Prevalence of and factors associated with post-traumatic stress disorder among French university students 1 month after the COVID-19 lockdown

The COVID-19 pandemic and resulting lockdowns globally impacted billions of people, raising concerns about mental health consequences. Prior research indicated negative psychological effects from quarantine, including post-traumatic stress symptoms. Post-traumatic stress disorder (PTSD), a trauma- and stressor-related disorder, is characterized by re-experiencing, avoidance, negative cognitions/mood, and hyperarousal. While some studies linked PTSD to COVID-19-related events, its prevalence in the general population due to the pandemic and lockdown remained debated. Earlier studies on university students, a high-risk population for mental health issues, showed high rates of distress symptoms during the pandemic. However, these studies lacked consistency in methodology and assessment tools (PCL-C vs. PCL-5). This study aimed to determine the prevalence of probable PTSD among French university students one month post-lockdown, identify associated factors, and analyze subjective perceptions of traumatic events related to the pandemic.

Existing literature highlighted the significant psychological distress experienced by university students during the initial phases of the COVID-19 pandemic. Studies reported high percentages of moderate to severe distress symptoms in student populations across different countries. However, there was a notable lack of consistency in the methodologies employed and the assessment tools utilized. The heterogeneity in prevalence estimates of PTSD among quarantined university students (ranging from 2.79% to 30.8%) underscored the need for a standardized approach to assess the mental health impact of the pandemic and lockdown. This study addresses this gap by using the DSM-5 aligned PCL-5 to measure PTSD symptoms, improving comparability with future research.

This study utilized data from the COSAMe survey, a repeated cross-sectional study of French university students. Data for this analysis came from the second measurement time (T2), conducted one month after the lockdown ended (June 15-July 15, 2020). A total of 22,883 students completed the online, anonymous questionnaire at T2. Probable PTSD was assessed using the French version of the PCL-5, a 20-item scale measuring PTSD symptom severity over the past month. A score above 32 indicated probable PTSD. An alternative assessment used DSM-5 criteria for PTSD diagnosis. Several variables were collected, including socio-demographic characteristics (age, gender), clinical information (psychiatric history, exposure to non-COVID-19 related trauma), social support (living alone, social ties quality), socioeconomic factors (income loss, housing quality), information quality regarding the pandemic, and COVID-19 exposure level (constructed scale based on seven items). For students who participated in both T1 (during lockdown) and T2, distress levels at T1 (using the IES-R) were analyzed to determine psychological response trajectories (resilience, persistence, recovery, delayed onset). Multivariate logistic regression analyzed the association between factors and probable PTSD.

The study revealed a 19.5% prevalence of probable PTSD among the 22,883 participating students, as measured by a PCL-5 score above 32. This aligns closely (91.7% agreement, Kappa=0.738) with the prevalence determined using DSM-5 criteria (19.4%). Multivariate logistic regression identified significant associations between probable PTSD and several factors: Female and non-binary genders showed higher odds of PTSD compared to males. Older age was associated with a slightly lower likelihood of PTSD. A history of psychiatric follow-up or exposure to another traumatic event increased the odds of PTSD substantially. Living alone during quarantine, poor quality of social ties, low sense of social integration, income loss, poor housing quality, low-quality information about the pandemic, and higher COVID-19 exposure scores were all significantly associated with an elevated risk of probable PTSD. Regarding subjective perceptions of traumatic events, over three-quarters of students with probable PTSD considered quarantine potentially traumatic, second only to the death of a relative due to COVID-19. Analysis of response trajectories among students who completed both T1 and T2 showed that 73.2% exhibited resilience, 10.3% recovery, 9.9% persistence, and 6.5% delayed onset of PTSD. The prevalence of probable PTSD at T2 varied significantly according to the level of distress experienced at T1 (ranging from 4.3% for no distress to 49.0% for severe distress).

The high prevalence of probable PTSD (19.5%) among French university students one month post-lockdown, significantly exceeding pre-pandemic rates (0.7%), highlights the pandemic's negative impact on mental health. The identified risk factors corroborate existing literature on PTSD, emphasizing the importance of social support, economic stability, and access to reliable information during crises. The study’s findings underscore the vulnerability of young adults, particularly students, to mental health challenges, even after the immediate crisis has passed. The various response trajectories observed highlight the heterogeneity of individual reactions to traumatic events, indicating the need for targeted interventions tailored to different needs and profiles. The study’s large sample size and use of validated instruments (PCL-5, IES-R) strengthen the reliability of its findings.

This study demonstrates a significant increase in probable PTSD prevalence among French university students following the COVID-19 lockdown. Several socio-demographic, clinical, social, economic, and pandemic-related factors were associated with increased PTSD risk. The findings highlight the need for mental health support systems to address the lasting mental health effects of the pandemic on vulnerable populations like university students. Future research should investigate longitudinal trajectories of PTSD and the efficacy of specific interventions to mitigate these effects.

The cross-sectional design limits causal inference. The reliance on self-reported data may introduce recall bias. The study is specific to French university students, potentially limiting the generalizability to other populations or contexts. The use of probable PTSD based on PCL-5 scores rather than clinical diagnoses should be considered. Future studies could address these limitations through longitudinal designs, incorporating objective measures, and broader sampling strategies.