Impact of COVID-19 on the mental health of Japanese university students (years II-IV)


 This study aimed to investigate the differences in mental health during COVID-19, specifically among second-, third-, and fourth-year Japanese university students (n = 2,157; n = 2,000; and n = 2,284; respectively). A one-way MANOVA was conducted to assess the association between year of enrollment (academic years 2020, 2021, and 2022) and each of the eight subscales of the counseling Center Assessment Psychological Symptoms-Japanese. For second-year students, depression and generalized anxiety mean scores were higher in 2021 than those in 2020 and 2022. Alcohol use mean scores got smaller each year. For third-year students, depression, generalized anxiety, social anxiety, hostility, and alcohol use were significantly higher in 2021 than those in 2020 and 2022. Among fourth-year students, means for all eight subscale categories were significantly higher in 2021 than those in 2020 and 2022. The findings found worsened mental health profiles during 2021, which recovered to approximate pre-pandemic levels in 2022. This study demonstrates that university students’ mental health has been negatively impacted by the COVID-19 pandemic, an effect which is more salient among fourth-year students. Further, it offers insights into mental health trends among Japanese university students and a possible foundation for learning about changes among university students worldwide.



Introduction
Coronavirus disease 2019  caused by acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was declared as a pandemic by the World Health Organization (WHO) on March 11, 2020 (Cucinotta and Vanelli, 2020).Nations worldwide began closing borders, locking down cities, restricting access to public venues, making masks mandatory, shifting education delivery to a remote platform, and directing residents to shelters (Hale et al., 2021).Hence, people were forced to adapt their lifestyles and routines quickly and drastically as they adjusted to the COVID-19 pandemic.
Anxiety, stress, and depression were exceedingly widespread during the COVID-19 pandemic, with strong correlations between them (Shah et al., 2021;WHO, 2022).Anxiety and depression were prevalent among young adults, including students, and quarantine for longer periods was associated with increased levels of stress, anxiety, and depression (Shah et al., 2021).Nationwide cross-sectional studies reported that the extended COVID-19-related lockdowns were associated with increased aggression in the U.S. (Killgore et al., 2021) and decreased emotional well-being in China (Qin et al., 2020).Currently, however, only a few studies have focused on the mental health of university students during the pandemic, which have been based in China (Chen et al., 2020), Lebanon (Hammoudi et al., 2021).
Japan did not implement lockdowns at the municipal level.Instead, a state of emergency at national and municipal levels was declared several times when the number of infected people spiked.This state of emergency promoted "The New Lifestyle," which comprises some basic rules: infection prevention, maintaining physical distance, wearing a mask, and washing hands.People avoided the "Three Cs" (crowded areas, closed spaces, and close-contact settings) and refrained from going out unnecessarily.The declaration imposed strict behavioral restrictions on students such as travel and dining with friends because almost all events were canceled or restricted in the number of people who could attend, and restaurants and pubs were closed or only for brief periods.
COVID-19 affected the mental health of Japanese university students.For example, less frequent conversations increased the risk of suicidal ideation among medical students (Yamazaki et al., 2022).The prevalence of anxiety and depressive symptoms were remarkably high regardless of perceived vulnerability (Sakai et al., 2022).Moreover, encouraging physical activity during the state of emergency may have had positive effects, especially on emotions like happiness (Nakahara-Gondoh et al., 2022).Although, most prior studies only collected a single set of data.
Our previous study at Gifu University investigated mental health using the counseling Center Assessment of Psychological Symptoms-Japanese (CCAPS-Japanese) tool among first-year Japanese university students.It concluded that the average score of mental health declined after the COVID-19 pandemic (2020) and returned to the pre-pandemic level (2019) over the next two years (2021).However, the number of students at high risk of mental health issues continued to rise from 2019 to 2021 (R. Horita et al., 2021Horita et al., , 2022)).Generally, these students experienced COVID-19 as they were entering university.
To advance previous research, this study examined the changes in mental health among upper-level (second-, third-, and fourth-year) Japanese university students before and during the COVID-19 pandemic.Most university students in Japan enter at the age of 18 or 19 years and graduate in four years with the same classmates, and the composition of each class continues from the first grade to fourth (final) grade.Therefore, it is reasonable to make grade-by-grade comparisons.These students faced restrictions during their university experience such as on-campus classes, club activities, and so on.In this study, we examined the impact of behavioral restrictions due to the pandemic.

Study design
This study was conducted over three years using online surveys with the same procedure and instruments.Participants comprised students at Gifu University-a national university in Japan.The STROBE guidelines were followed in the preparation of this study (von Elm et al., 2014).

Participants and procedures
An online survey was implemented during the annual health checkup usually conducted in February of each respective year before the start of the new academic year.In Japan, the health checkup is provided for all students annually and is mandatory based on the School Health and Safety Act.Thus, all students were eligible.Students were instructed and guided to complete the questionnaire through email, which they could access for 24 h from anywhere using either smartphones or personal computers.They answered the questions by selecting numerical responses.Participants were informed that their involvement in the study was unrelated to their academic evaluation or credits.
Regarding demographic data, 6441 students (3140 males, 3287 females, and 14 others) with a total mean age of 20.27 years met the inclusion criteria and were included in the analysis.Inclusion criteria were: 1) second-, third-, and fourth-year Gifu University students during the academic years 2020, 2021, or 2022; 2) completion of the CCAPS-Japanese questionnaire; and 3) no missing responses.
The dataset was separated by student class standing (second-, third-, and fourth-year student).The categorical, independent variable was the academic year: February 2020 (pre-pandemic), February 2021 (a year after pandemic onset), and February 2022 (two years after pandemic onset).The number of participants in 2021 was less than half that of other years.This may have been owing to a situation in which the survey was not well publicized to students, as 2021 was the first annual health checkup conducted with the pandemic.Participants' characteristics are summarized in Table 1.The number of COVID-19-infected students at Gifu University is summarized in Fig. 1.

Measure
The CCAPS-Japanese, which measures university students' psychological symptoms over the past two weeks, was used to assess mental health among participants (Horita et al., 2020(Horita et al., , R. 2021)).The CCAPS was originally created by the University of Michigan's counseling and Psychological Services in 2001 as a multidimensional measure designed to assess psychological issues among university students (Locke et al., 2011).The CCAPS-Japanese was adequately adapted and validated for Japanese students demonstrating rigorous factor structure, good internal consistency, adequate convergent validity, and good test-retest reliability (Horita et al., 2020(Horita et al., , R. 2021)).It comprises 55 items with eight-factor-derived subscales: depression (11 items), eating concerns (8 items), hostility (7 items), social anxiety (6 items), family distress (6 items), alcohol use (5 items), generalized anxiety (9 items), and academic distress (3 items); and four critical items: thought disturbance, suicidal ideation, violent behavior, and homicidal behavior.The CCAPS-Japanese is rated on a five-point Likert scale ranging from 0 (not at all like me) to 4 (extremely like me).Higher scores reflect higher levels of distress and symptoms (Horita et al., 2020).

Data analysis
For data analysis, SPSS 27.0 (IBM Corporation, Armonk, New York) was used.A cross-sectional analysis was conducted comparing the CCAPS-Japanese eight subscale scores (depression, generalized anxiety, social anxiety, eating concerns, family distress, academic distress, hostility, and alcohol use) against the academic year (2020, 2021, and 2022).A one-way multivariate analysis of variance (MANOVA) with post-hoc Tukey using a Bonferroni correction was conducted to assess the association between the independent variable (academic year) and the eight continuous dependent variables for each of the sub-population groups: second-, third-, and fourth-year students.Sex was used as a covariate.The partial eta squared value was reported for effect size, where 0 to <0.01 was "little to no effect," 0.01 to < 0.06 was "small effect," 0.06 to < 0.14 was "moderate effect," and ≥0.14 was "large effect."The data were considered statistically significant at the p < .05level.

Ethics considerations
All study procedures complied with the ethical requirements of the national and institutional committees that oversee human studies and with the 1964 Declaration of Helsinki and its revisions.The study design was reviewed and approved by the ethical review committee of the Graduate School of Medicine, Gifu University, Japan (no. 28-320).All participants provided written informed consent to participate.They were notified that their responses would remain confidential and anonymous.They were also notified that the participants could withdraw at any time and there were no compensation for participants.

Fourth-year students
The one-way MANOVA showed the main effect of group on the CCAPS-Japanese depression subscale score (F( 2 = 0.010).Post hoc analysis showed that the 2020 students (0.52±0.67) scored significantly higher on the alcohol use subscale results than did the 2021 and 2022 students (0.39±0.62, p < .01,0.39±0.60,p < .001).
The results of the one-way MANOVA and post-hoc analysis for fourthyear students are summarized in Table 4.

Discussion
The study aimed to clarify the extent that mental health has changed owing to the effect of the COVID-19 pandemic among Japanese second-, third-, and fourth-year university students.
Students in 2020 had all their classes on-campus, and there were no restrictions on club activities.Students in 2021 had most of their classes online and could not participate in club activities.Students in 2022 had a higher percentage (not all) of on-campus classes and club activities were partially resumed.As shown in Fig. 1, the population ratio of COVID-19 cases and deaths in Japan is very small compared to other countries.Therefore, the current results may reflect the effects of excessive behavioral restrictions rather than fear of COVID-19 infection.
The statistical test results suggest a general trend of worsening mental health profiles from 2020 (pre-pandemic before the university closed from April to June 2020), to 2021 (a year after pandemic onset), and a return to approximate pre-pandemic levels in 2022 (two years after pandemic onset) academic years.Fancourt et al. (2021) argued that the highest levels of depression and anxiety occurred in the early stages of lockdown but declined fairly rapidly, possibly because individuals adapted to circumstances.Robinson et al. (2022) also suggested that a small increase in mental health symptoms soon after the outbreak of the COVID-19 pandemic then decreased and was comparable to pre-pandemic levels by mid-2020 among most population sub-groups and in symptom types, including anxiety disorder, depression, and general mental health.
The analysis demonstrated a significant difference in means across the groups for depression, generalized anxiety, and alcohol use for the second-year students.For third-year students, generalized anxiety, social anxiety, hostility, and alcohol use significantly differed across groups.Among fourth-year students, the means for all eight subscale categories significantly differed across groups.
COVID-19 has been associated with increased anxiety and depression worldwide (Shah et al., 2021;WHO, 2022).Previously reported nationwide surveys suggested that lockdowns were associated with elevated aggression levels in the later months of the national pandemic response in the U.S. (Killgore et al., 2021) and with poor emotional states among adults in China (Qin et al., 2020).
Only a few studies on mental health among young adults, specifically university students, have been reported, and even fewer in Japan.Chen et al. (2020) showed that the isolation/lockdown aspect of the COVID-19 policy in China had a complex influence on the symptoms of obsessive-compulsive disorder, fear, hypochondria, and neurasthenia   Nagase et al., 2021).A previous study demonstrated the effect of remote learning (R. Horita et al., 2021) and COVID-19 (Horita et al., 2022) on the mental health of first-year Gifu University students.Although these previous studies (R. Horita et al., 2021Horita et al., , 2022) ) targeted first-year students, participants of studies by Chen et al. (2020) and Hammoudi et al. (2021) were recruited using random sampling through an online questionnaire or the snowball technique, respectively.Most campus-based cross-sectional studies did not account for academic year and changes in academic stress as students advanced to the next grade level.
Limited mental health studies have focused on differences based on academic year.An exception is the study by Fuse-Nagase (2022), who conducted a longitudinal study targeting second-year students in 2020 (after one year of entering university) and found that 18.3% of students scored one to two points higher and 8.2% scored higher than five points in 2021 on the K6 survey (a mental health screening tool for anxiety and depression).The effect of COVID-19 on the mental health of university students advancing to higher grades is unknown.Therefore, this study makes important contributions by investigating the mental health of second-, third-, and fourth-year undergraduate students from 2020 to 2022 at Gifu University, Japan during the COVID-19 era.
Generally, for second-year students, depression and generalized anxiety mean scores were higher in 2021 than those in 2020 and 2022.Alcohol use mean scores got smaller each year.The 2020 data were collected before the university switched to remote instruction and reflected "normal"/baseline conditions for depression and generalized anxiety for students of this class standing.A year after pandemic onset and the implementation of restrictions, feelings of isolation and uncertainty increased (Horita et al., 2022;Shah et al., 2021;WHO, 2022).This may have led to difference in depression and generalized anxiety scores on the CCAPS-Japanese questionnaire.Two years later, students may have found virtual niches; created positive coping mechanisms; and as school functions slowly began to return, may have accounted for the slight improvement in the depression and generalized anxiety scores in 2022 (Nagata et al., 2020).Additionally, alcohol use mean scores got smaller each year.As restrictions were implemented, students may not have had opportunities to participate in physical social circles/events where alcohol may have been consumed and thus may not have engaged in drinking as much in 2021 and 2022 compared to the pre-pandemic period (2020).
Regarding third-year students, similar trends for generalized anxiety and social anxiety were observed (the scores were higher in 2021 than those in 2020 and 2022), which may be explained by the same trend discussed above for second-year students.Moreover, hostility mean scores were higher in 2021 as compared to in 2020 and 2022.The extended shelter-in-place orders for a year after pandemic onset may have stimulated frustration that students may not have known how to channel or cope with and could have led to increased hostility scores (Killgore et al., 2021).As the shelter-in-place orders/pandemic continued into its second year, students may have found healthy coping mechanisms and habits, which could explain the change to below the pre-pandemic levels.Finally, alcohol use score was low in 2021, as previously stated, owing to the implemented restrictions.While second-year students did not see a rise in 2022, third-year students did, which may be attributed to the mean age being 20 years-the legal drinking age in Japan.
Finally, fourth-year students saw significant difference in all eight subscale categories.For all subscales except alcohol use, the general trend was that the scores were higher in 2021 than those in 2020 and 2022.Depression, generalized anxiety, and social anxiety can follow the previous explanation of heightened isolation and uncertainty related to the pandemic.Academic distress scores was high in 2021, potentially arising from the new online format for academic instruction; however, as students adapted by the year 2022, they felt less distress from the start of the pandemic and even compared to the pre-pandemic period.The eating concerns score was high in 2021, which may be explained by the general nutritional concerns noted among university students globally (Amatori et al., 2020;Bertrand et al., 2021;Ismail et al., 2020;Powell et al., 2021).Changes in food selection and meal quality among Japanese children (Horikawa et al., 2021) and changes in food preparation and intake among adults aged 20 to 69 years (Hayashi and Takemi, 2021) during COVID-19 were also recently reported.As there may have been general uncertainty and concerns relating to food intake, combined with overall stress, anxiety, and depression, these could have resulted in eating concerns and disorders among this at-risk group.The hostility trend may be justified similar to the results observed among third-year students, as they may have experienced increased frustration during the shelter-in-place orders, and eventually developed healthier coping mechanisms resulting in lower mean scores compared to the pre-pandemic period.Regarding family distress, as global trends showed increased anxiety, stress, and depression, these may have impacted personal relationships, including among family members, straining relations and causing distress among students (WHO, 2022).Furthermore, as students may have moved back home after having lived independently (such as in dormitories), the adjustment to living with family members may have been challenging to reacclimate, such as with having to take on new responsibilities or having less independence.Moreover, concern among students for their families' health and well-being during the pandemic may have added additional tension and distress relating to their family unit and translated to higher mean scores in 2021, which over time with potential adaption led to lower than pre-pandemic levels in 2022.Finally, alcohol use followed the same trend as second-year students wherein the means dropped from 2020 to 2021 and remained the same in 2022 as students may not have had the same social opportunities/access to engage in alcohol consumption owing to restrictions.
The general trend among all three groups demonstrated that during the 2021 academic year, there were worse mental health profiles when compared to 2020 and 2022.When comparing the aggregate results of all three grade levels, it is evident that mental health concerns increased with each academic level.Fourth-year students had greater mental health changes than did their younger counterparts.To our knowledge, this study is the first of its kind to present the difference in university students' mental health during COVID-19, with a focus on those in higher academic years.A few studies have investigated university students' mental health during the pandemic (Chen et al., 2020;Fuse-Nagase et al., 2021;Fuse-Nagase, 2022;Hammoudi et al., 2021;R. Horita et al., 2021R. Horita et al., , 2022)).However, none of these studies investigated the specific difference among upper-level university students.This research will help elucidate differences among upper-level university student populations, which may serve as the foundation for public health interventions and programs aimed at improving the mental health and overall health of students who have been sheltering-in-place.A prolonged pandemic might affect students' sense of self-esteem and identity because they would not be able to build interpersonal relationships or have social experiences such as club activities or part-time jobs throughout their college life.We speculate that the educational support should be required to create good human relationships and team work for the generation who spend campus life during pandemic.The insights obtained from this research and future program implementation that may result could serve as the foundation for culturally appropriate and long-term interventions among different populations worldwide.

Limitations and future directions
Some limitations must be acknowledged.First, there is a possibility of the social desirability bias arising from participants' self-reported responses to the CCAPS-Japanese questionnaire, which is an original limitation of the tool.Respondents may under-report socially desirable/ "bad" behaviors while over-reporting socially acceptable/"good" behaviors.Second, the difference in some CCAPS subscales by academic years was significant, but almost all effect sizes were small.Thus, it may suggest that our results lack full explanatory power.Third, the data do not follow the same students from 2020 to 2022 to understand the impact on the same person but instead considers new cohorts each year.Despite this constraint, the cohorts are representative of similar populations.Future studies could identify the specific changes in mental health adjusted for sex and educational department at the university.Furthermore, additional research may be conducted to compare mental health, lifestyle behaviors, and physical health data to acquire a better understanding of the complex links that exist across all three domains.

Conclusion
When comparing the general trends in the results, 2021 demonstrated worsened mental health profiles of students for all three grade levels.The findings denoted that COVID-19 negatively impacted the mental health of upper-level university students, which was most evident among fourth-year students.This novel study contributes considerably to the literature and provides professionals working with upper-level university students the opportunity to peruse data that can help them design appropriate public health programs to support students navigating their mental health during the pandemic.

Fig. 1 .
Fig. 1.Change the COVID-19 infected student number at Gifu University from January 2020 to February 2022.Survey I was conducted in February 2020(prepandemic).Survey 2 was conducted in february 2021 (a year after pandemic onset).Survey 3 was conducted in February 2022(two years after pandemic onset).

Table 1
Background characteristics of the participants.
SD; standard deviation, The mean age and male / female ratio are almost same in 2020, 2021, and 2022.N.Nagib et al.

Table 2
Sample survey of second-year students' CCAPS-Japanese scores and academic year.

Table 3
Sample survey of third-year students' CCAPS-Japanese scores and academic year.

Table 4
Sample survey of fourth year students' CCAPS scores and academic year.