Mental health and sexual risk behaviours in a South
African township: A community-based
J. Smita,b, L. Myera,c,*, K. Middelkoopa, S. Seedatb, R. Wooda,
L.-G. Bekkera, D.J. Steinb
aDesmond Tutu HIV Centre, University of Cape Town, Cape Town, South Africa
bMRC Unit for Anxiety and Stress Disorders, University of Stellenbosch, Stellenbosch, South Africa
cInfectious Diseases Epidemiology Unit, School of Public Health and Family Medicine, University of Cape
Town, Cape Town, South Africa
Received 3 May 2005; received in revised form 4 October 2005; accepted 12 January 2006
Available online 8 May 2006
human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) in
developing countries, there are few data on the association between different forms
of mental illness and sexual risk behaviours in resource-poor settings. The objective
of this study was to examine the association between mental illness and HIV risk
behaviours in a South African township.
Study design: A cross-sectional study was performed among 645 individuals living in
households selected at random.
Methods: A self-administered translated questionnaire investigated sexual risk
behaviours [including sexual partners, condom use, casual sexual contacts, and sex
in exchange for money, drugs or a place to stay (transactional sex)], depression
(measured using the Center for Epidemiological Studies Depression Scale), alcohol
abuse (from the Alcohol Use Disorders Identification Test), and post-traumatic stress
disorder (based on the Life Event Checklist).
Results: Of the 645 individuals who completed the survey, 33% reported depression,
17% reported alcohol abuse, and 15% reported post-traumatic stress disorder. After
adjusting for demographic characteristics, the presence of any of these three
conditions was strongly associated with experiences of forced sex [adjusted odds
ratio (AOR) 2.53; 95% confidence intervals (CI) 1.60–4.02], transactional sex (AOR
2.88; 95% CI 1.29–6.48) and increased condom use (AOR 2.07; 95% CI 1.32–3.25).
Conclusions: These findings emphasize the substantial burden of mental illness in
this setting, and its association with forced and transactional sex. The temporal
nature of these associations is not always clear from this cross-sectional study, and
Objectives: Despite the high prevalence of both mental illness and
Public Health (2006) 120, 534–542
0033-3506/$ - see front matter Q 2006 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.
*Corresponding author. Address: Infectious Diseases Epidemiology Unit, School of Public Health and Family Medicine, University of
Cape Town, Anzio Road, Observatory 7925, Cape Town, South Africa.
E-mail address: email@example.com (L. Myer).
additional prospective research is required. Public health interventions are needed
to address the dual burden of HIV/AIDS and mental illness in this and similar settings.
Q 2006 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights
With over 25 million people across the continent
already infected,1the human immunodeficiency
virus (HIV)/acquired immunodeficiency syndrome
(AIDS) epidemic is a major health concern facing
countries of sub-Saharan Africa. While a range of
structural conditions, such as migrant labour, high
levels of poverty and gender inequality, contribute
to the spread of the epidemic throughout Africa,
most research to date investigating the determi-
nants of the HIV epidemic has focused on individual-
level risk factors.2However, little attention has
been given to the impact of factors related to
mental illness, such as depression, trauma and
substance abuse, on HIV risk behaviours.3,4
Several studies have suggested that HIV risk
behaviours may be associated with mental illness,
including depression and post-traumatic stress
disorder (PTSD).5–11Despite the increasing concern
about the contribution of mental illness to the
burden of disease in developing countries,12and
growing evidence that most mental health pro-
blems in these settings go untreated,13there are
few data from resource-poor countries on the
potential links between these conditions and sexual
Mental illness may be a particularly important
factor shaping high-risk sexual behaviours in South
Africa. High levels of violence are well documen-
ted,14,15including sexual, physical and domestic
abuse,16which have in turn been associated with
PTSD, depression and substance abuse.17,18Several
studies among adolescents have shown that sexual
risk behaviours are associated with alcohol and drug
abuse,19but the association between sexual beha-
viours and other forms of mental illness, such as
depression and PTSD, has received little attention.
Given the lack of data on the relationship
between mental illness and individual risk beha-
viours, this study used data from a community-
based survey to examine the association between
HIV-related sexual risk behaviours and three key
psychiatric disorders that are prevalent in the South
African context: depression, alcohol abuse and
PTSD.20,21It was hypothesized that different
measures of mental illness would be associated
with increased risk-related behaviours, such as low
levels of condom use, exchange of sex for money,
sex under the influence of alcohol and drugs, or sex
with casual (rather than regular) sexual partners.
Subjects and methods
Data for this analysis were drawn from a larger
cross-sectional study of HIV risk conducted in a peri-
urban settlement outside Cape Town, South Africa.
The community is home to an estimated 12,000
Xhosa-speaking residents living on approximately
1300 demarcated plots. The local population is
young and of low socio-economic status. Unemploy-
ment is widespread. HIV prevalence in the commu-
nity is estimated to be over 20%.22
A total of 1127 individuals aged 15 years or older
living on 200 randomly selected plots were
approached to participate in the survey. The
purpose and procedures of the study were
explained to prospective participants by a trained
fieldworker, and individuals who agreed to partici-
pate provided their written informed consent.
Questionnaires were self-administered in the
language of the participant’s choice (Xhosa or
English) and took approximately 30–45 min to
complete. Fieldworkers were available to explain
questions to illiterate participants and to assist
participants in completing the questionnaire. Par-
ticipation was voluntary and anonymous.
If individuals living on a plot were unavailable
during the fieldworkers’ initial visit, up to three
repeated attempts were made to contact the house-
hold member before the participant was deemed a
non-respondent. Prior to the start of the study,
discussions were held with the community advisory
board to ensure that the survey procedures and
Ethics Committee of the University of Cape Town.
Four items explored specific sexual behaviours
during a reference period of the 6 months prior to
Mental health and sexual risk behaviours in South Africa535
the interview: (1) condom use; (2) sex in exchange
for money, drugs or a place to stay (transactional
sex); (3) sex after heavy use of alcohol or other
drugs, and (4) sex with a person known for less than
1 day. In addition, one question enquired about
previous experiences offorced sex and one question
asked about previous treatment for a sexually
transmitted disease (STD) as a surrogate marker of
unprotected sex. Responses were coded as ‘yes’,
‘no’ and ‘refuse to answer’. The ‘refuse to answer’
category was removed from the analysis and
accounts for minor variations in sample size
involving some items.
Depression was assessed using the Center for
Epidemiological Studies Depression Scale (CES-D).
This is a 20-item self-rating scale that assesses
current levels of depression as per the DSM-IV
(Diagnostic and Statistical Manual of Mental
Disorders-Fourth Edition) criteria and is scored
from 0 (rarely or none of the time) to 3 (most of
the time). A score of 20 indicates clinical
depression. The CES-D has been widely used in
cross-cultural studies23,24and has been validated
in South Africa in different language groups.25
Alcohol use was measured using the Alcohol Use
Disorders Identification Test (AUDIT). This is a 10-
item, self-rating questionnaire that assesses hazar-
dous drinking. It is scored from 0 (never) to 4 (daily
or almost daily), and a score of 8 or more indicates
the presence of alcohol abuse.26The AUDIT was
developed by the World Health Organization and
has been shown to be reliable in a variety of
different settings, including
For measurement of PTSD, the Life Event Check-
list (LEC) was used to identify events that would
qualify as traumatic or life threatening as defined
by the DSM-IV, consisting of 16 items that have been
adapted for the cultural context of this community.
Participants who answered ‘yes’ to any of the
questions on the LEC also completed the Harvard
Trauma Questionnaire (HTQ), a 30-item self-rating
scale. Symptoms are rated from 1 (not at all) to 4
(extremely), and a cut-off score of 75 indicates
symptoms of clinical significance. It is a valid and
reliable measure to screen for PTSD where cross-
cultural sensitivity is required,29,30and it has also
been used in a school sample in the Western Cape,
All questionnaires were translated, back trans-
lated and further scrutinized through a series of
focus groups to ensure clear and appropriate
phrasing. The questionnaires were piloted in the
community before the survey began, and minor
modifications were made to ensure participant
Data were analysed using SPSS 11.0 (SPSS Corpor-
ation, Chicago, USA). Age and number of sexual
partners were coded as continuous variables.
Education was categorized into primary, secondary
or tertiary, and employment status was coded as a
dichotomous variable (employed/unemployed).
Scores on each mental illness scale were categor-
ized as high (for scorers at or above the established
threshold of clinical significance) or low (for scorers
below the cut-off point) for each disorder as based
on established standards. Bivariate analyses used
Chi-square and Student’s t-test to examine crude
Separate logistic regression
developed to predict each sexual risk behaviour
measured, with condom use, previous STD treat-
ment, transactional sex, sex while using alcohol,
sex with a casual partner (defined as someone
known for less than 1 day) and forced sex treated as
independent variables in separate models. The
number of sexual partners was recoded into low
risk (one or less partners) and at risk (more than one
partner). All multivariate models included age,
gender, education, employment and all three
mental illnesses (to account for confounding by
comorbidity) as covariates. In addition, the three
psychological disorders were grouped together to
create an ‘any disorder’ variable (categorized as
disorder presentZ1; disorder absentZ0), and an
additional set of regression models were developed
to determine the relationship between having any
psychiatric disorder and sexual risk behaviours.
as odds ratios (OR) with 95% confidence intervals
(CI). All statistical tests were two-sided at aZ0.05.
In total, 725 participants (64% of those eligible)
completed the study questionnaire. An additional
80 questionnaires (7%) were excluded from the
analysis because they contained incomplete data
for sexual behaviour and/or mental illness
measures. Table 1 describes the 645 participants
included in the study analysis. Slightly more than
half of the participants were female and two-thirds
of the sample had at least some secondary
education. High rates of mental illness were
observed, with 33% of participants reporting
depression, 17% of participants reporting alcohol
J. Smit et al. 536
abuse, and 15% of participants reporting PTSD.
Women were significantly more likely to be
diagnosed with depression compared with men
(PZ0.009). Comorbidity between depression, alco-
hol abuse and PTSD was high (P!0.001 for each
positively associated with previous STD treatment,
but negatively correlated with condom use (data not
shown). Women were more likely to report having
been forced into sex (P!0.001) and having been
treated for an STD in the past (PZ0.038).
Sexual risk behaviours and mental health
In unadjusted analyses, depression was positively
associated with most sexual risk behaviours
(Table 2). However, when adjusted for socio-
demographic factors and psychiatric comorbidity,
only the associations involving self-reported con-
dom use (OR 5.32; 95% CI 2.74–10.29), experience
offorced sex (OR 3.12; 95% CI 1.67–5.84) or previous
treatment for an STD (OR 2.14; 95% CI 1.25–3.66)
remained statistically significant.
Alcohol abuse was significantly associated with
transactional sex, sex while using alcohol or drugs,
and sex with a casual partner in crude analyses. In
the multivariate model, only the association
between sex using alcohol or drugs remained
statistically significant (OR 2.53; 95% CI 1.18–5.42).
PTSD was associated with transactional sex, sex
while using alcohol or drugs, sex with a casual
partner, and previous treatment for an STD.
When these associations were examined in multi-
variate analysis, the associations that involved
participants in a household survey (nZ645) in a peri-urban community outside Cape Town, South Africa.
Demographic characteristics and prevalence of sexual risk behaviours and mental illness among
Total study population
Males (nZ252) Females (nZ357)
Mean age (SD)
Sexual risk behaviour
No condom use in last 6 months
Exchanged sex for money, drugs
or place to stay
Had sex when using alcohol or
Had sex with someone known for
less than 1 day
Forced or pressured into sex
Previous treatment for a sexually
Mean age of first sex (SD)
Mean number of sexual partners
30.3 (11.9) 32.4 (13.2)28.9 (11.3)
7611.8 2710.7 19 5.3
Post-traumatic stress disorder
Comorbidity of mental illness
SD, standard deviation.
Mental health and sexual risk behaviours in South Africa537
(nZ645) in a peri-urban community outside Cape Town, South Africa.
Unadjusted and adjusted associations between different forms of mental illness and sexual risk behaviours among participants in a household survey
Depression Alcohol abuse Post-traumatic stress disorder
Unadjusted analysisAdjusted analysis Unadjusted analysisAdjusted analysis Unadjusted analysisAdjusted analysis
OR95% CIOR 95% CIOR 95% CIOR 95% CIOR 95% CIOR 95% CI
Condom use in
last 6 months
or place to
Had sex while
Had sex with
known for less
than 1 day
More than one
2.95* 1.96–4.36 5.32*2.74–10.29 0.79 0.50–1.260.53 0.26–1.081.26 0.77–2.050.70 0.28–1.76
5.26* 2.67–10.392.35 0.92–5.98 3.43* 1.84–6.401.59 0.63–4.034.86* 2.61–9.05 3.58** 1.34–9.56
2.26**1.35–3.791.52 0.73–3.183.43* 2.01–5.852.53** 1.18–5.422.06** 1.16–3.661.32 0.52–3.33
1.5 0.93–2.440.80 0.39–1.63 1.95**1.14–3.341.27 0.60–2.632.30** 1.32–3.992.61** 1.09–6.25
1.86**1.26–2.713.12*1.67–5.84 1.44 0.89–2.331.620.84–3.13 1.21 0.74–2.010.40**0.17–0.93
2.27*1.58–3.28 2.14** 1.25–3.661.28 0.82–1.990.73 0.39–1.351.90** 1.20–2.991.77 0.83–3.77
2.83**1.05–7.632.15 0.53–8.701.270.53–3.031.22 0.35–4.26 0.810.31–2.141.050.18–6.17
Adjusted associations are from separate logistic regression models including age, gender, education, employment and other mental illnesses. OR, odds ratio; CI, confidence
interval. *P!0.001; **P!0.05.
J. Smit et al.
transactional sex (OR 3.58; 95% CI 1.34–9.56) and
sex with a casual partner (OR 2.61; 95% CI 1.09–
When the three separate psychiatric diagnoses
were combined into a single binary measure for the
presence of any mental illness (Table 3), having any
psychiatric diagnosis was associated with experi-
ences offorced sex [adjusted OR (AOR) 2.53; 95% CI
1.60–4.02], transactional sex (AOR 2.88; 95% CI
1.29–6.48) and increased condom use (AOR 2.07;
95% CI 1.32–3.25).
This is among the first studies on sexual risk
behaviours and mental illness in sub-Saharan Africa.
These data suggest that mental illness appears to be
highly prevalent in this setting, that measures of
different individual disorders appear to be associ-
ated with specific sexual risk behaviours, and that
the presence of any mental illness was associated
with having experienced forced sex and reporting
transactional sex and increased condom use.
The high rates of depression, alcohol abuse,
PTSD and comorbidity between these disorders
were consistent with other South African studies
of the prevalence of psychopathology in peri-urban
areas.21The relationship between mental illness
and sexual risk behaviours, such as multiple
partners, transactional sex, forced sex and previous
STD treatment, were also consistent with data from
developed countries,6,11,31as well as South Africa32
and other developing countries33that show high
rates of these risky phenomena.
This study found that having evidence of mental
illness was associated with experiencing forced sex,
transactional sex and increased condom use.
Recent studies in sub-Saharan Africa reported that
insistence on condom use was related to poor
relationship quality and more forced sexual encoun-
ters,34which may in turn result in psychopathology.
This may partly explain the study findings, with
depression and PTSD strongly associated with
experience of forced or transactional sex. It is
also plausible that individuals living in communities
such as this where HIV is highly prevalent are
acutely aware of the widespread risks of infection.
The positive association between these measures
of mental illness and increased condom use was
unexpected in both the individual depression
analysis and when all measured disorders were
grouped together. This is contrary to other studies
in the published literature which usually note an
inverse association involving condom use.11,31,34
While this finding requires further investigation, it
may suggest that condom use negotiation in this
population is mediated by a range of factors other
than mental illness. There are data from sub-
Saharan Africa showing that condom use is influ-
enced by relationship factors, including trust
between partners and fear of violence,35as well
as cultural factors, such as the meanings attached
to condom use in different sexual relationships.36,37
Future research may help to elucidate important
mechanisms for increasing levels of condom use in
resource-limited settings. Given the widespread
difficulties in promoting condom use, particularly in
high-risk populations (such as individuals with
underlying mental illness), the identification of
such mechanisms could point the way for interven-
tions to raise levels of condom use for the
prevention of STDs including HIV.
Interestingly, the measure of alcohol abuse
employed here did not emerge as a significant
independent risk factor for sexual risk taking in
post-traumatic stress disorder and/or depression) and sexual risk behaviours among participants in a household
survey (nZ645) in a peri-urban community of Cape Town, South Africa.
Unadjusted and adjusted associations between any measured mental illness (alcohol dependency,
Unadjusted analysisAdjusted analysis
OR 95% CIOR 95% CI
Condom use in last 6 months
Exchanged sex for money, drugs or place to stay
Had sex while using alcohol or drugs
Had sex with someone known for less than 1 day
Forced or pressured into sex
Previous treatment for sexually transmitted disease
More than one sexual partner
Adjusted associations are from a logistic regression model including age, gender, education and employment. OR, odds ratio; CI,
confidence interval. *P!0.001; **P!0.05.
Mental health and sexual risk behaviours in South Africa539
these data. Alcohol use has been linked with
transactional and forced sex in South Africa,38but
whether it causes sexual disinhibition or is taken in
specific situations to facilitate sexual engagement
remains unclear.39Self-medication with alcohol can
also mask other psychological problems.40,41It is
possible that the use of drugs other than alcohol,
strongly associated with sexual risk behaviours.
These results should be interpreted with caution
for several reasons. Although the proportion of
randomly selected individuals from the community
who participated in the study (64%) is acceptable for
survey research in resource-limited settings, the
possibility that non-respondents may have been
systematically different from participants cannot be
ruled out. If non-respondents were less likely to
practice and report high-risk sexual behaviours and
also less likely to report evidence of mental illness,
the observed associations in these data may be
artificially inflated. However, the profile of this
sample (mean age 30.3 years; 55% female) is broadly
similar to the demographic characteristics of the
data (mean age 29.9 years; 49% female), suggesting
that any non-response bias is unlikely to fully explain
the study results. While the forms of mental illness
sexual risk taking, it is also possible that, in some
instances, psychopathology may lead to high-risk
of mental illness on increasing risk for HIV infection
and visa versa, but cannot ascertain the temporal
relationship between these. Further prospective
research is needed. In addition, this study focused
on a peri-urban African community characterized by
low socio-economic status as well as high levels of
unemployment and illiteracy; this context is typical
prevalent, but these results may not be immediately
generalizable to developed-country populations or
other parts of sub-Saharan Africa.
It is important to note that the measures of
mental illness used here are screening tools that are
not clinically diagnostic, and thus these findings
cannot be interpreted as definitive data on the
prevalence or correlates of mental illness. In
addition, each of the forms of mental illness
evaluated here are based on self-report which
may have resulted in over-reporting of these
disorders by including symptoms related to other
psychiatric disorders, such as anxiety and somato-
form disorders not examined as part of the study.
The high comorbidity suggests that the self-report
scales may not differentiate adequately between
disorders, although other studies in similar
populations also indicate that these conditions
commonly co-occur.21To deal with this, the
present analysis assessed the behavioural corre-
lates of each form of mental illness separately
(using regression models to identify independent
associations) and as a binary measure of any vs no
Although sexual risk reporting was anonymous
(with no negative consequences for reporting high-
risk behaviours), demand characteristics may have
resulted in biased reports by possibly under-
reporting certain sexual behaviours, over-reporting
condom use42,43and over-reporting emotional
symptoms.44Furthermore, the validity of the
instruments may have been adversely influenced
by the use of either self- or interviewer-adminis-
tered questionnaires, which may have contributed
to unreliable measurements among some partici-
pants. Importantly, the study measures were not
validated, and further research on the validity of
the measures of sexual risk taking and mental
illness in this setting is warranted. However,
interviews were given in the participants’ home
languages (regardless of the medium of adminis-
tration). Furthermore, condom use was found to be
associated with sexual risk behaviours in this study,
suggesting either consistently biased reporting or
the result of previous HIV-related education to
promote condom use (e.g. as part of STD treat-
ment). Similarly, engaging in sexual risk behaviours
was also positively associated with having been
treated for an STD. The consistency in these
responses gives some weight to the credibility of
the data and indicates a level of accuracy in the
Although preliminary in nature, these results may
have important implications for policies to advance
Mental health has received relatively little attention
within efforts to alleviate global inequities in the
burden of disease; where mental health has received
consideration, it is typically considered separately
from HIV and other infectious diseases. However,
given the relatively high prevalence in this setting of
pathology related to alcohol abuse, depression and
PTSD, and the persistent associations between these
conditions and high-risk sexual behaviours, it is
possible that mental illness may play a role in the
spread of HIV/AIDS at population level. For policy
makers, this points to the need for increased
attention to mental health concerns within HIV
prevention efforts, as well as enhanced HIV preven-
tion as part of mental health services. Furthermore,
these inter-relationships suggest that clinicians and
researchers should include routine screening for
J. Smit et al. 540
psychiatric disorders as part of HIV-related
In summary, this study demonstrated that
despite considerable comorbidity, depression,
alcohol abuse and PTSD were independently associ-
ated with different forms of high-risk sexual
behaviours. As the temporal nature of these
associations is not always clear, additional pro-
spective research is required. These findings
emphasize the substantial burden of mental illness
in this setting, and suggest that policy makers,
clinicians and researchers alike should pay greater
attention to the possible interaction between
mental illness and high-risk sexual behaviours in
This research was supported by the South African
AIDS Vaccine Initiative and the Medical Research
Council of South Africa. Prudence Mthimunye was
instrumental in the recruitment, enrolment and
interviewing of participants. Noliswe Malashe,
Ayanda Mhlambiso, Phumla Madliwa and Skolweni
Zelanga conducted the fieldwork for the survey,
and Thabisa Maqweqwana, Ndiseka Nashwa, Lubu-
balo Vellem and Patricia Mhlambiso helped with
translations and the pilot study.
1. UNAIDS. Report on the global aids epidemic. UNAIDS/04.16E.
2004. Geneva: UNAIDS; 2004.
2. Myer L, Morroni C, Susser ES. The social pathology of the
HIV/AIDS pandemic. Int J Epidemiol 2003;32:189–92.
3. Dolezal C, Carballo-Dieguez A, Nieves-Rosa L, Diaz F.
Substance use and sexual risk behavior: understanding
their association among four ethnic groups of Latino men
who have sex with men. J Subst Abuse 2000;11:323–36.
relationship between alcohol use and HIV-related sexual risk-
taking in young people. Addiction 1995;90:319–28.
5. Allers CT, Benjack KJ, White J, Rousey JT. HIV vulnerability
and the adult survivor of childhood sexual abuse. Child
Abuse Negl 1993;17:291–8.
6. Hutton HE, Lyketsos CG, Zenilman JM, Thompson RE,
Erbelding EJ. Depression and HIV risk behaviours among
patients in a sexually transmitted disease clinic. Am
J Psychiatry 2004;161:912–4.
7. Hutton HE, Treisman GJ, Hunt WR, Fishman M, Kendig N,
Swetz A, et al. HIV risk behaviors and their relationship to
posttraumatic stress disorder among women prisoners.
Psychiatry Serv 2001;52:508–13.
8. Melzer-Lange MD. Violence and associated high-risk health
behavior in adolescents. Substance abuse, sexually trans-
mitted diseases, and pregnancy of adolescents. Pediatr Clin
N Am 1998;45:307–17.
9. Stiffman AR, Dore P, Earls F, Cunningham R. The influence of
mental health problems on AIDS-related risk behaviors in
young adults. J Nerv Ment Dis 1992;180:314–20.
10. Wilson AE, Calhoun KS, Bernat JA. Risk recognition and
trauma-related symptoms among sexually revictimized
women. J Consult Clin Psychol 1999;67:705–10.
11. ErbeldingEJ, HuttonHE,
Lyketsos CG. The prevalence of psychiatric disorders in
sexually transmitted disease clinic patients and their
association with sexually transmitted disease risk. Sex
Transm Dis 2004;31:8–12.
12. World Health Organisation. Mental health: new under-
standing, new hope. Geneva: World Health Organisation;
13. Demyttenaere K, Bruffaerts R, Posada-Villa J, Gasquet I,
Kovess V, Lepine JP, Angermeyer MC, Bernert S, de
Girolamo G,Morosini P,
Kawakami N, Ono Y, Takeshima T, Uda H, Karam EG,
Fayyad JA, Karam AN, Mneimneh ZN, Medina-Mora ME,
Borges G, Lara C, de Graaf R, Ormel J, Gureje O, Shen Y,
Huang Y, Zhang M, Alonso J, Haro JM, Vilagut G, Bromet EJ,
Gluzman S, Webb C, Kessler RC, Merikangas KR, Anthony JC,
Von Korff MR, Wang PS, Brugha TS, Aguilar-Gaxiola S, Lee S,
Heeringa S, Pennell BE, Zaslavsky AM, Ustun TB, Chatterji S.
WHO World Mental Health Survey Consortium. Prevalence,
severity, and unmet need for treatment of mental disorders
in the World Health Organization World Mental Health
Surveys. JAMA 2004;291:2581–90.
14. Bowley DM, Khavandi A, Boffard KD, Macnab C, Eales J,
Vellema J, et al. The malignant epidemic: changing patterns
of trauma. S Afr Med J 2002;92:789–802.
15. Stein DJ, Seedat S, Emsley RA. Violence in the world and in
South Africa—what is a doctor to do? S Afr Med J 2002;92:
16. Jewkes R, Penn-Kekana L, Levin J, Ratsaka M, Schrieber M.
Prevalence of emotional, physical and sexual abuse of
women in three South African provinces. S Afr Med J 2001;
17. Peltzer K. Posttraumatic stress symptoms in a population of
rural children in South Africa. Psychol Rep 1999;85:646–50.
18. Ward CL, Flisher AJ, Zissis C, Muller M, Lombard C. Exposure
to violence and its relationship to psychopathology in
adolescents. Inj Prev 2001;7:297–301.
19. FlisherAJ,Ziervogel CF,
Robertson BA. Risk-taking behaviour of Cape Peninsula
high-school students. Part VI. Road-related behaviour. S
Afr Med J 1993;83:486–90.
20. Bhagwanjee A, Parekh A, Paruk Z, Petersen I, Subedar H.
Prevalence of minor psychiatric disorders in an adult African
rural community in South Africa. Psychol Med 1998;28:
21. Carey PD, Stein DJ, Zungu-Dirwayi N, Seedat S. Trauma and
posttraumatic stress disorder in an urban Xhosa primary care
population: prevalence, comorbidity, and service use
patterns. J Nerv Ment Dis 2003;191:230–6.
22. Dorrington RE, Bradshaw D, Budlender D. HIV/AIDS Profile of
the Provinces of South Africa: Indicators for 2002. Cape
Town: Centre for Actuarial Research, Medical Research
Council and the Actuarial Society of South Africa, 2002.
23. Ghubash R, Daradkeh TK, Al Naseri KS, Al Bloushi NB, Al
Daheri AM. The performance of the Center for Epidemiologic
Study Depression Scale (CES-D) in an Arab female commu-
nity. Int J Soc Psychiatry 2000;46:241–9.
24. Iwata N, Buka S. Race/ethnicity and depressive symptoms: a
cross-cultural/ethnic comparison among university students
in East Asia, North and South America. Soc Sci Med 2002;55:
Polidori G, Kikkawa T,
Mental health and sexual risk behaviours in South Africa 541
25. Pretorius TB. Cross-cultural application of the center for
epidemiological studies depression scale: a study of black
South African students. Psychol Rep 1991;69:1179–85.
26. Saunders JB, Aasland OG, Babor TF, De la Fuente JR,
Grant M. Development of the alcohol use disorders
identification test (AUDIT): WHO collaborative project on
early detection of persons with harmful alcohol consump-
tion—II. Addiction 1993;88:791–804.
27. Chinyadsa E, Moyo IM, Katsumbe TM, Chisvo D, Mahari M,
Cock DE, et al. Alcohol problems among patients attending
five primaryhealth care clinics in Harare city. Cent Afr J Med
28. Medina-Mora E, Carreno S, De la Fuente JR. Experience with
the alcohol use disorders identification test (AUDIT) in
Mexico. Recent Dev Alcohol 1998;14:383–96.
29. Halepota AA, Wasif SA. Harvard Trauma Questionnaire Urdu
translation: the only cross-culturally validated screening
instrument for the assessment of trauma and torture and
their sequelae. J Pak Med Assoc 2001;51:285–90.
30. Mollica RF, Caspi-Yavin Y, Bollini P, Truong T, Tor S,
Lavelle J. The Harvard Trauma Questionnaire. Validating a
cross-cultural instrument for measuring torture, trauma,
and posttraumatic stress disorder in Indochinese refugees.
J Nerv Ment Dis 1992;180:111–6.
31. DiClemente RJ, Wingood GM, Crosby RA, Sionean C,
Brown LK, Rothbaum B, et al. A prospective study of
psychological distress and sexual risk behaviour among
black adolescent females. Pediatrics 2001;108:85–97.
32. Reddy P, Meyer-Weitz A, van den BB, Kok G. Determinants of
condom-use behaviour among STD clinic attenders in South
Africa. Int J STD AIDS 2000;11:521–30.
33. Clift S, Anemona A, Watson-Jones D, Kanga Z, Ndeki L,
Changalucha J, et al. Variations of HIV and STI prevalences
within communities neighbouring new goldmines in Tanza-
nia: importance for intervention design. Sex Transm Infect
34. Koenig MA, Lutalo T, Zhao F, Nalugoda F, Kiwanuka N,
Wabwire-Mangen F, et al. Coercive sex in rural Uganda:
prevalence and associated risk factors. Soc Sci Med 2004;58:
35. Campbell C. Selling sex in the time of AIDS: the psycho-social
context of condom use by sex workers in a Southern African
mine. Soc Sci Med 2000;50:479–94.
36. Myer L, Mathews C, Little F. Condom use and sexual
behaviors among individuals procuring free male condoms
in South Africa: a prospective study. Sex Transm Dis 2002;
37. Meyer-Weitz A, Reddy P, Weijts W, van den Borne B, Kok G.
The socio-cultural contexts of sexually transmitted diseases
in South Africa: implications for health education pro-
grammes. AIDS Care 1998;10(Suppl. 3):s39–s55.
38. Wojcicki JM. ‘She drank his money’: survival sex and the
problem of violence in taverns in Gauteng province, South
Africa. Med Anthropol Q 2002;16:267–93.
39. Justus AN, Finn PR, Stern RA. The influence of traits of
disinhibition on the association between alcohol use and risky
sexual behaviour. Alcohol Clin Exp Res 2000;24:1028–35.
40. Khantzian E. The self-medication hypothesis of substance
use disorders: a reconsideration and recent applications.
Harv Rev Psychiatry 1997;4:231–44.
41. Shrier LA, Harris SK, Sternberg M, Beardslee WR. Association
of depression, self-esteem and substance use with sexual
risk among adolescents. Prev Med 2001;33:179–89.
42. Catania JA. A framework for conceptualising reporting bias
and its antecendents in interviews assessing human sexu-
ality. J Sex Res 1999;36:25–38.
43. Dare OO, Cleland JG. Reliability and validity of survey data
on sexual behaviour. Health Transit Rev 1999;4(Suppl.):
44. Myers JK, Weissman MM. Use of a self-report symptom scale
todetect depression in
J Psychiatry 1980;137:1081–4.
J. Smit et al.542