The impact of the 2019 coronavirus disease (COVID-19), which began in China and spread around the world starting in December 2019 [1
], has profoundly altered the economic activities and daily lives of people everywhere [2
]. Increased stress and loneliness due to COVID-19 and the measures taken to slow the spread of COVID-19 have led to the deterioration of mental health in the general public and particularly in certain vulnerable subpopulations, such as students [3
COVID-19 has resulted in lost learning opportunities and economic stress for college students [4
]. About 30% of college students were reported to have mental health problems such as depression and anxiety due to the spread of COVID-19 [6
]. In Japan, the government asked primary and secondary schools across the nation to close temporarily on 2 March 2020. Nine days later, the World Health Organization declared COVID-19 a pandemic [8
]. On 7 April 2020, the Japanese government issued an emergency declaration and designated 13 prefectures, including Tokyo and Osaka, as special caution areas [9
]. That declaration required people to reduce interpersonal contact by 80% and to limit outings; it changed the lifestyles of many people [10
], which meant that many students could no longer go to school. In addition, especially for students living alone, going to school and going out, such as shopping, were restricted, which meant spending time alone at home.
For college students, participation in the college community provides a foundation for spiritual growth. Online lessons alone cannot compensate for the loss of interactions with that community. Students whose engagement with the college community is limited to online lessons at home alone are vulnerable to feelings of loneliness and anxiety, which adversely affect mental health. Therefore, it is not enough to deal with the loss of learning opportunities; it is also necessary to understand the mental vulnerabilities caused by the loss of community. In addition, it is necessary to consider specific risk factors and coping strategies that affect the deterioration of mental health. In particular, healthcare college students are at risk of worsening mental health compared to other college students, as are healthcare workers exposed to infection and social risks during a pandemic and a special work environment different from the general public [11
Coping strategies, defined as a person’s constantly changing cognitive and behavioral efforts to manage specific external and/or internal demands, are indispensable in stressful circumstances [14
]. Previous studies reported on the effects of COVID-19 restrictions on students’ mental health [2
] but did not provide any guidance on coping strategies to deal with them. Therefore, it is useful to investigate how students spend their lives and cope with stress in a lifestyle that has been transformed by COVID-19 to support their mental health in the COVID-19 pandemic.
We conducted a cross-sectional observational investigation of healthcare college students to identify specific risk factors and coping strategies that affect the mental health of COVID-19-affected college students.
2. Materials and Methods
We conducted a cross-sectional survey of third-year and fourth-year students at Kitasato University School of Allied Health Sciences using a web-based questionnaire (Google Form) from 28 April 2020 to 4 May 2020. Considering the mental burden on participants living in a pandemic to answer questions, they were not prompted for answers and were not linked to individuals to prevent duplicate answers.
Survey items included demographic information such as age, gender, number of people living together, region of residence, year of university, communication with friends and family, and financial status. Regions of residence were divided into the alert region (13 prefectures of the special alert area) and the non-alert region (the remaining 32 prefectures in Japan). Communication with family and friends was rated on a three-point Likert scale: “same as usual”, “less than usual”, and “more than usual”. The respondents’ financial situation was also rated on a three-point Likert scale: “no change”, “worse than usual”, and “better than usual”.
We assessed mental health status using the Japanese version of The General Health Questionnaire-12 (GHQ-12), which includes 12 questions regarding mental health status, with four options for each question. The GHQ-12 typically has two scoring methods, bimodal (0-0-1-1) and Likert scale (0-1-2-3) [16
]. We used bimodal scoring in the present study. In accordance with previous studies, we considered a GHQ-12 score ≥ 4 as indicative of mental health problems [17
We investigated health status and anxiety over COVID-19 using a visual analog scale, with scores ranging from 1 (“not at all healthy” or “not at all anxious”) to 10 (“very healthy” or “very anxious”). To capture satisfaction with daily life, we developed an original 10-point Likert scale based on the Canadian Occupational Performance Measure [19
], which included satisfaction with leisure, satisfaction with job, satisfaction with daily-life activities, and satisfaction with new activities started since COVID-19 restrictions began [20
]. A score of 1 indicated “not satisfied at all” or “no new activities”. A score of 10 indicated “very satisfied”. Participants were also asked to give an example of a new activity started since COVID-19 restrictions began.
Different types of behavioral motivation are related to stress and wellbeing [22
]. We asked participants about their motivation for self-restraint regarding COVID-19. The participants indicated on a visual analog scale the degree to which their motivation was due to intrinsic factors, such as not wanting to spread infection, or to extrinsic factors, such as requests from the government. A score of 1 indicated “self-determination based on extrinsic motivation”. A score of 10 indicated “self-determination based on intrinsic motivation”.
Since the spread of COVID-19 in Japan and the issuance of a state of emergency were completely unexpected to us, and the period was not clarified, these survey items have not been surveyed in advance.
To evaluate the respondents’ coping strategies, we categorized behavioral efforts to manage stress into classes used in previous research [23
], which included confrontive coping, distancing, self-control, seeking social support, accepting responsibility, escape/avoidance, planful problem solving, and positive reappraisal. Confrontive coping includes aggressive efforts to alter the situation and involves a degree of hostility and risk taking. Distancing includes efforts to detach oneself from the adverse situation. Self-control includes efforts to regulate one’s own feelings and actions. Seeking social support includes efforts to seek informational support, tangible support, and emotional support. Accepting responsibility means acknowledging one’s own role in the problem with a concomitant theme of trying to put things right. Escape-avoidance includes wishful thinking and behavioral efforts to escape or avoid the adverse situation. Planful problem solving includes deliberate problem-focused efforts to alter the situation coupled with an analytic approach to solving the problem. Positive reappraisal includes efforts to create positive meaning by focusing on personal growth. We determined which coping strategies each respondent used and calculated the overall use of each category among all the students.
We divided the respondents according to mental health status into a poor mental health group (GHQ-12 ≧ 4) and a normal group (GHQ-12 < 4). We performed Mann–Whitney tests using SPSS software (version 25, IBM Corp.) to compare subgroups based on mental health status, gender, year of university, and number of household members. We calculated the adjusted odds ratio (OR) with a 95% confidence interval (CI) for risk of poor mental (GHQ-12 ≧ 4) health associated with COVID-19 restriction. The multiple logistic regression analysis performed with adjustments for all potential factors as listed in Table 1
and Table 2
. We used p
< 0.05 as the threshold for significance in all statistical tests.
Of 828 eligible students, a total of 226 (27.3%) healthcare college students responded to the survey for 7 days from 24 April 2020 to 4 May 2020. Three respondents did not complete the survey and were excluded from the analysis.
3.1. Demographic Characteristics
The demographic data of the 223 (26.9%) respondents that completed the survey are shown in Table 1
. The participants included 175 females (78.5%). The age distribution was as follows: 20 years, 92 participants (41.3%); 21 years, 103 participants (46.2%); 22 years, 24 participants (10.8%); 23 years, 3 participants (1.3%); and 24 years, 1 participant (0.4%). A total of 121 participants (54.3%) were third grade students, and 102 participants (45.7%) were fourth grade students. In terms of living status, 81 (36.3%) lived alone, and 142 (63.7%) lived with two or more other people. A total of 197 participants (88.3%) lived in the alert region and 26 (11.7%) lived in the non-alert region.
Regarding financial situation, 112 participants (50.2%) answered “no change”, and 91 participants (40.8%) answered “worse than usual” and 20 participants (9.0%) answered “better than usual”. For amount of communication with family, 86 participants (38.6%) answered “same as usual”, and 11 participants (4.9%) answered “less than usual”, and 126 participants (56.5%) answered “more than usual”. Conversely, 22 participants (9.9%) answered “same as usual”, and 192 (86.1%) responded “less than usual”, and 9 participants (4.0%) answered “more than usual” for amount of communication with friends.
3.2. Severity of Mental Health Problems and Associated Factors
shows the medians, interquartile ranges (IQRs), and statistical comparisons of the GHQ-12, anxiety, health, self-determination, and satisfaction scores stratified by gender, year of university, and number of household residents. The Median score of GHQ-12 was 5 (interquartile range (IQR): 3–8). The other Median scores were 8 (IQR: 7–10) for anxiety over COVID-19, 8 (IQR: 6–10) for health condition, 7 (IQR: 5–9) for self-determination score, 3 (IQR: 1–4) for satisfaction with leisure, 4 (IQR: 2–6) for satisfaction with study, 4 (IQR: 3–7) for satisfaction with daily life activities, and 6 (IQR: 2–8) for satisfaction with new activities started since social distancing began.
Compared with men, women had higher anxiety regarding COVID-19 and lower satisfaction with daily-life activities (p = 0.001 and p = 0.04, respectively). Students in their third year of university were slightly more satisfied with their leisure activities than students in their fourth year (p = 0.02), although satisfaction with leisure activities was low in both groups. Compared with students who lived with other individuals, students who lived alone reported lower satisfaction with their studies and new activities (p = 0.03 for both comparisons).
shows demographic characteristics and mental health measurements for normal and severe mental health problems groups. The severe mental health problems group (GHQ-12 score ≥ 4) including 158 participants (70.9%) exhibited these characteristics: 128 (81.0%) were females; 83 participants (52.5%) were third grade students, and 75 participants (47.5%) participants were fourth grade students; 60 (38.0%) lived alone, and 98 (62.0%) lived with two or more other people; 144 participants (91.1%) lived in an alert region, and 14 participants (8.9%) lived in a non-alert region; 73 (46.2%) answered “same as usual”, and 71 (44.9%) responded “worse than usual” for financial situation; 58 (36.7%) answered “no change”, and 9 (5.7%) responded “less than usual” for communication with family; and 7 (4.4%) answered “same as usual”, and 144 (91.1%) responded “less than usual” for communication with friends.
Compared with the students with normal mental health status, the students with poor mental health status had less anxiety about COVD-19 (p
= 0.03), higher self-reported health status (p
< 0.001), higher self-determination scores (p
= 0.04), and greater satisfaction with leisure (p
< 0.001), studies (p
= 0.001), daily-life activities (p
< 0.001), and new activities (p
< 0.001; Table 4
3.3. Resilience Factors for Mental Health Outcomes
After adjustment for covariates in the logistic regression analysis, less communication with friends (odds ratio (OR), 5.38; 95% confidence interval (CI), 1.54–18.83; p
= 0.008) was independently associated with increased risk of mental health problems (Table 5
). By contrast, good health status (OR, 0.68; 95% CI, 0.55–0.83; p
< 0.001), satisfaction with leisure (OR, 0.68; 95% CI, 0.58–0.79; p
< 0.001), and satisfaction with new activities (OR, 0.88; 95% CI, 0.78–0.99; p
= 0.04) were associated with reduced risk of mental health problems.
3.4. Coping Strategies: New Activities Initiated since Social Distancing Began
shows the types of coping strategies reported by the students. Thirty-five of 65 respondents (53.8%) with normal mental health and 98 of 158 respondents with poor mental health (62.0%) reported some kind of stress coping. Figure 1
contrasts the poor mental health problem group with the normal mental health group, and shows the contrast of the adoption rate of each coping style with a line. Escape/avoidance was the most common coping strategy reported by students in both groups. The next most common strategy in both groups was seeking social support, which was more common among students with poor mental health than among students with normal mental health.