A study of the mental health status of Chinese medical students at risk of COVID-19 infection

Introduction

Since the outbreak of the novel coronavirus disease (COVID-19) late in 2019, global socioeconomic stability and mental health of the public have faced unprecedented challenges. According to the World Health Organization (WHO), the number of new cases of major depressive disorder and anxiety disorders worldwide in 2020 reached 53 million and 76 million, respectively, representing a 25% increase compared with pre-pandemic levels [1]. As the virus evolved and herd immunity changed, many countries and regions gradually adjusted their pandemic control strategies. On December 7, 2022, the Chinese government announced 10 new measures to optimize its prevention and control policies. Following these adjustments, infection rates increased sharply in the short term [2] significantly impacting the mental health of the population. Research shows that during this period, the prevalence of anxiety and depressive symptoms among the Chinese population reached 30.1% and 55.7%, respectively [3].

Medical students, as future healthcare professionals, represent a critical population whose mental health requires close attention during the pandemic. Research shows that medical students often face heavy academic workloads, intense competition, and career-related anxiety, making them particularly vulnerable to mental health problems [4]. A study by Saraswati et al. in India [5] found that during the pandemic, medical students experienced elevated levels of anxiety and stress due to disruptions in education process and increased risk of infection. Chinese medical students experienced similar problems [6]. The increased risk of infection may have exacerbated preexisting mental health problems.

However, research on medical student mental health after the lifting of pandemic restrictions remains limited. Some scholars suggest that easing restrictions may help reduce negative emotions caused by prolonged isolation [3]. Yet medical students, due to the demands of their studies and future work environments, are at higher risk of infection and exposure to viruses. As a result, their mental health may evolve differently than that of the general population. Therefore, this study aims to assess the mental health of Chinese medical students during the early stages of policy relaxation, which will serve as a basis for further research.

Material and Methods

Study Design

This cross-sectional study was conducted on a random sample of medical students at Huzhou university, Huzhou, Zhejiang Province, China. Data collection was performed from December 2022 through January 2023, in accordance with China’s pandemic policy adjustments. The study protocol was approved by the Ethics Committee of Huzhou University (approval number 202302-02), and all study participants were volunteers.

Sampling procedure and participant selection

This study used simple random sampling in accordance with epidemiological survey protocol. Participants were selected from 1,676 students majoring in Nursing, Oral Medicine, and Clinical Medicine at the College of Medicine during the 2022–2023 academic year. Inclusion criteria were current enrollment and voluntary participation with electronic informed consent. A diagnosed mental disorder was an exclusion criterion. Based on the college’s electronic medical records, 15 students who met the exclusion criterion were excluded, leaving 1,661 eligible participants.

The administration provided an anonymized list containing only serial numbers, gender, and education level information. Using the RAND() function in Microsoft Excel 2019, the research team randomly selected 1,112 serial numbers (67.0% of the sample) and returned them to the administration, which then distributed survey invitations through the university’s official communication platform, WeChat. The research team did not have access to any of the students’ personal data or contact information.

An anonymous online survey was administered using Wenjuanxing (https://www.wjx.cn), and participants were required to complete all items before submitting. To increase response rates, the administration sent two reminder messages, one week apart, to non-respondents. A total of 896 students completed the survey (a response rate of 80.6%). After quality control for completion time and the availability of response templates, 872 valid questionnaires (97.3% validity rate) were retained. Although two reminder emails were sent to increase participation, no additional follow-up was conducted to avoid coercion. The total number of non-responders (n=216) was recorded but individual demographic data were anonymized. To assess representativeness, aggregated data (gender and education level) were compared between invited and responding students (see Supplementary Table S1). The results showed no significant differences, thereby indicating minimum non-response bias.

Questionnaire

The questionnaire consisted of two main sections.

Based on established methodologies from similar studies [7,8], the first section collected sociodemographic data relevant to mental health during and after the pandemic. These included gender, age, education level, place of residence, occupation, self-assessed physical health, daily time spent on pandemic-related news, history of close contact with COVID-19 patients, and personal or family history of COVID-19 infection.

The second section assessed mental health outcomes using validated instruments measuring resilience, anxiety, depression, post-traumatic stress disorder (PTSD), and coping strategies. The following scales were used:

(1) Resilience was assessed using the 10-item version of the Connor-Davidson Resilience Scale (CD-RISC-10) [9]. This instrument includes 10 items; each item is scored on a five-point scale ranging from 0 (not at all) to 4 (almost always).

(2) Anxiety symptoms were assessed using the General Anxiety Disorder-7 (GAD-7) [10], with scores ranging from 0 to 21, where higher scores indicate greater severity.

(3) Depression symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9) [11] with scores ranging from 0 to 27.

(4) PTSD was assessed using the Impact of Event Scale-Revised (IES-R) [12], a widely used 22-item self-report instrument assessing three clusters: intrusion (obsessive thoughts), hyperarousal, and avoidance. Items are rated on a 5-point Likert scale ranging from 0 (not at all) to 4 (extremely).

(5) Coping skills were assessed using the 20-item Simplified Coping Style Questionnaire (SCSQ) [13]. The SCSQ includes two subscales: positive coping strategies (12 items) and negative coping strategies (8 items) rated on a 4-point Likert scale (0=not at all, 3=usually).

The results were interpreted as follows: (1) GAD-7: ≤4 (normal), 5-9 (mild), ≥10 (moderate to severe) anxiety [14]; (2) PHQ-9: ≤4 (normal), 5-9 (mild), ≥10 (moderate to severe) depression [15]; (3) IES-R: ≤23 (normal), 24-32 (mild), ≥33 (moderate to severe) PTSD [16].

The two-section structure of the questionnaire was strictly adhered to for all participants: demographic data were administered in a fixed order and validated scales were administered sequentially. No variations or conditional branching were implemented.

Statistical data processing

Categorical data were described using frequencies (%), and univariate analysis was conducted using the chi-squared test and Fisher’s exact test. Based on the univariate analysis results, statistically significant variables were included in multivariate logistic regression models to identify sociodemographic and health-related factors associated with different levels of anxiety, depression, and PTSD symptoms (normal, mild, and above moderate). One-way analysis of variance (ANOVA) was employed to compare differences in resilience and positive/negative coping strategies among individuals with different levels of anxiety, depression, and PTSD symptoms. Tukey’s post hoc test was performed for statistically significant results. Statistical data processing used R 4.4.1 software, while graphs were created using GraphPad Prism 9.0. A p-value of <0.05 was considered statistically significant.

Sociodemographic characteristics of the study participants

The study included 872 medical students with a mean age of 19.26 years (SD=1.47). Of these, 590 (67.7%) were women and 282 (32.3%) were men. The participants were enrolled in three academic majors: Clinical Medicine (419 students, 48.0%), Nursing (321 students, 36.8%), and Oral Medicine (132 students, 15.1%). Most participants were first-year medical students (63.0%) and reported good health (95.5%). Detailed characteristics are presented in Table 1.

Table 1. Sociodemographic parameters of the study participants

Results

Anxiety levels among medical students

Univariate analysis revealed that freshmen (22.18% vs. 28.11%/32.11%), those without a history of COVID-19 (20.00% vs. 28.12%), and those in good health (24.01% vs. 46.15%) had lower rates of anxiety symptoms (shown in Table S2, Figure S1A). Multivariate logistic regression analysis (Table 2) showed that a history of infection (OR=1.532, 95% CI [1.087-2.160], p=0.015) and poor health status (OR=2.168, 95% CI [1.075-4.374], p=0.031) were statistically significantly associated with higher rates of mild anxiety symptoms. In contrast, higher education level was significantly linked with higher levels of above-moderate anxiety (OR=1.700, 95% CI [1.021-2.831], p=0.041), while poor health increased the risk of above-moderate anxiety as well (OR=5.871, 95% CI [1.804-19.106], p=0.003).

Table 2. Results of the multivariate logistic regression model for anxiety levels

Depression levels among medical students

Univariate analysis revealed that medical students exhibited lower rates of depressive symptoms if they were freshmen (31.45% vs. 41.62%/44.53%), had no close contacts with COVID-19 patients (28.31% vs. 38.13%), had no history of COVID-19 infection (29.25% vs. 39.66%), and exhibited good physical health (34.21% vs. 66.66%) (Table S3 and Figure S1B). The multivariate model (Table 3) showed that higher education level was significantly associated with mild depression (OR=1.282, 95% CI [1.053-1.559], p=0.013) and above-moderate depression (OR=1.438, 95% CI [1.016–2.035], p=0.040). History of infection was associated with mild depression (OR=1.438, 95% CI [1.005-2.058], p=0.047). Poor health was independently associated with both mild depression (OR=2.730, 95% CI [1.295-5.756], p=0.008) and above-moderate depression (OR=7.223, 95% CI [2.895-18.021], p<0.001). In contrast, close contacts were not statistically significantly associated with either mild or above-moderate depressive symptoms (all p>0.05).

Table 3. Results of the multivariate logistic regression model for depression levels

PTSD prevalence among medical students

Medical students residing in urban areas (19.8% vs. 24.5%), as well as students with good physical health (21.1% vs. 43.6%), demonstrated a lower prevalence of PTSD symptoms (Table S4 and Figure S1C). Multivariate logistic regression analysis (Table 4) revealed that poor health was significantly associated with higher levels of PTSD symptoms. Specifically, it was linked with mild PTSD (OR=2.484, 95% CI [1.070-5.769], p=0.034) and moderate PTSD (OR=3.253, 95% CI [1.448-7.305], p=0.004).

Table 4. results of the multivariate logistic regression model for PTSD levels

Differences in resilience and positive coping strategies across groups with varying levels of anxiety, depression, and PTSD symptom severity

levels of anxiety, depression, and PTSD symptoms (ANOVA results are presented in Tables S5-S7; Tukey’s post hoc test results are presented in Tables S8-S10). The analysis revealed that individuals with normal mental health had significantly higher resilience scores than those in the mild and above-moderate symptom groups across all three mental health indicators (p<0.001). Additionally, the mild symptom group showed significantly higher resilience than the above-moderate symptom group<0.001). Regarding positive coping strategies scores, the normal mental health individuals had significantly higher scores than the above-moderate symptom group (p<0.001). However, no statistically significant differences were observed between the normal group and the group with mild symptoms on measures of anxiety and PTSD.

Figure 1 illustrates the differences in resilience (Figure 1A) and positive coping strategies (Figure 1B) across groups with varying.

Figure 1. Distribution of CD-RISC-10 and SCSQ (positive coping strategies) scores according to the level of anxiety, depression, and PTSD. (A) Distribution of CD-RISC-10; (B) Distribution of SCSQ (positive coping strategies).

Examining the relationships between mental health factors

To test the hypothesized relationships between mental health constructs, we used structural equation modeling (SEM) (see Supplementary methods section for hypothesis development). The model structure with standardized path coefficients is presented in Figure 2. The model fit the data satisfactorily, as evidenced by the following indices: CFI=0.955, TLI=0.946, SRMR=0.033, RMSEA=0.073. The results confirmed all hypothesized paths: positive coping was a significant positive predictor of resilience (β=0.277, p<0.001) and a significant negative predictor of mood disorders (β=-0.201, p<0.001). Resilience, in turn, was inversely associated with mood disorders (β=-0.523, p<0.001), which were strongly and directly associated with PTSD symptom severity (β=0.790, p<0.001). Descriptive statistics and bivariate correlations for all variables are presented in Table S11.

Figure 2. Sequential effects of coping strategies, stress resilience, mood disorders, and posttraumatic stress disorder. ***P<0.001.

Discussion

The impact of pandemic-related policy changes on anxiety, depression, and PTSD symptoms in medical students

This study assessed the mental health of medical students after China’s pandemic policy adjustments. We found lower rates of anxiety (25.0%) and depression (35.7%) compared with both pre-adjustment medical students (29.6%, 52.5%) [17] and the general population after the adjustment (30.1%, 55.7%) [3,6]. We hypothesize that their medical education served as a protective factor. Their expert knowledge potentially facilitated a more scientific interpretation of pandemic information protecting them from widespread psychological stress and promoting better understanding and confidence in national policies, thereby reducing the impact on their mental health.

Our results indicate that students with a high education level demonstrated higher levels of anxiety and depressive symptoms, which is consistent with Shao’s findings [18]. These students are more likely to undergo clinical rotations, where they are often exposed to high-risk healthcare environments. Challenges such as long work hours, sleep deprivation, and limited clinical experience may contribute to elevated stress levels in medical students [19]. In addition to clinical rotations, they also face important career milestones, including taking important national licensing exams, competitive residency placement processes, and beginning a clinical career. Recent studies conducted during the COVID-19 pandemic revealed elevated levels of anxiety in this group due to these career-related challenges [20]. The post-pandemic period, characterized by increased risk of infection, has likely introduced additional uncertainty and anxiety into these already stressful career-defining stages [21]. This is consistent with findings that career-related uncertainty and the pressure of important examinations are significant factors contributing to psychological stress among medical students [22]. However, potential non-response bias should be considered, as students who did not complete the survey may differ in their levels of psychological distress or pandemic experience, which may slightly influence the observed estimates. However, comparison of aggregated demographic data (gender and educational level) between invited and responding students revealed no significant differences, suggesting that the magnitude of such bias was likely negligible. Furthermore, medical students who did not contract COVID-19 reported lower levels of anxiety and depressive symptoms. Numerous studies deminstrated that infected individuals are at a significantly higher risk of developing mental disorders such as anxiety and depression vs. their uninfected peers [23]. Uninfected individuals may have been vaccinated earlier, reducing both the risk of infection and the associated psychological burden [24]. Furthermore, long-term physical consequences of infection, such as fatigue and sleep disorders, may further exacerbate the risk of depression and anxiety [25].

We also established that chronic illnesses were associated with poorer mental health outcomes. Previous research findings suggested that persistent physical discomfort, pain, and functional limitations associated with chronic illnesses may increase vulnerability to anxiety and depression [26, 27]. Conversely, anxiety and depressive symptoms may exacerbate the progression and severity of chronic conditions. Furthermore, prolonged stress associated with managing a chronic illness may lead to dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis, leading to abnormal secretion of cortisol [28]. This neuroendocrine dysregulation resembles the HPA axis dysfunction observed in patients with PTSD [29], suggesting that HPA axis alterations may serve as a biological pathway linking chronic illness to PTSD symptoms.

The protective role of resilience and positive coping strategies

Our results suggest that medical students with higher levels of anxiety, depression, or PTSD symptoms tend to have lower scores on resilience and positive coping strategies. This implies that resilience and positive coping strategies may help protect medical students from psychological stress and support their mental well-being. Resilience is defined as the ability to adapt and recover from adversity or trauma by drawing on internal and external resources [30]. Highly resilient individuals are better able to cope with negative emotions and maintain a positive outlook during challenging times. This allows them to view challenges as opportunities for personal growth while remaining focused and motivated. In contrast, positive coping strategies aim to improve well-being and strengthen social support networks [31]. For medical students, this may include building strong relationships with peers to collaboratively address academic and clinical challenges, thereby creating opportunities for mutual support and encouragement. These social and emotional resources may improve overall well-being and the ability to cope with stress.

Our analysis also revealed that better mental health outcomes were associated with higher resilience and more extensive use of positive coping strategies. These traits may act as protective factors supporting psychological health in challenging environments. This is consistent with previous research [32,33] and highlights the importance of providing mental health support to medical students. In addition to existing counseling and education programs [34], future efforts should include training in resilience skills and positive coping strategies in the medical school curriculum. For senior students and students with chronic illnesses, individualized guidance, such as career planning and self-management support [35,36], may help them prepare for the future and improve overall well-being.

This study has several limitations. First, the sample was drawn from students at a single medical school, and the overrepresentation of female participants may limit the generalizability of the results. Second, a cross-sectional study does not allow for causal inferences. Third, although the administration sent reminder messages to increase participation, non-respondents may have differed from those who responded in their psychological state or level of engagement, potentially biasing the results. Finally, the data were based solely on self-reports without accounting for clinical diagnoses, thereby potentially introducing measurement bias.

Conclusion

The study revealed that after China adjusted its pandemic policies, many medical students reported psychological symptoms, including anxiety, depression, and PTSD. Those most affected were senior students, those in close contact with COVID-19 patients, those previously infected, and students with chronic health conditions. Future support efforts should focus on these high-risk groups by providing targeted psychological services, resilience training, and mental health promotion activities tailored to their needs.

Conflict of interest

The authors declare no conflicts of interest.

Disclosure of generative AI use

Generative artificial intelligence was not used in the preparation of this manuscript.

Acknowledgments

The author expresses deep gratitude to Mr. Gu and Ms. Fang for their solid academic support throughout this study. Special thanks are extended to Dr. Shen for his scientific vigilance in providing critical methodological clarifications and intellectual rigor during the editing process.

Ethical approval

This study protocol was reviewed and approved by the Biomedical Research Ethics Committee of Huzhou University (202302-02). In accordance with the ethical principles outlined in the Declaration of Helsinki, all participants provided digital informed consent, ensuring their voluntary participation and understanding of the goal and procedures of the study. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Table S4. Univariate analysis of PTSD symptom level among medical students

Table S5. One-way ANOVA results of CD-RISC-10 and SCSQ scores at different anxiety levels

Table S6. One-way ANOVA results of CD-RISC-10 and SCSQ scores at different depression levels

Table S7. One-way ANOVA results of CD-RISC-10 and SCSQ scores at different PTSD severity levels

Table S8. Tukey’s post-hoc test to compare scores between different groups with different anxiety levels

Table S9. Tukey’s post-hoc test to compare scores between different groups with different depression levels

Table S10. Tukey’s post-hoc test to compare scores between different groups with different PTSD levels

Table S11. Descriptive statistics and correlations between studied variables

Figure S1. Percentage stacked bar chart of anxiety, depression, and PTSD levels by demographic characteristics among medical students. (A) Anxiety levels; (B) depression levels; (C) PTSD levels. 

Figure S2. Proposed structural model of the relationship between coping tendencies, resilience, mood disorders, and PTSD.