https://orcid.org/0000-0001-8503-4127
Figures
Abstract
Purpose
Adolescent girls are at high risk for depression and human immunodeficiency virus (HIV) acquisition. Poor mental health can increase vulnerability to risky sexual behaviours. Therefore, this study aims to determine the prevalence of depressive symptomology and explore the convergence of HIV risk factors with depressive symptoms amongst cis-gender adolescent girls and young women (AGYW) in rural KwaZulu-Natal (KZN) and peri-urban Western Cape (WC) communities in South Africa.
Methods
Cross-sectional survey data from two sites in South Africa was used - the rural Vulindlela community in KZN and the peri-urban Philippi East community in the WC. Study inclusion criteria included being sexually active with at least one male partner, and not planning to relocate in the next 12 months. The PHQ-9 scale was used to determine depressive symptomology, a socio-behavioural questionnaire was used to determine sexual behaviours, odds ratios and confidence intervals derived from logistic regression models were used to explore the associations between depressive symptomology and socio-behavioural factors associated with HIV acquisition.
Results
The cohort consisted of 274 adolescent girls, 38.6% from the WC site and 61.4% from the KZN site. Overall, 15.7% (43/274) of AGYW reported depressive symptoms. Participants from the peri-urban WC site were more likely to experience depressive symptoms (OR 8.34; 95% CI 3.80–18.30) compared to those living in the rural KZN site. Depressive symptoms were less likely to occur in adolescent girls between the ages of 14 to 17 as compared to those between the ages of 18 and 19 (OR 0.44; 95% CI 0.22–0.90). Socio-behavioural HIV risk factors associated with depressive symptoms include: age disparate relationships (OR 2.98; 95% CI 1.52–5.84), high (four or more) numbers of lifetime partners (OR 8.15; 95% CI 3.60–18.45) and engaging in sex under the influence of alcohol (OR 2.58; 95% CI 1.32–5.04). Multivariate analysis showed that participants from the WC site (AOR 5.25; 95% CI 1.95–14.17) had higher odds of experiencing depressive symptoms while participants with four or more lifetime partners (AOR 3.46; 95% CI 1.24–9.60) were at higher odds of experiencing depressive symptoms.
Citation: Gumbi Z, Mehou-Loko C, Masson L, Mdladla M, Maphumulo N, Passmore J-A, et al. (2025) Exploring the relationship between established HIV risk factors and depressive symptoms amongst young women without HIV in two sites in South Africa. PLoS ONE 20(1): e0317732. https://doi.org/10.1371/journal.pone.0317732
Editor: Denise Evans, University of the Witwatersrand, SOUTH AFRICA
Received: November 24, 2023; Accepted: January 5, 2025; Published: January 29, 2025
Copyright: © 2025 Gumbi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: This research was made possible by the support of an NIH funded program under the award number R01 AI128792 (to JAP and HBJ). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Progress has been made in reducing new HIV infections with a notable 60% decrease observed globally HIV in 2023. However, women and girls represented 44% of all new infections during this period [1]. HIV still continues to be a significant factor contributing to poor health outcomes, particularly for Adolescent Girls and Young Women (AGYW) living in sub-Saharan Africa (SSA). Among those who acquired HIV in 2023, 62% were AGYW aged 15–24 years in SSA [1]. Moreover, AGYW were three times more likely to contract HIV compared to their male peers among adolescents with new infections. Mental health is acritical mediator of poor HIV treatment outcomes and may also increase the risk of HIV acquisition. Understanding the impact of mental health issues on HIV acquisition risk in endemic regions is essential for developing more effective interventions.
Depression - a depressed mood or loss of pleasure, or interest in activities for long periods [2] - is the most prevalent mental health disorder globally [3,4]. There are an estimated 280 million cases of depression worldwide [2], with ~ 80% of these occurring in low-and middle-income (LMIC) countries [4]. Similar to trends in HIV, young women aged 14–25 years are particularly vulnerable to depression, being three times more likely to report depression than their male peers [5]. In South Africa, data from a large national survey conducted in 2014 reported that one in four individuals experienced depression, with heterogeneity in prevalence across genders [6].
Clear evidence demonstrates a link between depression and HIV in sub-Saharan Africa. Depression negatively influences treatment adherence and mortality amongst people living with HIV, and an HIV diagnosis may contribute to experiencing depression as it is a life-threatening condition, is stigmatised and may have serious negative social consequences. Depression’s role in enhancing HIV acquisition remains an area requiring more research. Data from a randomised control trial conducted in Mpumalanga, South Africa suggests that socio-behavioural mediators explained two-thirds (indirect effect: OR = 2.4, 95% CI: 0.2–4.5) of the association between depressive symptoms and HIV incidence [7]. In AGYW living in Kenya, moderate-to-severe depression was associated with higher odds of being in age disparate relationships, engaging in transactional or forced sex, experiencing intimate partner violence and higher alcohol use, all of which are also well described risk factors for HIV infection [8]. Additionally, economic stress, lack of family support, substance use, unemployment, food insecurity, alcohol use, intimate partner violence, and childhood traumas have been shown to enhance depressive symptomology and are also risk factors for HIV infection [9–13].
There is a lack of evidence on depression amongst AGYW in general in South Africa, and in particular amongst rural AGYW. Understanding the association between depressive symptomology and HIV risk behaviours, and how this differs for AGYW living in urban and rural areas provides novel insight into this relationship and could inform future intervention development. The present analysis aimed to investigate the relationship between depressive symptoms (measured using the PHQ-9) and HIV risk behaviour among AGYW in two different HIV-endemic locations in South Africa.
Methods
Study setting
The Centre for the AIDS Programme of Research in South Africa (CAPRISA) site is based in the rural Vulindlela area in the uMgungundlovu district of KwaZulu-Natal (KZN), South Africa. The area is predominantly isiZulu speaking marked by high levels of poverty, poor infrastructure and high levels of sexually transmitted infections (STI) and HIV prevalence [14]. The Desmond Tutu Health Foundation site is located in the Mitchell’s Plain township area in Cape Town, Western Cape (WC), South Africa. Participants in that area face high levels of gangsterism, crime and violence, substance abuse, teenage pregnancy, and high levels of STI, HIV and tuberculosis [15].
Cohort description
This study was nested within a longitudinal cohort study called MIST/MISC (Mucosal Injury from Sexual Trauma/Contact). MIST was a prospective, longitudinal observational study that enrolled adolescent girls (14–19 years) and adult women (25–35 years) at two sites in South Africa: 1) a rural site based at the Vulindlela Clinical Research (VRC) site at CAPRISA in Mafakathini, KZN from the 1st of June 2018 until the 30th of June 2021, and 2) a peri-urban site based within the Desmond Tutu Health Foundation (DTHF) Adolescent Clinic situated in Philippi East, WC from 1st of July 2019 until the 30th of November 2021. The MIST study aimed to investigate unique socio-behavioural, reproductive tract anatomical and biological characteristics around sexual debut in adolescent females (14–19 years) in response to early sexual exposure, male semen, vaginal products, mucosal trauma and wound healing compared to a group of older women (25–35 years). Both sites recruited study participants by mobilising existing networks and general community outreach activities. Eligibility criteria at both sites included being sexually active with at least one male partner to meet the aims of the parent study, being willing to provide contact information and not planning on moving out of the area in the next 12 months. Participants in the study were screened for pregnancy using urine pregnancy tests and for HIV using Uni-Gold™ Recombigen® HIV-1/2 (Trinity Biotech, Ireland) and Determine™ HIV-1/2 test (Abbott Diagnostics Scarborough, Inc., USA) in KZN. In WC, they were tested for HIV using One Step Anti-HIV (1&2) (Intec Products Inc., China), and HIV 1 + 2 Rapid Test (Wantai BioPharm, China). Participants were excluded if they were pregnant (due to extensive genital specimen sampling) or if they tested positive for HIV. They were then referred for appropriate antenatal care and/or HIV management. Participants aged 16 years and older provided written informed consent. The ethics committees in KZN and WC waived the requirement for parental consent for 16–17-year-olds due to the sensitive nature of the study. For girls aged 14–15 years, both assent and parental consent were obtained. Both study sites received ethical approval from their respective ethics committees, the Biomedical Research Ethics Committee at the University of KwaZulu Natal (UKZN) (BF504/17) and the Human Research Ethics Committee at the University of Cape Town (UCT) (696/2017).
Study tools
At enrolment, a socio-behavioural questionnaire was completed at both study sites by participants with the assistance of trained nurses. Socio-behavioural variables were selected for inclusion based on a review of previous literature about factors that may be associated with increased HIV vulnerability and depression as well as factors hypothesized to be linked to depression and HIV in the South African context, including: (1) socio-demographic factors: age, location, education, employment status, financial support, and death in the household; (2) sexual history and behaviour: consent and age at sexual debut, contraception use, number of lifetime partners, partner age gap (a peer partner can be defined as a partner who is ≤ 5 years younger than participant, and a non-peer is ≥ 5 years older than participant), condom use at last sex act, use of sexual enhancers, engagement in transactional sex or sex under the influence of alcohol, and pregnancy history; and (3) diagnosed biological factors and self-reported substance use: body mass index (BMI), Chlamydia trachomatis (CT), Neisseria gonorrhoeae (NG), Trichomonas vaginalis (TV), bacterial vaginosis (BV), and substance use ever. STI testing was done using the Xpert® CT/NG assay (Cepheid, USA; KZN site), OSOM® Trichomonas Rapid Test (Sekisui Diagnostics, USA; KZN site) or genesig® kits (Primerdesign™ Ltd, UK; WC site). BV was diagnosed using Nugent scoring.
A patient health questionnaire (PHQ)-9 was completed by all women with assistance from a trained lay counsellor. The PHQ-9 is a depressive symptom screening tool that scores each of the nine DSM-IV (Diagnostic and Statistical Manual of Mental Disorders) criteria from “0” meaning not at all, to “3” meaning nearly every day [16]. This depression screening tool has been validated in South Africa [17]. The sum of each participant’s reported symptoms is then calculated and classified as being mild (scores of 5–9 out of 27), moderate (scores of 10–14 out of 27), moderately severe (scores of 15–19 out of 27), or severe (scores of 20–27 out of 27) [16]. For the current analysis, we considered a participant to exhibit depressive symptomology if she scored ≥ 5. Those who scored in the categories of mild to severe depressive symptomology received in-house counselling from qualified lay counsellors and were provided with a referral to a local social worker, helpline, or relevant local mental health services. In addition, those responding affirmatively to a question about contemplating self-harm or expressing suicidal ideation were immediately referred for assistance, and a facilitated referral to specialist services was completed.
The study teams translated both the socio-behavioural and PHQ-9 questionnaires from English to isiZulu and isiXhosa, then back-translated them into English to ensure reliability. Participants could choose the language of the questionnaire they were most comfortable with; isiZulu in KwaZulu Natal, isiXhosa in Cape Town, or English at both sites.
Statistical analysis
The associations between sociodemographic characteristics, sexual behaviour, health-related behaviours, and substance use of participants with depressive symptoms, defined by a PHQ-9 score of ≥ 5, were compared to participants without depressive symptoms, defined by a PHQ-9 score of ≤ 4 using logistic regression. The logistic regression model has become the primary approach for examining relationships between binary outcomes and a set of exposure variables, and it offers the advantage of controlling for multiple confounders [18,19]. A univariable regression was performed first to assess whether there is an association between the outcome and each predictor. Multivariable regression was then used to investigate the association between depressive symptomology and a combination of several variables selected based on the literature and those found to be significant in the univariable regression (level of significance = p ≥ 0.05).
The association is described using unadjusted and adjusted odds ratios (AOR) with 95% confidence interval (CI) to provide evidence of precision. An OR of 1 or a CI that includes 1 indicates no association between the predictor and depression. The total number of participants included in the logistic regression model (univariable and multivariable) varied depending on the completeness of the predictor variables. Missing values were excluded from the regression model, the proportion of missingness was low ranging from 0% to approximately 6% except for miscarriage status (15.7%). The proportion of missing values in the multivariable model was 5.5%. We assessed the goodness of fit on the final logistic regression model using the Hosmer-Lemeshow test. All statistical analysis was done using SAS, version 9.4 [20].
Results
Demographic characteristics of the study cohort
Of the 274 sexually active AGYW participants in this study, 106/274 (38.6%) were from the WC site and 168/274 (61.4%) were from the KZN site (Table 1 and S1 Fig Consort diagram). Based on the PHQ-9 assessment, 15.7% of the participants were classified as experiencing depression. The majority of participants with depressive symptoms at both the WC and KZN sites scored in the mild depression range, with 9.5% and 2.9% in these ranges, respectively. Only 0.7% of participants from the WC site scored in the severe depression range, while none from the KZN site scored in this category. PHQ-9 scores distributions indicated that the WC site had participants across all PHQ-9 categories, while most participants from the KZN site scored within a mild range (Fig 1). The median PHQ-9 score was significantly higher for the WC site (4; IQR 2–5) compared to the KZN site (0; IQR 0–0) (p < 0.001). Most study participants from both locations received financial support from family and were high school students, though 1.8% of participants from the WC site reported having no formal education. The prevalence of sexually transmitted infections was high, with an overall prevalence of 56.8% (Table 1).
Factors associated with depressive symptomology
In the univariate analysis (Table 2), younger adolescents (14–17 years) were less likely to exhibit depressive symptoms compared to older adolescents (Odds ratio (OR) 0.44; 95% confidence interval (CI) 0.22–0.90). Location was a strong predictor, with AGYW from the WC site having significantly higher odds of experiencing depressive symptoms (OR 8.34, 95% CI 3.80–18.30). Several risky sexual behaviours, including having more than four lifetime partners (OR 8.151; 95% CI 3.60–18.45), having a non-peer partner (OR 2.98, 95% CI 1.52–5.84), and engaging in sex under the influence of alcohol (OR 2.58; 95% CI 1.32–5.04) were all associated with increased odds of depressive symptomology.
Conversely, receiving financial support from more than one source (such as family, social grants, or partners) (OR 0.13; 95% CI 0.02–1.01) and the use of sexual enhancers (OR 0.24; 95% CI 0.10–0.57) were associated with lower odds of depressive symptomology, irrespective of location. Condom use was associated with higher odds of depressive symptoms in AGYW (OR 2.57; 95% CI 1.32–5.00). Additionally, experiences of miscarriage (OR 7.42; 95% CI 1.20–45.93), pregnancy termination (OR 5.13; 95% CI 2.00–13.17), and contraceptive use (OR 3.46; 95% CI 1.31–9.16) were all linked to increased odds of depressive symptoms.
Variables that were significantly associated with depressive symptoms, in the univariate analysis, were further examined to identify which factors most contributed to the PHQ-9 depression categories (Fig 2). The PHQ-9 “Mild” category (scores of 5–9) was most prevalent across each significant variable. Variables such as having a higher number of lifetime sexual partners (2–3 partners), engaging in sex under the influence of alcohol, having a non-peer partner, not using a condom during intercourse, and terminating a pregnancy showed relatively high proportions within the “Severely Depressed”range (PHQ-9 scores of 20–27). The distribution of participants across each PHQ-9 category by site, as well as the socio-behavioural variables contributing to each level of depressive symptoms is shown in Fig 2.
The effects of age, location, number of lifetime partners, age difference with sexual partner, condom use, sex under the influence of alcohol and financial support were assessed using multivariate logistic regression (Table 2). The Hosmer-Lemeshow test results (Table 3) indicate that the model fits the observed data well, with a p-value of 0.17, suggesting no significant difference between observed and predicted values. Location was associated with more than a fivefold increase in the adjusted odds of experiencing depressive symptoms among participants in the study cohort (adjusted odds ratio [AOR] = 5.25; 95% CI 1.95–14.17). Additionally, having a higher number of lifetime partners (four or more) tripled the odds of depressive symptoms (AOR = 3.46; 95% CI 1.24–9.60).
Discussion
Adolescent depression is an important public health issue that requires urgent and focused attention, especially in a setting as complex and diverse as South Africa. The prevalence of depressive symptomology obtained in this study (15.7%) is quite similar to what has been reported in the literature. Craig et al. suggest that the prevalence of depression in the South African adolescent population using the PHQ-9 ranges from 14.7% to 38.3% [21]. Age, location, multiple sexual partners, sex under the influence of alcohol, partner age gap, previous miscarriage, termination of pregnancy and contraceptive use were all significantly associated with increased odds of depressive symptomology in the univariate analysis. These are factors that, according to the literature, have been known to influence HIV risk in AGYW. However, these did not present as significant predictors in the multivariate analysis. Multivariate analysis found location and multiple sexual partners to be key influencers in the risk of presenting depressive symptoms.
Although the univariate data does not provide evidence of a cause, it is evident in research that multiple sexual partners, sex under the influence of alcohol and age disparate relationships have been associated with increased HIV acquisition risk [22–24]. The experience of a stressor such as the termination of an unexpected pregnancy results in a higher risk of depressive symptomology and binge drinking [25]. Our univariate findings suggest that these socio-behavioural factors are interconnected with depressive symptomology and HIV acquisition risk, and these interactions should be explored further.
The striking difference between the prevalence of depression between the two sites and the odds related to the WC site (WC site = 34% prevalence versus KZN site = 9% prevalence; 5 times greater odds of experiencing depressive symptomology when residing in WC) could be attributed to the differences in the characteristics of each community. The rural Vulindlela KZN site is a community characterised by traditional settlements or farmlands and an HIV prevalence of 36.3% [26,27], while the WC site in the area of peri-urban Philippi East is characterized by high unemployment rates, low per capita income, frequent instances of gangsterism and an HIV prevalence of 24%, the highest in the WC province [28–30]. Our study showed similar results to studies [31,32] where AGYW in urban settings were at greater risk of experiencing depression compared to those in more rural settings. Township settings often present increased exposure to various forms of interpersonal violence, crime, and socio-economic deprivation, that can negatively impact the mental and emotional wellbeing of those living in these areas [33]. Additionally, township settings often include people from multiple locations who may have less social coherence, and fewer formal support structures [34]. Rural settings often include households characterised by strong family connections, long established households, and a stronger sense of community with more collective support structures in place [34]. Structural differences between provinces may also mediate the risk of reporting depressive symptoms. A study by Herman et al [35], concurrent with our findings, showed that the WC province had the highest prevalence of mental health disorders when compared to other South African provinces. Geographical differences may be due to the Western Cape being one of the first regions to be colonised and is highly urbanised as compared to provinces with more rural areas such as KZN [35].
Secondly, our analysis shows that having multiple partners (4 or more) was strongly associated with increased odds (3.5 times likelihood) of being depressed. Considering that having multiple partners is also associated with increased HIV risk, this may indicate a point of synergy between depression and increased HIV risk. This result is similar to findings from other African settings, where reporting multiple partners; a known risky behaviour for HIV acquisition, was associated with increased odds of having depressive symptoms [36]. It is unclear whether depressive symptoms could increase the likelihood of AGYW engaging in risky sexual practices such as having multiple partners or rather that engaging in risky sexual behaviour may cause depression. This relationship needs to be explored in longitudinal research studies. Importantly, the higher odds of having multiple partners amongst AGYW showing depressive symptomology have important implications for their risk when we consider the high prevalence of other risk factors present in this sample. The majority of AGYW in our study reported not using a condom during sexual acts (69.4%), a high prevalence of non-peer relationships (26.4%), high rates of sex under the influence of alcohol (28.7%) and a prevalence of sexually transmitted infections (56.8%).
Collectively, the findings in this study highlight a possible synergy between depressive symptomology and HIV risk behaviours. If social factors (i.e., poverty, social dependence) enhance the risk of depressive symptomology and HIV, and depressive symptomology also increases the risk of HIV infection on its own, depression may serve to enhance the risk of both on its own and by affecting other pathways resulting in increased HIV risk.
There are some limitations to the current analysis. The data is cross-sectional and therefore we cannot establish causality. Future longitudinal studies are needed to investigate the causal pathways between the risk factors associated with HIV and depression. This study is nested under the ongoing parent study, and therefore the risk factors that were analysed were those limited to the data already collected from the parent study. Future exploration will require more in-depth and targeted data on all factors that could impact the risk of depression in the two different contexts. Nonetheless, this study highlights a need to establish more varied and nuanced risk profiles as all AGYW are not equally at risk of HIV or depression, however, those who present with one condition should also be screened for the other. Finally, future work should explore the rural site/urban divide within the same province, rather than different provinces.
Conclusion
In conclusion, this research suggests an important connection between depression and HIV risk factors, highlighting the need for more comprehensive HIV prevention strategies. To effectively address the HIV epidemic, interventions must consider mental health screening and provide tailored psychosocial support, particularly for individuals living in urban and township areas. Furthermore, HIV risk assessments and prevention measures, including pre-exposure prophylaxis, should also incorporate mental health evaluations. By recognising and addressing these interconnected factors, we can work towards a more holistic approach to HIV prevention that may improve health outcomes and contribute to the resilience of communities.
Supporting information
S1 Fig. Consort diagram describing the cohort of adolescent girls included in the depressive symptomology sub-study.
https://doi.org/10.1371/journal.pone.0317732.s001
(TIF)
Acknowledgments
The study team would like to extend their utmost gratitude to the study participants and the study research team in KwaZulu-Natal and Cape Town.
References
- 1.
Joint United Nations Programme on HIV/AIDS (UNAIDS). Global HIV & AIDS statistics - fact sheet. Retrieved October 2024. Available from: https://www.unaids.org/en/resources/fact-sheet
- 2. World Health Organization (WHO). Depressive disorder (depression) fact sheet. Retrieved 19 June 2024. Available from https://www.who.int/news-room/fact-sheets/detail/depression
- 3. Patel V, Flisher AJ, Nikapota A, Malhotra S. Promoting child and adolescent mental health in low and middle income countries. J Child Psychol Psychiatry. 2008;49(3):313–34. pmid:18093112
- 4. Ferrari AJ, Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJL, et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS Med. 2013;10(11):e1001547. pmid:24223526
- 5. Albert PR. Why is depression more prevalent in women?. J Psychiatry Neurosci. 2015;40(4):219–21. pmid:26107348
- 6. Mutyambizi C, Booysen F, Stornes P, Eikemo TA. Subjective social status and inequalities in depressive symptoms: a gender-specific decomposition analysis for South Africa. Int J Equity Health. 2019;18(1):87. pmid:31196102
- 7. Goin DE, Pearson RM, Craske MG, Stein A, Pettifor A, Lippman SA, et al. Depression and Incident HIV in adolescent girls and young women in HIV prevention trials network 068: targets for prevention and mediating factors. Am J Epidemiol. 2019;189(5):422–32.
- 8. Larsen A, Kinuthia J, Lagat H, Sila J, Abuna F, Kohler P, et al. Depression and HIV risk behaviors among adolescent girls and young women seeking family planning services in Western Kenya. Int J STD AIDS. 2020;31(7):652–64. pmid:32538330
- 9. Bonner CP, Carney T, Browne FA, Ndirangu JW, Howard BN, Wechsberg WM. Substance use and depressive and anxiety symptoms among out-of-school adolescent girls and young women in Cape Town, South Africa. S Afr Med J. 2020;111(1):40–5. pmid:33404004
- 10. Cheng Y, Li X, Lou C, Sonenstein FL, Kalamar A, Jejeebhoy S, et al. The association between social support and mental health among vulnerable adolescents in five cities: findings from the study of the well-being of adolescents in vulnerable environments. J Adolesc Health. 2014;55(6 Suppl):S31–8. pmid:25454000
- 11. Gibbs A, Dunkle K, Washington L, Willan S, Shai N, Jewkes R. Childhood traumas as a risk factor for HIV-risk behaviours amongst young women and men living in urban informal settlements in South Africa: a cross-sectional study. PLoS One. 2018;13(4):e0195369. pmid:29624612
- 12. Gibbs A, Govender K, Jewkes R. An exploratory analysis of factors associated with depression in a vulnerable group of young people living in informal settlements in South Africa. Glob Public Health. 2018;13(7):788–803. pmid:27533487
- 13. Kuringe E, Materu J, Nyato D, Majani E, Ngeni F, Shao A, et al. Prevalence and correlates of depression and anxiety symptoms among out-of-school adolescent girls and young women in Tanzania: a cross-sectional study. PLoS One. 2019;14(8):e0221053. pmid:31419238
- 14. George G, Beckett S, Cawood C, Khanyile D, Govender K, Kharsany ABM. Impact of HIV testing and treatment services on risky sexual behaviour in the uMgungundlovu District, KwaZulu-Natal, South Africa: a cross-sectional study. AIDS Res Ther. 2019;16(1):20. pmid:31434574
- 15. Chetty R. Social complexity of drug abuse, gangsterism and crime in Cape Flats’ schools, Western Cape.. Acta Criminologica. 2015;2015(sed-3):54–65.
- 16. Kroenke K, Spitzer RL, Williams JB. The PHQ-9 validity of a brief depression severity measure. J Gen Intern Med. 2001.
- 17. Cholera R, Gaynes BN, Pence BW, Bassett J, Qangule N, Macphail C, et al. Validity of the Patient Health Questionnaire-9 to screen for depression in a high-HIV burden primary healthcare clinic in Johannesburg, South Africa. J Affect Disord. 2014;167:160–6. pmid:24972364
- 18. Fekedulegn D, Andrew M, Violanti J, Hartley T, Charles L, Burchfiel C. Comparison of statistical approaches to evaluate factors associated with metabolic syndrome. J Clin Hypertens (Greenwich). 2010;12(5):365–73. pmid:20546380
- 19. Barros AJD, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol. 2003;3:21. pmid:14567763
- 20.
SAS Institute Inc. SAS/STAT® 15.3 User’s Guide. Cary, NC: SAS Institute Inc.; 2023.
- 21. Craig A, Rochat T, Naicker SN, Mapanga W, Mtintsilana A, Dlamini SN, et al. The prevalence of probable depression and probable anxiety, and associations with adverse childhood experiences and socio-demographics: a national survey in South Africa. Front Public Health. 2022;10:986531. pmid:36388391
- 22. Delany-Moretlwe S, Nanoo A, Nagpal A, Moultrie H, Delany S, Rees H. P1-S5.14 Risk factors associated with HIV acquisition: a comparative analysis of older and younger women who participated in the MDP301 trial in Johannesburg. Sex Transm Infect. 2011;87(Suppl 1):A179.3–80.
- 23. Naicker N, Kharsany ABM, Werner L, van Loggerenberg F, Mlisana K, Garrett N, et al. Risk Factors for HIV Acquisition in high risk women in a generalised epidemic setting. AIDS Behav. 2015;19(7):1305–16. pmid:25662962
- 24. Meier A, Erickson GA, McLaughlin H. Older sexual partners and adolescent females’ mental health. Perspect Sex Reprod Health. 2016;48(1):25–33. pmid:26918326
- 25. Olsson CA, Horwill E, Moore E, Eisenberg ME, Venn A, O’Loughlin C, et al. Social and emotional adjustment following early pregnancy in young Australian women: a comparison of those who terminate, miscarry, or complete pregnancy. J Adolesc Health. 2014;54(6):698–703. pmid:24438851
- 26. Kharsany ABM, Cawood C, Khanyile D, Grobler A, Mckinnon LR, Samsunder N, et al. Strengthening HIV surveillance in the antiretroviral therapy era: rationale and design of a longitudinal study to monitor HIV prevalence and incidence in the uMgungundlovu District, KwaZulu-Natal, South Africa. BMC Public Health. 2015;15:1149. pmid:26588902
- 27. Kharsany ABM, Cawood C, Khanyile D, Lewis L, Grobler A, Puren A, et al. Community-based HIV prevalence in KwaZulu-Natal, South Africa: results of a cross-sectional household survey. Lancet HIV. 2018;5(8):e427–37. pmid:30021700
- 28. Nardell MF, Lee YS, Rousseau E, Julies R, Klaas P, Vundhla P, et al. “You are not alone”: a qualitative study to explore barriers to ART initiation and implications for a proposed community-based youth treatment club among young adults newly diagnosed with HIV in South Africa. AIDS Care. 2021;33(7):952–61. pmid:33345593
- 29. Nash S, Dietrich J, Ssemata AS, Herrera C, O’Hagan K, Else L, et al. Combined HIV Adolescent Prevention Study (CHAPS): comparison of HIV pre-exposure prophylaxis regimens for adolescents in sub-Saharan Africa-study protocol for a mixed-methods study including a randomised controlled trial. Trials. 2020;21(1):900. pmid:33121503
- 30. Petrus T, Kinnes I. New social bandits? a comparative analysis of gangsterism in the Western and Eastern Cape provinces of South Africa. Criminol & Criminal Just. 2018;19(2):179–96.
- 31. Ajaero CK, Nzeadibe CT, Igboeli EE. Rural-urban differences in the prevalence and predictors of depression among adolescents in South Africa. S Afr J CH. 2018;12(2b):71–4.
- 32. Barhafumwa B, Dietrich J, Closson K, Samji H, Cescon A, Nkala B, et al. High prevalence of depression symptomology among adolescents in Soweto, South Africa associated with being female and cofactors relating to HIV transmission. Vulnerable Child Youth Stud. 2016;11(3):263–73.
- 33. Naledi T, Little F, Pike C. Women of worth: the impact of a cash plus intervention to enhance attendance and reduce sexual health risks for young women in Cape Town, South Africa.. J Int AIDS Soc. 2022;25(6):e25938. pmid:35700052
- 34. Mckenzie K. Urbanization, social capital and mental health. Glob Soc Pol. 2008;8(3):359–77.
- 35. Herman A, Stein D, Seedat S. The South African Stress and Health (SASH) Study: 12-month and lifetime prevalence of common mental disorders. S Afr Med J. 2008;98(4).
- 36. Othieno CJ, Okoth R, Peltzer K, Pengpid S, Malla LO. Risky HIV sexual behaviour and depression among University of Nairobi students. Ann Gen Psychiatry. 2015;1416. pmid:25873984