–
–

–
–
–

208
238
114

37
43
20

–
–

–

–
–
–

P-values refer to differences between recruitment sites across categories, using v2 statistics. ***P < = .001
a

Categories are not mutually exclusive

from communities (data not shown). Unemployment
was higher among participants from Vung Tau
compared to the other two cities (56% vs. 36% in Ho
Chi Minh City and 31% in Dong Nai, P < 0.001); those
recruited from Ho Chi Minh City were more likely to
be financially independent from their families (39% vs.
22% in Vung Tau, and 4% in Dong Nai, P < .001).
Men and women in the study were similar in age,
education and employment status; however, more
women lived on their own without financial support
(data not shown). They were more likely to be married
(19% vs. 10% of men, P < .01), live apart from their
families (22% vs. 11% of men, P < .01) and be

123

financially independent of them (24% vs. 15% of men,
P < .01). The economic status of their families was
also lower—82% of women were from poor or financially struggling families compared to 74% of the

families of men surveyed (P < .05).
Drug Use Behaviors
Table 2 highlights the drug use patterns of subjects
recruited from rehabilitation centers and the community. The mean length of time using drugs was 2.7 years
(median 3 years, range 1–12 years); those in treatment
had been using for slightly longer—about half had


AIDS Behav (2006) 10:S47–S56

S51

Table 2 Drug use patterns among 800 young drug users by type of recruitment site
Community

All
Current Age 13–18 (N = 196) **
Used drugs £2 years
Used drugs >2 years
Current age 19–24 (N = 604)*
Used drugs £2 years
Used drugs >2 years
Current methods of drug usea
Injection
Smoking
Snorting/swallowing
Currently share injection equipment
Drugs used currentlya
Heroin
Opium

Non opiates or marijuana
Reason for initiating drug use
Peer pressure
Personal problems
Family problems
Other
Used drugs with whom initially***
No one/alone
Friends
Other people
Previously been in treatment***

Treatment centers

N

%

N

%

240

30

560

70

42
15

74
26

72
67

52
48

98
85

54
46

187
234

44
56

147
91
18
35/126

61

38
8
28

306
257
31
57/274

55
46
6
21

170
77
12

71
32
5

550
6
15

98
1
3

187
22
15
16

78
9
6
7

444
39
37
40

79
7
7
7

14
220
6
76

6
92
3
32

91
449
20
293

16
80
4
52

P-values refer to differences between recruitment sites across categories, using v2 statistics. *P < = .05 **P < = .01 ***P < = .001
a

Not mutually exclusive categories

been using drugs for more than 2 years, and some up to
12 years. Nearly all those in treatment were using
heroin and only 1% used opium, whereas 32% of those
from the community were using opium. There were
some differences in sharing practices by city of
recruitment (data not shown): 24% of injectors in
HCMC reported sharing, compared to 22% in Vung
Tau and 19% in Dong Nai. Almost 80% stated that
peer pressure was their main reason for using drugs
initially, and most began using with friends. Many,
52% of those recruited from treatment centers and
32% of those in the community, had previously been in
treatment, and roughly similar proportions believed
they could eventually give up drugs (65% of those in
treatment vs. 37% from the community). Only 11

participants reported selling drugs (data not shown).
At initiation of drug use, the majority (87%) of
subjects smoked and only 9% injected (Table 3). This
changed significantly over time—when surveyed, the
majority of users (57%) reported injection practices.
Among those who had switched method of use, the
mean time to change was 14.1 months. Sharing needles

or drug paraphernalia was reported by only onequarter of injectors, and was the same at drug initiation
(26%) and at the time of the survey (23%). Heroin was
by far the most commonly used substance; at drug
initiation, 94% used heroin: 88% exclusively and 6%
mixed it with other drugs; only 2% were using opium
and 8% marijuana. By the time of the study, 90%
overall were using heroin, and the proportion of opium
users had increased slightly (10%). In general, those
using heroing continued to do so—only 7% of heroin
users switched to opium, whereas among those using
opium initially, 67% continued to do so. Among
marijuana users, 83% were smoking or injecting heroin
when surveyed (data not shown). Overall, 17% had
changed their primary drug since initiation. Frequency
of drug use increased from a mean of 1.6 to 2.5 times
per day.
More than half the survey participants (53%) began
using drugs at an entertainment place such as a cafe´,
restaurant or karaoke bar. There was a general trend
from initially smoking heroin at entertainment venues,
to injecting heroin at home or with friends. Even so, a

123


S52

AIDS Behav (2006) 10:S47–S56

Table 3 Change in drug use: at drug initiation and currently, among 800 young drug users
At drug initiation

Method of usea
Smoking
Injecting
Snorting/swallowing
Changed method of use since initiation
No. months to change, mean (SD)
Share injection equipment
Drug used and methoda
Heroin,
Smokeb
Inject
Snort/Swallow
Opium,
Smokeb
Inject
Marijuana
Medication
Changed drugs used since initiation
No. times/day use drugs
£1

2–3
‡4
Primary location of usea
Shooting gallery
Own house
Entertainment placec
Friend’s house
Other
Purchase drugs wherea
Shooting gallery
Entertainment place
Friend
Other

Currently

N

%

N

%

696
74
101
–
–
19/72

87
9
13
–
–
26

348
453
51
403
14.1
92/400

44
57
6
50
SD 10.6
23

750
659
62
98
13
5
9
67

20
–

94
82
8
12
2
1
1
8
3
–

720
339
376
50
83
6
82
27
37
135

90
42
47
6
10

1
10
3
5
17

399
360
31

50
46
4

158
499
143

20
62
18

87
133
425
160
45

11
17

53
20
6

133
262
269
156
46

10
33
34
20
6

–
–
–
–

–
–
–
–

552
36
148
72

69
5
19
9

a

More than one response possible

b

Proportion smoking, injecting etc. are calculated using the entire cohort (800) as the denominator

c

Entertainment places include restaurants, cafes, karaoke bars, etc

significant proportion (28% of injectors, 40% of
smokers, and 35% of heroin users) continued to use
drugs at entertainment sites (Table 4). Opium was
used most commonly at shooting galleries (54%). All
drugs, including opiates, marijuana and other drugs
were purchased primarily at shooting galleries.
Sexual Risk Behaviors
Table 5 describes the sexual behaviors of study subjects. Most, but not all, were sexually experienced: 82%
of women, 71% of men, and 67% of those who had
never been married. Among the 586 who had had sex,
20% had sex with friends in their drug-using group,
37% with friends outside of their using group, and 24%

with casual partners. About twice as many women

123

(41%) as men (20%) reported sex with casual partners.
We did not ask women whether they had engaged in
commercial sex, although 32% of men had visited female sex workers. About half of respondents reported
never using condoms, and this was not significantly
different between men and women. Condom use with
spouses was low—6% reported always using them.
Among those who were married and had other
relationships, 32% never used condoms with extramarital partners (data not shown).
Knowledge and Attitudes about HIV/AIDS
Participants in the study were generally knowledgeable
about HIV/AIDS (Table 6). Most (86%) could correctly answer all questions about HIV transmission,


AIDS Behav (2006) 10:S47–S56

S53

Table 4 Current primary location of drug use and purchase among those who used heroin and/or opium (N = 800)
Method of intake and drug used

Primary location of drug use
Shooting
gallery

Entertainment
placea

Friend’s house

Own house

N

%

N

%

N

%

N

%

Method
Injectors N = 453
Smokers N = 348
Drug
Heroin N = 720
Opium N = 83

96
37

21
11

130
144

29
41

86
65

19
19

154
112

34
32

85
50

12
60

256
13

36
16

144
13

20
16

250
12

35
15

Current drug used

Primary place of purchase
Shooting
gallery

Drug
Heroin N = 720
Opium N = 83
a

Entertainment
placea

Friend

Other

N

%

N

%

N

%

N

%

478
77

66
93

34
3

5

4

146
2

20
2

71
1

10
1

Entertainment places include restaurants, cafes, karaoke bars, et al.

and 80% could correctly answer all questions about
means of preventing HIV. However, only 43% of
injectors and 21% of non-injectors believed they were
at risk for HIV. Most had favorable attitudes about
how society should respond to people living with HIV/
AIDS. Eighty-four percent believed the identity of
those with HIV infection should not be disclosed in
order to avoid discrimination, and an equal number
believed that people living with HIV should be allowed
to work and live as usual. Almost all (97%) believed
that family and health services should provide care and
support for people living with HIV. Those in treatment
were more than twice as likely to have undergone HIV

testing than participants from the community (67% vs.
28%). A very small number of those who reported
having been HIV tested (440), indicated that they were
HIV infected (8%), and was higher (12%) among
injectors compared to non-injectors (1%).

Discussion
This study provides important information about how
youth in southern Vietnam begin using drugs and how
this changes over time. Heroin was by far the drug of
choice, and opium use was comparatively uncommon.

Table 5 Marital status and sexual risk behaviors of 800 young drug users
All

Total
Sexually experienceda
Sex with whom?a,b
Female sex worker
Friend in drug using group
Other friend
Casual partner
If married, sex with non-marital partner
Frequency of condom usec (N = 532)
Always
Sometimes
Never
a

Women

Men

N

%

N

%

N

%

800
586

100
73

150
123

19
82

650
463

81
71

147
117
217
143
37/91

25
20
37
24
41

–
32
34
51
13/28

–
26
27
41
46

147
85
183

92
24/63

32
18
40
20
38

122
125
285

23
23
54

22
29
57

20
27
53

100
96
228

24

23
54

Includes sexually active single persons, as well as those who were married, divorced, or separated

b

Categories are not mutually exclusive

c

Condom use among 108 women and 424 men who were single and sexually active, or married with non-marital partners

123


S54

AIDS Behav (2006) 10:S47–S56

Table 6 Knowledge and attitudes about HIV (N = 800 drug
users)

HIV transmission knowledgea
All 4 questions correct
Knowledge about HIV preventiona
All 3 questions correct
No. discriminatory attitudes towards PLWHAa
None
1

2–5
Perceive oneself at risk for HIV
Non-injectors (N = 345)
Injectors (N = 455)
Previously tested for HIV
Recruited from community (N = 240)
Recruited from treatment centers (N = 560)
a

N

%

685

86

637

80

543
95
162

68
12
20

73

194

21
43

67
373

28
67

Items are described in the Methods section of the text

This is in contrast to a decade ago, when opium use was
more prevalent, particularly in the southern part of the
country. One study from the mid-1990s reported that
96% of drug users in treatment were using opium
(Tran, Williams, Truong, & Do, 1998). Most young
people in our study reported smoking drugs initially,
with one-half transitioning to injection within a year.
This quick transition to injection likely results from
increasing addiction, the need for more drug per dose,
and the fact that injection is comparatively cheaper.
Other studies of drug users have found that transition
to injection occurred within 7 months to 1.85 years
(Hien et al., 2004; Nguyen, Hoang, Pham, & Detels,
2001). Needle sharing in our study was reported by
only one-quarter of those injecting drugs, and is considerably less than in other surveys in which up to 70%
of injectors in urban settings reported sharing equipment (Hien et al., 2001; Nguyen et al., 2001; Tran et al.,
1998). Other reasons for varying rates of sharing may

be related to cost, availability of clean needles and
concerns about arrest if caught with injection
equipment.
Until now, a policy of sending drug users to large
treatment centers for up to 2 years of detoxification
and ‘‘rehabilitation’’ has been widely pursued (Rekart,
2002; Vu, 2001). Vast resources have been dedicated to
expanding existing drug treatment centers and building
new ones. At the time of this study there were 8 such
sites in HCMC; by 2005, there were 22, mostly run by
the government with a few private clinics, and housing
30,000 users. Previous data show that upon leaving a
rehabilitation center, at least 90% of clients begin
using again (International Narcotics Control Strategy
Report, 1999). This is consistent with reports of prior
treatment among participants in our study—55% of
those in treatment and one-third of those in the

123

community had previously been in rehabilitation.
Because the maximum 2-year detention has now
expired for many with mandatory sentences, large
numbers of drug users are being released into the
community, up to 16,000 in 2006 in HCMC alone
(‘‘3,097 drug addicts in Ho Chi Minh City finish
rehab’’, 2005). This poses a large challenge in finding
ways to support abstinence, education, and reintegration into society. A step-down program of 1–3 years in
which former drug users can be employed and live in
more controlled settings is being pursued.

With the large efflux of drug users from treatment,
information about initiation of drug use will be
important for prevention programs and outreach. We
found that drug use was most often initiated at entertainment venues, such as karaoke bars or cafes, and
occurred under peer pressure. Even over time, both
purchase and use of drugs continued to occur in these
public venues, although most drugs were obtained at
shooting galleries even if they weren’t used there.
Developing peer outreach at these sites may be very
important in preventing recidivism as well as reducing
experimentation with highly addictive drugs.
We were somewhat surprised to find many young
drug users were fairly well-educated, middle class and
typically living with and/or receiving financial support
from their families. This was particularly true of drug
users in treatment, who in many cases were sent to
treatment by family members. Many families were
also a source of money for buying drugs. This may
occur because relatives would rather provide funds
than have children steal and bring shame to the family.
Because of this involvement, an important role may
exist for families in supporting prevention, harm
reduction and drug abstinence. However, this also
poses a dilemma for families, many of whom are
reluctant to have former drug users return home
following rehabilitation.
Despite concerns about the use of rehabilitation
centers, the HIV prevalence rates in HCMC declined
significantly in the several years during which the
majority of drug users were in treatment. Data from

HCMC sentinel surveillance show that HIV prevalence
among drug users dropped from 83% to 48% from
2002 to 2005, concomitant with reductions in HIV
among other at-risk populations. As the majority of
HIV infected persons were in rehabilitation centers,
HIV transmission both sexually and parenterally was
reduced. In addition, the presence of drug mafia and
dealers declined. Unfortunately, as former drug users
re-enter society, HCMC may witness a return of a
ready drug supply as well as the emergence of methamphetamine use.


AIDS Behav (2006) 10:S47–S56

We did not perform HIV testing and counseling as
part of this study. By self-report, only 55% of participants claimed to have undergone HIV testing. The
greater proportion in treatment who had been tested
may be a result of national HIV sentinel surveillance
which is performed primarily in rehabilitation centers.
A very small number, only 8% of the sample, admitted
to being HIV seropositive. This likely reflects underreporting and fear of disclosure due to stigma.
Sentinel surveillance data among injection drug users
indicates that 65% of surveyed injection drug users in
HCMC and 20% in Dong Nai were HIV infected
during the period when this study was performed (Hien
et al., 2004). Even though half of those recruited in our
study were injecting drugs, high HIV prevalence rates
among non-injecting heroin users have been noted in
many reports. In a study of 500 drug users in Hai
Phong, HIV prevalence among non-injectors was 46%

(Nguyen et al., 2001). In HCMC, 25% of surveyed
heroin smokers were HIV infected in 2000 (unpublished surveillance data, HCMC AIDS Committee).
The reasons for high rates among non-injectors may
be due to sexual transmission and the overlap among
drug users, their sexual partners, and female sex
workers (National AIDS Committee Bureau of Vietnam, 2001). Overall, one-half of those who were
sexually active in our study never used condoms with
their partners, and one-third of men had visited
female sex workers. Although we didn’t ask women
whether they sold sex, a large proportion, 41%,
admitted to having casual sex partners. It is possible
that many of these women were trading sex for drugs
or money.
The HIV prevalence among sex workers has risen
steadily in Vietnam, particularly in the southern part
of the country. Concomitantly, drug use among
female sex workers is a growing problem. Studies of
sex workers in HCMC and in Hanoi have indicated
that 25–45% of them inject drugs, and among them
HIV rates are close to 50% (Bain et al., 2003; Tran,
Detels, Long, & Lan, 2005). Thus, HIV among
female sex workers may be due to injecting drug use
and sexual transmission from drug using sex partners.
It is not clear whether many women identified as sex
workers and detained in rehabilitation centers, are
primarily addicted to drugs and compelled to sell sex
to support a habit; or whether they are female sex
workers who have become addicted to drugs. This
distinction may only be relevant in pointing out
that women who sell sex may be in need of harm

reduction programs.
There were several limitations to this study. Use
of peer educators as interviewers may have biased

S55

responses of participants toward socially desirable
answers, particularly from among those who were in
treatment. However, all efforts were made to maintain confidentiality and encourage clients to respond
accurately. The representativeness of the communitybased sample may be reduced because participants
were recruited from among networks of drug users
already known to peer outreach workers. Likewise,
participants enrolled from rehabilitation centers may
not be representative of drug users in general. We
do not have information about whether risk behavior, related to drugs or sexual relationships, was
taking place during rehabilitation. Because males and
females are separated and centers are inpatient
facilities, it is assumed that drug use and heterosexual sex are infrequent; however, to our knowledge
this has not been formally evaluated. Questionnaire
items were not designed to identify specific recall
periods, and we have assumed that ‘‘current’’ drug
use refers to the period prior to entry in rehabilitation centers. In addition, recall periods were not
specified for questions relating to sexual activity and
condom use.
This study and many others point to the need to
address the growing population of young drug users in
Vietnam who are fueling and most likely to succumb to
the HIV epidemic (Giang, Luyen, Thao, & Narimani,
1999; Hien et al., 2004). Many of those in our study
were less than 18 years of age, and some had started

using drugs by the time they were 13. Innovative
prevention programs need to be developed, such as
conducting outreach at entertainment sites where
young people smoke and exchange drugs. Recent
evidence shows that community-based interventions
for drug users are effective in reducing risk behaviors,
yet there remains a gap in the number who could
benefit from such interventions and those who actually
receive them (Needle et al., 2005). Young users are
sexually active and do not use condoms regularly. This
puts them at additional risk of HIV acquisition, and
requires that interventions also promote fewer
episodes of unprotected sex (Des Jarlais & Semaan,
2005; Go, Quan, Yoytek, Celentano, & Nam, 2006;
Nguyen et al., 2001). The recent official recognition of
needle exchange and drug substitution as valid
components of harm reduction and its inclusion into
law will be a big step in moving Vietnam’s HIV
prevention programs forward (JVnet, 2006). However,
this will only be successful if drug users can obtain
clean paraphernalia without fearing arrest, and if
methadone becomes available to those who need it.
Until then, high rates of drug use could spell disaster
for many of Vietnam’s youth.

123


S56

References
International Narcotics Control Strategy Report (INCSR) (1999).
Bureau for International Narcotics and Law Enforcement
Affairs, U.S. Department of State.
Vietnam: Country profile (2002). Retrieved March 20, 2006, from
/>3,097 drug addicts in Ho Chi Minh City finish rehab. (October 19,
2005). Nhan Dan.
Bain, D. L., Thinh, T., Thi, M. D. A., Giang, L. T., Mandel, J. S.,
& Lindan, C. P. (2003). Vietnam’s first anonymous HIV
testing and counseling site. World AIDS Foundation Symposium, Paris.
Beyrer, C., Razak, M. H., Lisam, K., Chen, J., Lui, W., & Yu, X.
F. (2000). Overland heroin trafficking routes and HIV-1
spread in south and south-east Asia. AIDS, 14(1), 75–83.
Des Jarlais, D. C., & Semaan, S. (2005). Interventions to reduce
the sexual risk behaviour of injecting drug users. The
International Journal on Drug Policy, 16S, S58–S66.
Giang, L. T., Luyen, N. H., Thao, L. T. L., & Narimani, P. (1999).
Promoting self-help activities for people living with HIV/
AIDS in Ho Chi Minh City, Vietnam. Sexual Health
Exchange, 3, 4–6.
Go, V. F., Quan, V. M., Yoytek, C., Celentano, D., & Nam, L. V.
(2006). Intra-couple communication dynamics of HIV risk
behavior among injecting drug users and their sexual partners in northern Vietnam. Drug Alcohol Depend, Epub
ahead of print.
Hammett, T. M., Johnston, P., Kling, R., Liu, W., Ngu, D., Tung,
N. D., Binh, K. T., Dong, H. V., Hoang, T. V., Van, L. K.,
Donghua, M., Chen, Y., Des Jarlais, D. C. (2005). Correlates
of HIV status among injection drug users in a border region
of southern China and northern Vietnam. Journal of
Acquired Immune Deficiency Syndromes, 38(2), 228–235.

Hien, N. T., Giang, L. T., Binh, P. N., Deville, W., van Ameijden,
E. J., & Wolffers, I. (2001). Risk factors of HIV infection
and needle sharing among injecting drug users in Ho Chi
Minh City, Vietnam. Journal of Substance Abuse, 13(1–2),
45–58.
Hien, N. T., Long, N. T., & Huan, T. Q. (2004). HIV/AIDS
epidemics in Vietnam: Evolution and responses. AIDS
Education and Prevention, 16(Suppl A), 137–154.
JVnet (2005). HR working group—minutes of Nov meeting.
Retrieved December 2, 2005.

123

AIDS Behav (2006) 10:S47–S56
JVnet (2006). The on-going seminar on substitution treatment.
Retrieved March 6, 2006.
Khoat, D. V., West, G. R., Valdiserri, R. O., & Phan, N. T.
(2003). Peer education for HIV prevention in the Socialist
Republic of Vietnam: A national assessment. Journal of
Community Health, 28(1), 1–17.
Lindan, C. P., Lieu, T. X., Giang, L. T., Lap, V. D., Thuc, N. V.,
Thinh, T., Lurie, P., & Mandel, J. S. (1997). Rising HIV
infection rates in Ho Chi Minh City herald emerging AIDS
epidemic in Vietnam. AIDS, 11(Suppl 1), S5–13.
Ministry of Health Vietnam (2005). HIV/AIDS estimates and
projections 2005–2010.
National AIDS Committee Bureau of Vietnam (2001). Report of
behavioral sentinel surveillance, first round.
Needle, R. H., Burrows, D., Friedman, S. R., Dorabjee, J.,
Touze, G., Badrieva, L., Grund, J-P. C., Kumar, M. S., Nigro, L., Manning, G., & Latkin, C. (2005). Effectiveness of

community-based outreach in preventing HIV/AIDS among
injecting drug users. International Journal of Drug Policy,
16S, S45–S57.
Nguyen, T. A., Hoang, L. T., Pham, V. Q., & Detels, R. (2001).
Risk factors for HIV-1 seropositivity in drug users under
30 years old in Haiphong, Vietnam. Addiction, 96(3), 405–
413.
Reid, G., & Costigan, G. (2002). Revisiting ‘the hidden epidemic’’: A situation assessment of drug use in Asia in the
context of HIV/AIDS (report). Melbourne: The Centre for
Harm Reduction, The Burnet Institute.
Rekart, M. L. (2002). Sex in the city: Sexual behaviour, societal
change, and STDs in Saigon. Sexually Transmitted Infections, 78(Suppl 1), i47–i54.
Tran, S. D., Williams, A. B., Truong, H. T., & Do, N. H. (1998).
Risk factors for HIV seropositivity in a sample of drug users
in drug treatment in Ho Chi Minh City, Vietnam. Journal of
Acquired Immune Deficiency Syndrome and Human Retrovirology, 17(3), 283–287.
Tran, T. N., Detels, R., Long, H. T., & Lan, H. P. (2005). Drug
use among female sex workers in Hanoi, Vietnam. Addiction, 100(5), 619–625.
UNAIDS, and WHO (2005). AIDS epidemic update.
Vu, T. (2001). Harm reduction for injecting drug users in
Vietnam: A situation assessment. Melbourne, Australia:
Macfarlane Centre for Medical Research.