Why counting traumatic events says little about student mental health in South Africa

Introduction

Walk into almost any lecture hall at a South African university and most of the students in it will have lived through something frightening. That is roughly what the survey data suggest. Padmanabhanunni and Pretorius (2025), who studied 491 students at a Western Cape university, begin by pointing to earlier work in which 97.6% of students reported at least one traumatic event, with physical assault and transport accidents at the top of the list. The national student survey of 28,268 students at 17 universities puts post-traumatic stress disorder among the most common conditions, with 30-day prevalence estimates between 21.0% and 24.5% (Bantjes et al., 2023). And in a mostly female, mostly Black African sample across nine public universities and TVET campuses, 21% of students had thought about suicide (Machisa et al., 2022).

My argument is simple enough to state. When exposure is close to universal, a count of traumatic events has very little power to tell anyone who is struggling and why. The more useful questions are which exposures matter, in what relationships and conditions they happen, which students the usual checklists overlook (Adedolapo et al., 2025), and whether the university has anything workable to offer afterwards (Wao et al., 2025). The evidence for answering those questions is thinner than the headline figures make it look, since much of it is cross-sectional, self-reported and drawn from a few institutions (Bantjes et al., 2023; Myers et al., 2021). Even so, put side by side it points in a fairly consistent direction, and that direction is what this whole thing hangs on. The paper looks first at what high prevalence figures can and cannot show, then at childhood maltreatment and sexual and partner violence, then at students who fall outside the standard picture, and finally at services.

What a very high prevalence figure can and cannot show

A figure like 97.6% leaves almost nothing to explain. If virtually every student has been exposed, exposure by itself cannot be what separates the student who copes from the one who stops coming to class. Padmanabhanunni and Pretorius (2025) report that in the earlier work men were more likely to report physical assault and assault with a weapon, while women were more likely to report unwanted sexual contact and sexual assault. So there is some pattern underneath the headline, and it is gendered. But their own study draws on a single university, with 64.8% women and 92.5% undergraduates, and the authors accept that a cross-sectional design and self-report limit what can be concluded (Padmanabhanunni & Pretorius, 2025). The number is a decent alarm bell. Beyond that it does very little work.

Outcome figures wobble a lot depending on who measured what and how. The national survey, using weighted data, estimated 30-day prevalence of any anxiety disorder at 37.1%, any mood disorder at 16.3% and any substance use disorder at 6.6% (Bantjes et al., 2023). A survey of undergraduates at the University of the Witwatersrand reached a much higher number, with 48% (439 of 910) screening positive for probable depression (Croock et al., 2023). Those two results can look like they disagree, but a brief screening tool is a different animal from a disorder estimate, and the Wits survey had a response rate of only 8.4%, so the students who answered may well have been the ones with something to say. Meintjes (2024) gives a third angle. Among first-year students in an extended programme, 32.3% in 2022 reported severe or extremely severe depression on the DASS-21, almost the same as the 31.2% found in a 2017 cohort, which suggests the pandemic left little mark on those figures.

What the South African context changes is that exposure and inequality arrive together. Bantjes et al. (2023) found that prevalence varied by the historical segregation status of institutions, and their conclusion singles out Black students at historically white institutions as one group needing attention. At Wits, reporting white race and having more spending power were each linked to lower odds of probable depression, with an adjusted odds ratio of 0.44 for students who had enough money for luxuries and extras (Croock et al., 2023). Neither study measured trauma in a way that lets me say unequal trauma explains unequal distress, and I would not claim that. Still, the pattern implies that a student’s starting position, and not just what happened to them, shapes how heavily exposure lands.

Childhood maltreatment, partner violence and rape

Myers et al. (2021) took a different line by asking first-year students about childhood maltreatment. In a web-based survey of 1,290 students at two large urban universities, 48.4% reported some form of maltreatment, and emotional abuse was the most common at 26.7%. After adjusting for the other types and for the number of types, emotional abuse was the only kind independently associated with 12-month major depressive disorder, generalised anxiety disorder or drug use disorder. That sits awkwardly beside checklists that lead with assault and accidents (Padmanabhanunni & Pretorius, 2025). Campus talk about trauma tends to circle around the dramatic events, whereas this finding implies the quieter ones, the emotional abuse at home, may matter more to everyday mental health than anyone is counting. Two campuses and self-report mean it is a signal and not a settled result (Bantjes et al., 2023).

Sexual and partner violence show up differently. Machisa et al. (2022) surveyed 1,292 students aged 18 to 30 at nine public universities and TVET campuses. About 50% had binge drunk, 43% reported depressive symptoms, 9% reported PTSD symptoms and 21% had suicidal thoughts. Using structural equation modelling, they found that childhood trauma, other traumatic events, food insecurity, recent intimate partner violence and non-partner rape each fed into a combined mental ill-health outcome, and that partner violence carried part of the effect of childhood and other trauma. In plain terms, harm early in life seems to travel into harm in relationships at university. The 9% PTSD figure looks low next to the 21.0% to 24.5% in the national survey (Bantjes et al., 2023), but one is a 30-day prevalence estimate from a weighted national sample and the other a symptom count from a purposefully selected set of nine campuses, so they cannot be lined up neatly.

Alcohol sits awkwardly in that model. Binge drinking is often described as something students do to cope, and it may well be, but the model also had binge drinking directly raising the chance of non-partner rape, and partner violence mediated the link between binge drinking and other traumatic events. So drinking appears on both sides of the arrow, as a possible result of trauma and a possible route to more of it. Because the data were cross-sectional, the direction of those arrows came from the researchers’ model and not from watching students over time (Machisa et al., 2022). Mahlangu et al. (2025) describe a similar loop from the qualitative side, because lesbian, bisexual and queer women students in their focus groups tied violence in their relationships to past traumas and previous experience of violence.

The gender gap is probably the most consistent thing in these studies. In Meintjes’s (2024) 2022 cohort, 39.4% of women reported severe or extremely severe depression against 18.8% of men, and 35.6% of women reported severe or extremely severe stress against 5.9% of men. That study did not measure trauma at all, so it cannot say why, and I do not want to read violence into numbers that do not contain it. Still, it fits with the higher reports of sexual assault among women (Padmanabhanunni & Pretorius, 2025) and with the national survey’s call to pay attention to female students (Bantjes et al., 2023).

Students the usual checklists tend to miss

Xenophobia is a harm that a standard trauma checklist would probably not register at all. Adedolapo et al. (2025) surveyed African postgraduate students at one South African higher education institution between December 2020 and March 2021. Of the 161 who completed the survey, 72, close to 45%, reported experiencing xenophobia, and those experiences were associated with depression, anxiety and PTSD symptoms but not with alcohol use. Social support moderated the link between xenophobia and PTSD severity. The response rate was under 13% and the sample was one of convenience, so the prevalence is shaky (Adedolapo et al., 2025). Even so, the link with PTSD symptoms is what matters here, because it implies that trauma-type distress can come from everyday hostility and not only from a single event of the kind that tops the prevalence figures summarised by Padmanabhanunni and Pretorius (2025).

Mahlangu et al. (2025) show a different gap. Through focus group discussions with 56 Black African students who identified as lesbian, bisexual or queer women on three campuses of one university, they documented physical, sexual and emotional partner violence that was often bidirectional, along with discrimination and disregard for bisexual women. Participants linked the violence to past traumas, heteronormativity and gender norms, and described depression, anger, self-hate and damage to their academic work. The authors note that research on partner violence has mostly looked at heterosexual relationships, which means a lot of what we think we know about campus violence rests on a narrow base (Mahlangu et al., 2025). At national level, Bantjes et al. (2023) describe the risks tied to sociodemographic factors as modest, yet they still point to gender nonconforming and sexual minority students as groups needing attention. Small qualitative work like this is where the mechanisms actually become visible, in a way a prevalence table never manages.

Then there is the matter of hunger and money. In the Machisa model food insecurity was linked with mental ill-health, and non-partner rape carried part of that link, which suggests that some of what gets called trauma is entangled with deprivation (Machisa et al., 2022). The Wits results line up loosely with this, since students with more spending power had lower odds of probable depression (Croock et al., 2023). I would not push this too far. Nobody here has shown that going without food causes rape or that money prevents depression, and the 8.4% response rate leaves the Wits prevalence open to question. But the practical point stands. If a counselling service takes a student’s exposure history and ignores the empty fridge, it may be treating only half the problem.

What campuses can realistically offer

Formal help is hard to reach, as far as the African evidence goes. Wao et al. (2025) reviewed 18 studies on mental health services for university students in Africa and found more informal than formal support available, accessible and used, with parents and lecturers the usual informal sources. Low mental health literacy and money problems got in the way, and neither is easy to fix from inside a counselling centre. That is a continental review and not a South African one, so I would be careful about transferring it wholesale. Still, Croock et al. (2023) reached a comparable conclusion about Wits, calling for greater awareness and use of counselling among undergraduates, which hints that even where a service exists many students do not get to it. If lecturers are the first people students turn to, they are, whether they like it or not, part of the trauma response (Wao et al., 2025).

There is some encouraging news on delivery. Bantjes et al. (2024) randomised 371 South African undergraduates with anxiety or depression to remote group CBT skills training, a gamified self-guided app called SuperBetter, or a mood-monitoring control. Remission of joint anxiety and depression at three and six months was higher in both active arms, with adjusted risk differences of roughly 19 to 23 percentage points for the group programme and 13 to 22 for the app. Low-cost remote delivery of that sort could suit campuses that cannot hire counsellors fast enough, which is the direction Wao et al. (2025) lean towards when they call for targeted, low-cost interventions. The trial, though, was built around anxiety and depression outcomes. Whether it would help a student recovering from rape, or from years of emotional abuse, remains unknown, and the trial cannot tell us.

Resilience training gets mentioned a lot, so the recent Western Cape results deserve a look. Padmanabhanunni and Pretorius (2025) found that higher smartphone addiction went with higher PTSD symptoms across all clusters, and that resilience moderated that link for avoidance and negative changes in thinking and mood but not for re-experiencing or hyperarousal. The authors read this as suggesting that the physiologically driven symptoms are less open to cognitive or emotional coping resources. If that holds up elsewhere, workshops that build resilience would help with some symptoms and leave the loudest ones alone. The violence studies point the same way. Mahlangu et al. (2025) argue that interventions should address both victimisation and perpetration, while Machisa et al. (2022) call for campus-based mental health promotion, psychosocial services and treatment together.

So what would I actually do? With exposure this widespread, screening students on whether they have ever been through something terrible seems pointless, because nearly everyone will say yes. Screening on current symptoms and then asking about the kind of exposure looks more sensible, and it matches the national finding that anxiety and PTSD are common enough to plan for (Bantjes et al., 2023). Services would then need clear pathways for the groups already identified, among them women facing partner violence, queer students, international students facing xenophobia, and students with a history of emotional abuse who might otherwise be treated for depression alone (Adedolapo et al., 2025; Myers et al., 2021). That is a judgement. None of the studies discussed here tested it.

Conclusion

Trauma exposure among South African university students is so common that the raw count has stopped being informative. What the better studies do is break it apart. Childhood emotional abuse stands out in one sample of first-years (Myers et al., 2021), partner violence and rape sit inside chains of earlier harm and deprivation in another (Machisa et al., 2022), and groups such as African international students and queer women face harms that standard lists barely register (Adedolapo et al., 2025; Mahlangu et al., 2025).

The evidence base is the weak point. Samples come from a handful of campuses, response rates fell as low as 8.4% at Wits (Croock et al., 2023) and under 13% in the xenophobia study (Adedolapo et al., 2025), designs are mostly cross-sectional, and prevalence figures built on different instruments do not line up. Nobody has yet shown which interventions work for students whose main problem is trauma, as against anxiety or depression (Bantjes et al., 2024). Longitudinal work that follows students from first year and measures cumulative exposure, deprivation and service use together would settle more than another prevalence count. Until then universities would do better to assume that nearly every student carries something, and to put their effort into finding out who is not coping, and why.

Reference list

Adedolapo, A., Feyasa, M. B., Salie, M., & Seedat, S. (2025). Mental health in relation to xenophobic experiences among African university students in South Africa. Journal of Racial and Ethnic Health Disparities. Advance online publication. https://doi.org/10.1007/s40615-025-02682-5

Bantjes, J., Hunt, X., Cuijpers, P., Kazdin, A. E., Kennedy, C. J., Luedtke, A., Malenica, I., Petukhova, M., Sampson, N., Zainal, N. H., Davids, C., Dunn-Coetzee, M., Gerber, R., Stein, D. J., & Kessler, R. C. (2024). Comparative effectiveness of remote digital gamified and group CBT skills training interventions for anxiety and depression among college students: Results of a three-arm randomised controlled trial. Behaviour Research and Therapy, 178, Article 104554. https://doi.org/10.1016/j.brat.2024.104554

Bantjes, J., Kessler, M., Lochner, C., Breet, E., Bawa, A., Roos, J., Davids, C., Muturiki, M., Kessler, R. C., & Stein, D. J. (2023). The mental health of university students in South Africa: Results of the national student survey. Journal of Affective Disorders, 321, 217-226. https://doi.org/10.1016/j.jad.2022.10.044

Croock, J., Mpinganjira, M. G., Gathoo, K., Bulmer, R., Lautenberg, S., Dlamini, Q., Londani, P., Solontsi, A., Stevens, C., & Francis, J. M. (2023). Probable depression and its correlates among undergraduate students in Johannesburg, South Africa. Frontiers in Psychiatry, 14, Article 1018197. https://doi.org/10.3389/fpsyt.2023.1018197

Machisa, M. T., Chirwa, E., Mahlangu, P., Nunze, N., Sikweyiya, Y., Dartnall, E., Pillay, M., & Jewkes, R. (2022). Suicidal thoughts, depression, post-traumatic stress, and harmful alcohol use associated with intimate partner violence and rape exposures among female students in South Africa. International Journal of Environmental Research and Public Health, 19(13), Article 7913. https://doi.org/10.3390/ijerph19137913

Mahlangu, P., Brooke-Sumner, C., Sikweyiya, Y., Dunkle, K., Mabena, N., Jewkes, R., Duma, S., Khuzwayo, N., Dartnall, E., Pillay, M., & Machisa, M. T. (2025). Intimate partner violence among lesbian, bisexual, and queer women students on campuses in South Africa: A qualitative study exploring context, drivers, and impacts. BMC Global and Public Health, 3(1), Article 33. https://doi.org/10.1186/s44263-025-00149-7

Meintjes, R. (2024). The mental health state of extended programme students at a South African university. South African Journal of Psychology, 54(1), 76-89. https://doi.org/10.1177/00812463241229163

Myers, B., Bantjes, J., Lochner, C., Mortier, P., Kessler, R. C., & Stein, D. J. (2021). Maltreatment during childhood and risk for common mental disorders among first year university students in South Africa. Social Psychiatry and Psychiatric Epidemiology, 56(7), 1175-1187. https://doi.org/10.1007/s00127-020-01992-9

Padmanabhanunni, A., & Pretorius, T. B. (2025). The impact of smartphone addiction on PTSD symptoms among South African university students: Resilience as a protective factor. Healthcare, 13(23), Article 3087. https://doi.org/10.3390/healthcare13233087

Wao, H., Wao, M. A., Omollo, S. B., & Mutua, C. K. (2025). Availability, accessibility, and utilization of mental health services or support among university students in Africa: A mixed methods systematic review with meta-analysis and meta-synthesis. BMC Psychiatry, 25(1), Article 1082. https://doi.org/10.1186/s12888-025-07529-1