COVID-19 and common mental disorders among university students in South Africa

The study investigates the impact of the COVID-19 pandemic on the mental health of first-year university students in South Africa. The context is a global rise in depression and anxiety reported in the general population and among students since the pandemic began, alongside a pre-existing concern about high rates of common mental disorders (CMDs) in university populations. Many prior studies lacked direct pre- and during-pandemic comparisons within the same populations and methods, especially in low- and middle-income countries (LMICs). Given links between student mental health and academic outcomes, understanding COVID-19’s effects is important for planning interventions that support educational retention and economic recovery. The research question asks whether the prevalence and onset patterns of major depressive episode (MDE), generalized anxiety disorder (GAD), and suicidal ideation among South African first-year students changed during COVID-19 compared to earlier periods (2015 and 2017).

Pre-COVID studies showed high rates of CMDs among university students worldwide, with the WHO WMH-ICS reporting a 12-month prevalence of about 35% for at least one CMD among first-year students, and substantial prevalence of suicidal ideation, plans, and attempts. CMDs in students are linked to developmental stressors and coincide with typical onset in late adolescence/early adulthood, producing significant role impairment in academic and social functioning. Global analyses during COVID-19 estimated marked increases in depression and anxiety cases. Numerous studies across countries (e.g., UK, China, Poland, Bangladesh) documented elevated psychological distress in students during the pandemic, although global suicide rates generally did not show increases in the early pandemic period. In South Africa, student mental health concerns predate COVID-19, with local factors including socio-economic adversity, crime and violence exposure, HIV/TB syndemics, food and housing insecurity, and campus disruptions such as the #FeesMustFall protests that affected learning and may have impacted mental health prior to the pandemic.

Design and aim: Cross-sectional analyses comparing first-year university students' mental health at three time points, before and during COVID-19, to evaluate changes in prevalence and age of onset for MDE, GAD, and suicidal ideation, and to assess whether demographic shifts explain observed changes. Settings and participants: All first-year students at Stellenbosch University (SU) in 2015 (N=5338), 2017 (N=5163), 2020 (N=5584), and at the University of Cape Town (UCT) in 2017 (N=3662) and 2020 (N=5408) were invited via institutional email to complete an anonymous online survey. A total of 2271 students completed surveys across years. Inclusion criteria: enrollment as first-year student at SU or UCT and informed consent. 2020 data were collected between 14 April and 10 July 2020. Measures: Sociodemographic variables included age, parents’ education (first-generation if neither parent completed tertiary education), serious physical impairment or chronic illness (disability), gender identity (female, male, gender non-conforming), population group (black [black African, coloured, Asian] vs white), and sexual orientation (heterosexual vs sexual minority). CMDs were assessed using adapted items from the Composite International Diagnostic Interview Screening Scales (CIDI-SC) for lifetime and 12-month MDE and GAD, which have good concordance with SCID-based diagnoses. Suicidal ideation (lifetime and 12-month) was assessed with a modified Columbia-Suicide Severity Rating Scale (C-SSRS) capturing passive and active ideation. Analysis: Data were weighted by population group and gender within each institution and year to adjust for sample-population differences and demographic changes over time. Analyses included: lifetime and 12-month prevalence estimates for MDE, GAD, suicidal ideation, and any of these outcomes; lifetime cumulative and conditional incidence before COVID-19 (2015, 2017) vs during COVID-19 (2020); 12-month persistence among prior lifetime cases; age-of-onset curves with log-rank tests; and regression models estimating adjusted risk ratios (corrected odds ratios) with 95% CIs to test whether changes were attributable to demographic shifts. Alpha=0.05. Ethical approval was obtained from Health Sciences Research Ethics Committees at SU and UCT; institutional permission for student contact was secured.

- No clear or consistent pattern of increased CMD prevalence was observed following the start of COVID-19 when compared to prior periods; lifetime CMD prevalence had been rising since 2015. - Lifetime prevalence at Stellenbosch University (SU): • MDE: 20.5% (2015) → 27.4% (2017, p=0.009) → 47.6% (2020, p<0.0001 vs 2017) • GAD: 20.2% (2015) → 23.3% (2017, ns) → 29.6% (2020, p=0.039 vs 2017) • Suicidal ideation: 42.9% (2015) → 49.6% (2017, p=0.035) → 60.0% (2020, p=0.003 vs 2017) • Any disorder: 51.6% (2015) → 57.8% (2017, p=0.054) → 70.9% (2020, p<0.0001 vs 2017) - Lifetime prevalence at University of Cape Town (UCT): between 2017 and 2020 no significant changes for MDE, GAD, suicidal ideation, or any disorder. - 12-month prevalence: • SU: 2015→2017 increases for MDE 16.8%→24.4% (p=0.002), suicidal ideation 24.1%→36.9% (p<0.0001), any disorder 36.6%→47.8% (p=0.0004); GAD ns. 2017→2020 only MDE increased 24.4%→40.3% (p<0.0001); GAD, suicidal ideation, and any disorder showed no significant change. • UCT: 2017→2020 suicidal ideation decreased 48.9%→36.0% (p=0.006); MDE 39.9%→45.3% (ns), GAD 29.7%→29.7% (ns), any disorder 63.7%→58.5% (ns). - Persistence among lifetime cases (12-month prevalence among those with prior lifetime disorder): • SU: 2015→2017 increased for suicidal ideation 56.2%→74.4% (p<0.0001) and any disorder 71.0%→82.7% (p=0.002). 2017→2020 decreased for suicidal ideation 74.4%→54.4% (p<0.0001); other outcomes showed no significant changes. • UCT: 2017→2020 decreased for suicidal ideation 76.8%→58.0% (p=0.001); other outcomes showed no significant changes. - Age of onset: Onset curves indicate earlier ages of onset in 2020 versus 2017 across disorders; log-rank tests showed significantly different failure times across years at both institutions (p<0.0001). - Regression controlling for demographic changes (risk ratios, 95% CI): • SU (lifetime as of 2 years prior): MDE increased 2015→2017 RR=1.4 (1.1–1.8) and 2017→2020 RR=1.8 (1.5–2.1), with no significant difference between the two intervals (p=0.327); GAD 2017→2020 RR=1.3 (1.0–1.7); suicidal ideation 2017→2020 RR=1.2 (1.1–1.4); any disorder 2017→2020 RR=1.3 (1.2–1.4). • UCT (lifetime as of 2 years prior): MDE increased 2017→2020 RR=1.4 (1.1–1.8); GAD, suicidal ideation, and any disorder showed no significant changes. • Additional analyses: at SU, 12-month prevalence among those without a lifetime diagnosis as of 2 years prior increased for MDE 2017→2020 RR=3.1 (1.6–6.0); among lifetime cases, suicidal ideation decreased 2017→2020 RR=0.6 (0.4–0.9). At UCT, among lifetime cases, suicidal ideation decreased 2017→2020 RR=0.4 (0.3–0.7). - 2020 12-month prevalence levels: any disorder SU 53.7%, UCT 58.5%; MDE SU 40.3%, UCT 45.3%; suicidal ideation SU 32.7%, UCT 36.0%; GAD SU 27.6%, UCT 29.7%. These exceed general population estimates.

The study addressed whether COVID-19 exacerbated common mental disorders among South African first-year university students beyond existing trends. Findings indicate no consistent increase in 12-month CMD prevalence after COVID-19’s onset, aside from an increase in MDE at SU and a decrease in suicidal ideation at UCT, and lifetime increases largely continued pre-existing trends. Earlier ages of onset in 2020 suggest that rising lifetime prevalence is partly driven by earlier development of symptoms. Regression analyses show that some observed increases are attributable to shifting demographics (e.g., increased enrollment of higher-risk groups such as female students). The South African context—including prolonged campus disruptions from the #FeesMustFall protests, high levels of violence and crime exposure, and the ongoing HIV/TB syndemic—may have produced high pre-pandemic distress, making COVID-19 one added stressor rather than a singular driver of increased psychopathology. Alternative interpretations include student resilience, limitations of diagnostic criteria in capturing subthreshold pandemic-related distress, and the possibility that the 2020 survey timing (April–July, early pandemic) was too soon to capture later effects.

Analyses of first-year students at two South African universities before and during the COVID-19 pandemic show no clear, consistent increases in MDE, GAD, or suicidal ideation attributable to COVID-19 beyond prior trends; where increases exist, many began before the pandemic. In a context of ongoing adversity and campus disruptions, COVID-19 appears to be one additional stressor rather than a uniquely amplifying factor for student mental health. The high prevalence of CMDs underscores the need for sustainable, scalable campus-based interventions to promote student well-being. Future research should track cohorts beyond the first pandemic year, include senior students, and examine institutions in rural and less resourced settings to assess differential impacts over time.

- Low response rate may bias prevalence estimates of trends, although the rate did not change over time, reducing the likelihood that this alone explains the lack of trend. - Significant changes in prevalence before the pandemic complicate attribution of effects to COVID-19. - Data reflect the first year of the pandemic; effects might have increased subsequently. - The sample includes only first-year students from two well-resourced Western Cape universities; patterns may differ among senior students and at rural or less well-resourced institutions.