The first case of a novel coronavirus was identified in late December 2019, and the subsequent coronavirus disease 2019 pandemic has altered people’s lives. National governments imposed interventions including lock-down, social distancing, and school closures. Reference Haug, Geyrhofer and Londei1 Such measures led to frustration and anxiety. Reference Brooks, Webster and Smith2–Reference Xiong, Lipsitz and Nasri5 In addition, the world’s gross domestic product decreased by at least 6% in 2020. 6 Economic difficulties potentially led to disproportionate mental health problems, particularly for individuals with lower socioeconomic status. Reference Witteveen and Velthorst7
Students in higher education faced academic difficulties. Closure of academic institutions forced students to shift from face-to-face to online lectures, but some students lacked the infrastructure or resources to make this change. Reference Sahu8 They also lost part-time jobs due to the economic downturn. Reference Aristovnik, Keržič and Ravšelj9 Financial insecurity was an important risk factor for worse mental health. Reference Aristovnik, Keržič and Ravšelj9–Reference Wathelet, Duhem and Vaiva12 In particular, 1st-year and 4th-year students were at risk because 1st-year ones could be more isolated due to lesser opportunities of visiting their campuses, and 4th-year ones could be concerned about uncertainty about their postgraduation careers and the economy. Reference Aristovnik, Keržič and Ravšelj9,Reference Kecojevic, Basch and Sullivan11,Reference Tang, Hu and Hu13
Japan’s Ministry of Education, Culture, Sports, Science and Technology (MECSST) reported that the proportion of students who dropped out or took a leave of absence between April and October 2020 did not largely differ from the same period in 2019. 14 Also, a previous study reported that the prevalence of 2020 first-year students who had psychological distress was lower than in 2019. Reference Horita, Nishio and Yamamoto15 On the other hand, the prevalence of academic distress in 2020 first-year students was higher than that in 2019. The reason might be that they had to prepare and adapt to an unexpected educational environment, such as online lectures. A report showed that 35.4% of the undergraduates expected a decline in their family’s income and 50.4% of those expected a decline in their total income from part-time jobs due to the pandemic. 16 Students with economic insecurity had more anxiety and worry than those without insecurity. Reference Tsurugano, Nishikitani and Inoue17 It was, therefore, important to explore the living and socio-economic environment of undergraduates in relation to psychological distress for the prevention of more serious mental disorder or possible adverse consequences for academic achievement. To our knowledge, only one study has reported the impacts of lifestyle changes due to the coronavirus disease 2019 pandemic on psychological status among Japanese medical students. Reference Nishimura, Ochi and Tokumasu18 In Japan, high school graduates enroll in university and take 6 y to pass through undergraduate medical education. Medical students who had concerns about a shifting to online education had an increased risk for depression. Requesting food aid was also a risk factor. However, to our knowledge, no study has examined this topic among the students in higher education in a wide range of Japanese population. This study aimed to examine factors potentially associated with psychological distress among undergraduate students in Japan during the coronavirus disease 2019 pandemic.
Methods
Ethical Approval
The study protocol was reviewed and approved by the Research Ethics Committee of the Osaka International Cancer Institute (approved on June 19, 2020; approval number 20084). All participants provided Web-based informed consent before responding to the online questionnaire. They could choose to respond and quit answering at any point. A credit point known as “Epoints,” which could be used for Internet shopping and cash conversion, was provided to the participants as an incentive. The exact value of “Epoints” was not disclosed at the Internet research agency’s request.
Data Sources and Participants
This cross-sectional study used data from the Japan “COVID-19 and Society” Internet Survey (JACSIS). This Web-based and self-administered questionnaire survey was conducted by a large Internet research agency in Japan. The approximately 2.2 million registered panelists covered not only students in higher education but also individuals with a wide range of social categories in Japan. The Internet research agency recruited participants from the registrants by means of accessing a designated website to answer the self-administered questionnaire. The recruitment based on random sampling stratified by sex, age, and prefecture continued until collecting 28,000 participants aged 15-79 y between August 25 and September 30, 2020 (37 d), approximately 3 mo after the first state of emergency in Japan ended. 19 Data collection ended when the target sample was reached. The response rate of 12.5% (28,000/224,389) was low due to the nature of Internet surveys. We excluded 2518 participants with invalid or inconsistent responses using the algorithm we developed earlier. Reference Matsuyama, Aida and Takeuchi20,Reference Okubo, Yoshioka and Nakaya21 Current analyses were restricted to university and college students not including postgraduates (n = 1030) based on a self-administered question. Medical students were included as undergraduates. We excluded 47 students who reported any kind of mental disorder because they likely had psychological stress since before the pandemic. Additionally, because grade skipping is rare for under 17 y old, 25 participants aged under 17 y or over 25 y at April 1, 2020 and age unknown were also excluded, leaving 958 students reported attending university or college.
Dependent Variable: Psychological Distress According to the Kessler Psychological Distress Scale
The dependent variable was nonspecific psychological distress during the past 30 d according to the Kessler Psychological Distress Scale (K6). Reference Kessler, Andrews and Colpe22–Reference Furukawa, Kawakami and Saitoh24 The K6 consists of 6 items measured on a 5-point scale (0-4): total score ranges from 0 (no distress) to 24 (maximum distress). We defined the presence of psychological distress as 5 points or more of the total K6 score, Reference Sakurai, Nishi and Kondo23 as used in previous studies. Reference Arima, Takamiya and Furuta25,Reference Fuse-Nagase, Kuroda and Watanabe26 In a secondary analysis, we also used the cutoff point of 13 or more, which is defined as severe psychological distress.
Independent Variable: Factors Potentially Related to Psychological Distress
Based on previous studies, Reference Wathelet, Duhem and Vaiva12,Reference Nishimura, Ochi and Tokumasu18 we selected variables potentially associated with psychological distress: (1) demographics, (2) socioeconomic status, (3) school-related factors, and (4) social networks.
-
(1) Demographics
Demographic variables included age at April 1, 2020 (18, 19, 20, 21, and 22-25 y), sex (men and women), residential prefecture (prefectures under special restrictions in the first declaration of a state of emergency [Tokyo, Kanagawa, Saitama, Chiba, Osaka Hyogo, Fukuoka, Hokkaido, Ibaraki, Ishikawa, Gifu, Aichi, Kyoto] and others), and living status (living alone, living with parents, and others). Age on April 1, 2020, was calculated using registered date of birth.
-
(2) Socioeconomic Status
In light of the wider context of socioeconomic status, we included a type of higher education institution (private university, public university, professional training college, junior college, and technical college), household income change during the pandemic (increased to >100%, unchanged at 100%, decreased to 50-99%, and decreased to 0-49%), receiving a student loan or scholarship (not received and received), receiving the Special Cash Payment (not received and received), part-time job loss (no, yes, not applicable, and do not know), unpaid wages (no, yes, started pre-pandemic), insufficient money to buy necessities (no, yes, started pre-pandemic), insufficient money to pay school fees (no, yes, started pre-pandemic), and insufficient money to buy food (no, yes, started pre-pandemic).
To determine household income change during the pandemic, we asked the following question: “If your previous household income was set as 100, how has your current household income changed? For example, if it has decreased by half, please answer 50, and if it has doubled, please answer 200.” The range of possible answers was 0 to 200 and “do not know.” We also used the following question to obtain information on unpaid wages, insufficient money to buy necessities, to pay school fees, and to buy food: “Since April 2020, have you had any of the following experiences?” The potential answers were “yes,” “no,” and “started pre-pandemic” for each item. We focused only on the answer of yes and no for ease of interpretation.
Typically, student loans bring in monthly monetary support of approximately 50,000 yen (100 yen ≈ 1 US dollar). Reference Sato, Watt and Saijo27 The Special Cash Payment (100,000 yen) in Japan is a financial assistance scheme for all residents, regardless of their financial situation. In 2018, more than 80% of undergraduates had a part-time job. 28
-
(3) School-Related Variables
School closure during the pandemic (no, yes, not applicable, do not know) and attending online lectures (no, yes, not applicable, do not know) were included. To obtain information on school closure, attending online lectures, and part-time job loss (referred to in the socioeconomic status section), we asked the following question: “Around April/May 2020, have you had any of the following experiences?” Participants selected from “yes,” “no,” “not applicable,” and “do not know” for each item. We focused only the on answer of yes and no for ease of interpretation.
-
(4) Social Network-Related Variable
To determine a frequency of communication with friends and acquaintances by means of email and other messages, we asked “How often have you done each of these in the last month?” with potential answers: “not done,” “once a month,” “2 or 3 times a month,” “once a week,” “2 to 3 times a week,” “4 to 5 times a week,” “almost every day (6 or 7 times a week).” We defined the first answers as “none,” the second and third as “ 1 to 3 times a month,” and fourth to last as “at least once a week.”
Statistical Analysis
We built 2 Poisson regression models with robust error variance, to calculate prevalence ratios (PRs) and 95% confidence intervals (CIs). Reference Zou29 PRs can be interpreted as relative risks. Reference Zou29 First, we estimated age- and sex-adjusted PRs of each independent variable. Second, we estimated PRs adjusted for all independent variables. Supplemental Table 1 shows the information on missing values in each variable. Supplemental Table 2 shows PRs for severe psychological distress (more than 13 cutoff points). To impute the missing values, we used a k-nearest neighbor imputation (the R package “VIM”) with the assumption of missing at random. Reference Kowarik and Templ30 A P-value <0.05 (2-tailed) was considered statistically significant. All analyses were conducted in R (ver. 4.1.0; R Foundation for Statistical Computing, Vienna, Austria).
Results
Table 1 shows the characteristics and psychological distress of the participants in this study. The proportion of psychological distress was 40.0% (383/958). The median age was 20 (1st and 3rd quartile were 19 and 21, respectively) and 56.8% (n = 544) were women. Among the participants, 56.4% were private university students, 33.0% public university students, and 10.6% professional training college, junior college, and technical college students. A total of 51.9% experienced decreases in household income, and 39.0% remained unchanged. The percentage of students who lost a part-time job or had unpaid wages was 39.5% and 2.5%, respectively. A total of 5.7% of students had insufficient money to buy necessities, 3.0% had insufficient money to pay school fees, and 4.6% had insufficient money to buy food. School closure occurred among 76.2%, of participants and 84.4% attended online lectures. A total of 16.6% did not communicate with friends and acquaintances by means of email and other messages more than once a week.
Table 1. Basic characteristics of participants and factors potentially associated with psychological distress
Note: Psychological distress was defined as 5 points or more of the total Kessler Psychological Distress Scale (K6).
Table 2 shows the results from the Poisson regression models after imputation. In the age- and sex-adjusted models, the following factors were associated with an increased proportion of psychological distress: decreasing household income, receiving student loan or scholarship, experiencing unpaid wages, having insufficient money to buy necessities, to pay school fees, or to buy food, and less frequent communication with friends and acquaintances by means of email and other messages. School closure and attendance at online lectures during the pandemic were protectively associated with psychological distress.
Table 2. Associations between psychological distress and potentially related factors from Poisson regression models after imputation
Note: Fully adjusted model simultaneously includes age, sex, residential prefecture, living status, type of higher education institution, household income change during the pandemic, receiving a student loan or scholarship, Special Cash Payment, part-time job loss, unpaid wages, insufficient money to buy necessities, to pay school fees, to buy food, school closure during the pandemic, attending online lectures, and frequency of communication with friends and acquaintances by means of email and other messages. Psychological distress was defined as 5 points or more of the total Kessler Psychological Distress Scale (K6).
Abbreviations: CI, confidence interval; PR, prevalence ratio.
In the fully adjusted model, decreases in household income of between 50% and 99% were associated with psychological distress compared with unchanged (PR = 1.48; 95% CI = 1.23, 1.77). Receiving a student loan or scholarship was associated with a PR of 1.27 (95% CI = 1.04, 1.54). Experiencing unpaid wages was associated with an increased risk for psychological distress compared with not experiencing it (PR = 1.44; 95% CI = 1.07, 1.92). Insufficient money to buy necessities was also associated with a PR of 1.45 (95% CI = 1.07, 1.95) compared with not experiencing it. Compared with communication with friends and acquaintances at least once a week, communication 1 to 3 times a month was associated with an increased risk for psychological distress (PR = 1.22; 95% CI = 1.00, 1.50). School closure during the pandemic was protectively associated with psychological distress (PR = 0.78; 95% CI = 0.63, 0.98).
In the secondary analysis, decreasing household income, insufficient money to buy food, and less frequent communication with friends and acquaintances by means of email and other messages were significantly associated with increased risk for severe psychological distress (Supplemental Table 2).
Discussion
This cross-sectional study showed the proportion of psychological distress according to the K6 was 40.0% among undergraduate students in Japan during the coronavirus disease 2019 pandemic. Experiencing decreasing household income, receiving a student loan or scholarship, experiencing unpaid wages, having insufficient money to buy necessities, to pay school fees, or to buy food, and less frequent communication with friends and acquaintances by means of email and other messages were associated with psychological distress. In contrast, school closure and attendance at online lectures during the pandemic were protective factors for psychological distress.
Economic hardship is considered the most significant predictor for worse mental health among students during the coronavirus disease 2019 pandemic. Reference Aristovnik, Keržič and Ravšelj9 Consistent with previous studies, decreasing household income, unpaid wages, having insufficient money to buy necessities, pay school fees, and buy food predicted phycological distress. In particular, the prevalence of decreases in household income was 51.9%, and the PRs for psychological distress were 1.55 and 1.54. Higher education institutions and the government of Japan have already provided financial supports, 14 which might be beneficial to mental health. Although the prevalence of having insufficient money to buy necessities and food was relatively low, the risk for psychological distress was high. Support for providing necessities and food might also be important for students’ mental health.
We initially considered that the monthly monetary support from student loans or scholarships would have a positive impact on mental health; however, receiving a student loan or scholarship was associated with psychological distress. In Japan, scholarships are rare, and student loans are the norm. Reference Sato, Watt and Saijo27 In a previous study, student loan was associated with psychological distress after graduation because loan repayment might create a financial burden. Reference Sato, Watt and Saijo27 In the coronavirus disease 2019 pandemic situation, students might have felt uncertain about their future career paths and also be worried about future loan repayment. Furthermore, as students who receive loans are from lower socioeconomic status, Reference Sato, Watt and Saijo27 they might be more vulnerable to the pandemic situation. Additional economic support for students receiving student loans might be useful for their mental health.
School closure and attendance at online lectures during the pandemic were associated with less psychological distress. The Centers for Disease Control and Prevention has warned of a potential to transmit severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) among adolescents attending school although this was still not established. 31 Graduates in Japan might also feel that school closure and attendance at online lectures could decrease the risk of infection, and might be satisfied with the measures. However, in a previous study, students concerned about changing to online lectures had a high risk for depression. Reference Nishimura, Ochi and Tokumasu18 Support may still be needed for students who have difficulty adapting to online classes. Our study also showed that students who seldom communicated with friends and acquaintances had an increased risk for psychological distress. As school closure and online lectures are recognized as producing social isolation, support for opportunities to make and communicate with friends might be also effective.
Limitations
This study had 3 limitations. First, as it is cross-sectional, the temporal relationship can be reversed. In addition, we could not assess the changes in independent variables before and after the pandemic. This means that some independent variables might not directly reflect the pandemic’s effects on student life. However, in this study, socioeconomic and lifestyle change factors due to the pandemic were also included, which were important predictors of psychological distress. Therefore, our study suggests the impact of difficulties due to the pandemic, as well as just socioeconomic difficulties. Second, all the information in this study was obtained using the self-administered questionnaire, and some questions were not validated because we did not conduct a pilot study. However, we have been conducting the Japan “Society and New Tobacco” Internet Survey (JASTIS) using similar questions. Reference Tabuchi, Shinozaki and Kunugita32 Also, the outcome in this study, the K6 score, has been previously validated. Reference Kessler, Andrews and Colpe22–Reference Furukawa, Kawakami and Saitoh24 Nevertheless, this study has limitations in relation to the use of self-reports such as misclassification. For example, we determined student status based on the self-administered questionnaire. Furthermore, there was no information on the departments and majors of the students, which might modify the associations. However, these misclassifications are thought to be non-differential to psychological distress. Third, there might be selection bias due to Web recruitment, which means the potential presence of the healthy volunteer effect. In 2020, there were approximately 3.68 million students in higher education. 33 Our study included only 958 undergraduate students, which might include students on leave. Furthermore, among 1030 graduates, just 47 students reported any kinds of mental disorder, which was considerably lower than a nationwide study. Reference Nishi, Ishikawa and Kawakami34 Furthermore, compared with earlier reports, the proportion of borrowing for student loans was approximately 30% lower. Reference Sato, Watt and Saijo27,28 As the study population could have high socioeconomic status, the prevalence of economic difficulties and its risk for psychological distress could be underestimated. Also, previous studies reported the prevalence of psychological distress according to K6 at 5 or higher was 28.5% among medical students Reference Arima, Takamiya and Furuta25 and 18.4% among first-year students Reference Fuse-Nagase, Kuroda and Watanabe26 during the coronavirus disease 2019 pandemic in Japan. Although the direct comparison is difficult due to the different target populations, the proportion of psychological distress from this study was approximately 10-20% higher than previous studies. We should consider this limitation when interpreting the results of this study.
Conclusions
In conclusion, among undergraduate students in Japan, economic difficulties significantly predicted psychological distress. Empowerment for educational institutions to provide emergency material and mental support to vulnerable students might be needed. In particular, financial support might be effective for students, as experiencing unpaid wages and a decrease in household income were associated with psychological distress. Furthermore, higher education institutions should provide opportunities to communicate with classmates, even if on the Web. Both educational institutions and the general public should recognize the socioeconomic difficulties faced by students and their impact on mental health.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/dmp.2022.245
Data availability
The data used in this study are not available in a public repository because they contain personally identifiable or potentially sensitive patient information. Based on the regulations for ethical guidelines in Japan, the Research Ethics Committee of the Osaka International Cancer Institute has imposed restrictions on the dissemination of the data collected in this study. All data enquiries should be addressed to the person responsible for data management, Dr. Takahiro Tabuchi at the following e-mail address: tabuchitak@gmail.com
Acknowledgments
We thank all the participants who voluntarily shared their time and experience for the JACSIS. We also thank Dr Julia Mortimer for her English language editing.
Author contributions
Study concept and design: Takahiro Tabuchi and Yukihiro Sato. Acquisition of data: Takahiro Tabuchi. Analysis and interpretation of data: All authors. Drafting of the manuscript: Yukihiro Sato. Critical revision of the manuscript for important intellectual content: All authors. Final approval of the version to be published: All authors
Agreement to be accountable for all aspects of the work: All authors.
Funding
This study was funded by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grants (grant number 17H03589; 19K10671; 19K10446; 18H03107; 18H03062; 19H03860), the JSPS Grant-in-Aid for Young Scientists (grant number 19K19439), Research Support Program to Apply the Wisdom of the University to tackle COVID-19 Related Emergency Problems, University of Tsukuba, and Health Labor Sciences Research Grant (grant number 19FA1005; 19FG2001).
Conflict of interests
None reported.
