A Study on the Psychological Wound of COVID-19 in University Students

fpsyg-12-589927 January 20, 2021 Time: 15:53 # 1 ORIGINAL RESEARCH published: 26 January 2021 doi: 10.3389/fpsyg.2021.589927 Edited by: Eduard Brandstätter, Johannes Kepler University Linz, Austria Reviewed by: Ceymi Doenyas, Koç University, Turkey Brigitte Lueger-Schuster, University of Vienna, Austria *Correspondence: Estrella Romero [email protected] Specialty section: This article was submitted to Personality and Social Psychology, a section of the journal Frontiers in Psychology Received: 31 July 2020 Accepted: 04 January 2021 Published: 26 January 2021 Citation: Padrón I, Fraga I, Vieitez L, Montes C and Romero E (2021) A Study on the Psychological Wound of COVID-19 in University Students. Front. Psychol. 12:589927. doi: 10.3389/fpsyg.2021.589927 A Study on the Psychological Wound of COVID-19 in University Students Isabel Padrón1, Isabel Fraga1, Lucía Vieitez1, Carlos Montes2and Estrella Romero3* 1Cognitive Processes & Behavior Research Group, Department of Social Psychology, Basic Psychology, and Methodology, University of Santiago de Compostela, Santiago de Compostela, Spain, 2Department of Social, Psychology, Basic Psychology, and Methodology, University of Santiago de Compostela, Santiago de Compostela, Spain, 3UNDERISK, Department of Clinical Psychology and Psychobiology, University of Santiago de Compostela, Santiago de Compostela, Spain An increasing number of studies have addressed the psychological impact of the COVID-19 crisis on the general population. Nevertheless, far less is known about the impact on specific populations such as university students, whose psychological vulnerability has been shown in previous research. This study sought to examine different indicators of mental health in university students during the Spanish lockdown; we also analyzed the main sources of stress perceived by students in relation to the COVID-19 crisis, and the coping strategies adopted when faced with the situation. Data was collected from 932 students (704 women) through a web-based platform. Measures of anxiety (i.e., GAD-7), depression (PHQ-9), irritability, and self-perceived change in mental health were administered, as well as ad hoc measures of stressors and coping strategies. Results indicated that students experienced considerable psychological problems during the confinement, with higher rates of emotional difficulties in women and undergraduate students than in men and postgraduates, respectively. Psychological distress was mainly related to several specific domains of stressors, as perceived by the participants: academic future, task overload, worsening of interpersonal conflicts, and restrictions in pleasant social contact; and far less related to the spread of the disease and its consequences for physical health. As regards coping strategies, both reframing skills and daily routines were shown to be the most effective. A path-analysis model integrating stressors, coping, and mental health revealed that coping strategies partially mediated the effect of stressors on psychological health. In general, results suggest that students’ psychological health was substantially affected by the COVID-19 situation and that the academic and relational changes were the most notable sources of stress. This study reinforces the need to monitor and promote mental health in university students to boost resilience in times of crisis. Our results on effective coping strategies may inform preventive programs aimed at helping students to deal with challenges like the COVID-19 pandemic. Keywords: university students, psychological impact, COVID-19, stressors, coping, COnVIDa-20 Frontiers in Psychology | www.frontiersin.org 1January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 2 Padrón et al. The Psychological Wound of COVID-19 INTRODUCTION In late December 2019, a new coronavirus disease (COVID- 19) emerged in the Chinese city of Wuhan. This new disease, caused by SARS-CoV-2 (Severe Acute Respiratory Syndrome Coronavirus-2), spreads very easily from one person to another and thus rapidly affected other parts of China (Wang et al., 2020). Within a few weeks, the first cases emerged in other countries, and COVID-19 soon became a global threat. Indeed, in March 2020, the World Health Organization (WHO) declared COVID- 19 to be a global pandemic. As of July 06, 2020 (the date of writing this article), about 9,843,073 confirmed cases, including 495,760 deaths, have been reported by the WHO (World Health Organization [WHO], 2020a). Following the initial outbreak in China, further outbreaks occurred in Italy and Spain, the first Western countries to be affected, and thus the first countries to face a problem that had hitherto seemed a very distant one to the populations of Western societies. In fact, at the time of writing, Spain is one of the three countries (after Belgium and the United Kingdom) with the highest rate of confirmed cases and deaths per million inhabitants (July 06, 2020, cf. World Health Organization [WHO], 2020c). In response to the COVID-19 outbreak, many countries were forced to adopt severe restrictive measures to slow down its propagation. In the case of Spain, the Spanish government declared a state of emergency on March 14th, and a population lockdown became mandatory 2 days later, creating an unprecedented situation. Citizens only could leave their homes for food, to go to a pharmacy, and for other essential needs. Schools and universities were closed throughout the national territory. Employers (public and private) were obliged to work from home whenever possible, and many lost their jobs temporarily or permanently. In a matter of days, millions of people’s lives changed dramatically, leading to important questions about how the pandemic was affecting not only the physical but also the mental health of the population. Research initially focused mainly on the impact of COVID- 19 on physical health and its clinical characterization (e.g., Cao and Li, 2020;Lvov et al., 2020), with studies on the psychological impact of the quarantine subsequently beginning to appear (e.g., Cao et al., 2020;Wang and Zhao, 2020). Findings of all studies (the first of these on the Chinese population, and soon after on other populations of Western countries) revealed a significant and severe increase in depressive symptoms, anxiety, and stress levels due to confinement (e.g., Cao et al., 2020; Chen et al., 2020;Wang and Zhao, 2020) which was generally more pronounced in women than in men (Flesia et al., 2020; Rodríguez-Rey et al., 2020;Wang and Zhao, 2020). Moreover, as recent reviews suggest (Brooks et al., 2020;Hossain et al., 2020), other negative psychological effects, such as post-traumatic stress symptoms, anger, panic, irritability, low self-esteem, and lack of self-control, are commonly found among individuals affected by physical isolation. Data from other pandemics and natural disasters revealed similar effects (see Brooks et al., 2020, for a review). A recent technical report on the Spanish population between 18 and 75 years of age, starting 26 days after the first state of emergency, also revealed significant rates of symptoms of depression (22.1%) and anxiety (19.6%) (Valiente et al., 2020). Importantly, results from this and other studies (e.g., Flesia et al., 2020;Ozamiz-Etxebarria et al., 2020;Rodríguez-Rey et al., 2020) reveal that not all groups were affected in the same way, with young adults (18–24) being more affected compared to other age groups. To design action plans aimed at protecting and helping citizens who may be affected in different ways by these types of situations, it is first necessary to establish the specific effects of the pandemic in different populations. Interestingly, although studies on the general population are accumulating, the impact on university students is still not well known. There is, however, an abundance of work indicating that most mental health disorders have first onset in young adulthood (e.g., Kessler et al., 2001;Eisenberg et al., 2007). Several studies have also reported that students have consistently higher levels of mental health problems than the general population (Overbeek et al., 2003;Zivin et al., 2009;Auerbach et al., 2016). Moreover, research in this field notes the importance of personal and psychosocial factors in the emergence and development of mental disorders in university students (Galindo et al., 2009). Besides academic issues, university students are exposed to multiple stressors which are unique to this developmental period, such as the abandonment of the family home, adjusting to new social and geographical environments, making new friends and social relationships, life-stage transitions, time management, economic resources, etc. (Beiter et al., 2015;Fried, 2020). The lockdown substantially affected these conditions, and the well-being of the university student population would also be expected to have been affected. In brief, university students constitute a population that is particularly vulnerable in terms of mental health, one that even before the pandemic showed a high prevalence of mental disorders. Moreover, many universities suspended normal class-based teaching and moved online, with the result that the lives of students changed drastically (Sahu, 2020) and their psychosocial functioning was negatively disrupted, thus altering the social integration of some individuals (Elmer et al., 2020). It is within this situation, then, that we ask how students’ psychological health might have been affected by the pandemic and the confinement. Thus far only a few papers have addressed this issue specifically. Recent studies on Chinese undergraduate students reported higher levels of anxiety during the COVID-19 outbreak (e.g., Cao et al., 2020;Wang and Zhao, 2020). In the Cao et al. (2020) study, for example, the authors found that 24.9% of students experienced symptoms of anxiety, with 0.9% of cases being severe and 21.3% mild anxiety. Moreover, it was reported that some factors, such as place of residence, source of parental income, whether living with parents or not, and having a relative or an acquaintance infected with COVID-19, were associated with increased anxiety. Another recent study, looking at the initial psychological impact of the COVID-19 outbreak on a population of Spanish university students (77%), academic staff (13%), and administrative staff (9%), revealed significantly higher anxiety, depression, and stress scores in the students than in the other two groups (Odriozola-González et al., 2020). Meanwhile, Elmer et al. (2020) analyzed changes in social networks and Frontiers in Psychology | www.frontiersin.org 2January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 3 Padrón et al. The Psychological Wound of COVID-19 mental health in a sample of Swiss undergraduate students during the COVID-19 crisis (2 weeks after the lockdown) in relation to the two preceding years. They also analyzed individual and social factors associated with changes in mental health during the outbreak. In line with previous work, students were found to feel more depressed, anxious, stressed, and lonely than half a year before, and these negative effects were more prominent in women. Also, concerns about family and friends, future careers, living alone, and having less social contact and support, were linked to worse mental health. Even though some studies have not found significant changes in students’ mental health during the crisis (Fried, 2020), the general results are congruent with previous research on the psychological impact of previous pandemics in China, in which students presented high levels of stress and anxiety during SARS (Jia et al., 2003;Wong et al., 2007) or N1H1 influenza (Li et al., 2011). Although a small number of studies have focused on student populations, we know very little about the specific sources of stress that affected university students during the most acute stage of the crisis, and about the differential impact that these had on the psychological well-being of students. In addition to the stressors present in the general population, such as prolonged isolation periods, fear of infection, frustration and boredom, inadequate information (excess or confusing information), financial loss, and inadequate supplies, as noted by Brooks et al. (2020) and Hossain et al. (2020), university students were also greatly affected by significant academic changes resulting from the pandemic. All Spanish universities suspended face-to-face teaching and moved to online classes and examinations, a transition likely to have had a serious impact on students’ feelings of anxiety and uncertainty. As Sahu (2020) has observed, the quality of online education, and its consequences, is a critical issue here, because students without adequate Internet access will experience even more stress, which can have detrimental effects on their academic performance and achievement, as well as on their mental health. We might add that not all individuals possess equal levels of technological proficiency or confidence, and that the Internet itself was often slow and unreliable during this period, all of which meant that the interactive online environment was sometimes a less than optimal experience. With respect to these specific sources of stress, a study by Okruszek et al. (2020) with Polish young adults found that the COVID-19 risk perception (e.g., contact risk, severe symptoms risk, and financial problem risk) and the feeling of loneliness led to affective responses to the situation during the first 2 weeks of the crisis. However, it is also interesting to note that some individual stressors, such as work overload, strict schedules, Fears of Missing Out on social life (FoMO) or competition among students might in fact have been mitigated during the crisis situation for some students (Elmer et al., 2020). There is also little prior evidence about coping strategies used by young adults in this situation. Although the structure of coping strategies is still a controversial issue (Stanisławski, 2019), many different coping behaviors have been considered in the long tradition of research on stress: some of the coping behaviors are considered more “active,” i.e., directed to cognitively or behaviorally change the stressor (e.g., problem solving, cognitive reframing; Tobin et al., 1989;Gaudreau, 2017), while others are more “passive” or “disengaged” behaviors (e.g., distracting activities, substance use, social withdrawal). Apart from those well-known strategies, some other coping behaviors have been identified in collective crises; for example, following routines or involving in healthy habits (Fullana et al., 2020). Likewise, previous research has suggested the coping value of altruistic behaviors in wide-scale stressful situations (e.g., Sharma and Kar, 2019). Some decades ago, Midlarsky (1991) proposed that helping may be considered as a coping mechanism, which may be effective through different psychological processes; for example, orientation to others may distract the individual from own troubles; it may also enhance the sense of competence and selfefficacy and may provide a meaning to life in the middle of adversity; additionally, helping others may also promote social integration, and evoke reciprocal support from other people. Recent studies have also emphasized the coping function of helping (Vollhardt, 2009), and physiological and neurochemical pathways have been identified (e.g., activation of parasympathetic system, oxytocin levels, and dopaminergic activity) to explain the buffering effects of helping behavior (Raposa et al., 2016). During the COVID-19, some preliminary studies have also reported the use of helping as a coping mechanism (Balluerka et al., 2020). Nevertheless, research on coping during the pandemics is still limited; particularly, the impact of the various types of coping on students’ adjustment in the COVID-19 crisis is largely unknown. For all the above reasons, the present study was conducted, and had three main goals. The first of these was to study the psychological impact of the COVID-19 pandemic on mental health in a sample of Spanish university students after 6 weeks of quarantine. The second was to analyze the main sources of stress associated with the COVID-19 context, including stressors arising from the pandemic, measures of social isolation, and changes experienced by students in the academic environment. Finally, we sought to address the issue of which coping strategies were used by students, and how such strategies were related to psychological health during the lockdown. Results were expected to help explore questions of how psychological health was affected during the acute part of the crisis, as well as which stressors and coping behaviors may explain the differences in mental adjustment when faced with the challenges of COVID-19. MATERIALS AND METHODS Participants This study forms part of the wider COnVIDa-20 project, which aims to identify the psychosocial needs and challenges, plus the skills and resources, of Spanish university students during the COVID-19 pandemic; this is the first report on the data from the COnVIDa-20 project. A total of 932 Spanish students participated in the current survey, the majority being women (see Table 1), and with all levels of university education represented. Most participants were students from one of the 3 public universities of the Autonomous Region of Galicia, an area in the Frontiers in Psychology | www.frontiersin.org 3January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 4 Padrón et al. The Psychological Wound of COVID-19 TABLE 1 | Participants’ main demographic variables and psycho-social characteristics. Variables N% Gender Women 704 75.5 Men 222 23.8 Others 6 0.6 Age 18–20 280 30.0 21–23 417 44.7 24–26 112 12.0 27–29 47 5.0 30 or more 76 8.2 University Universidade de Santiago de Compostela 661 70.9 Universidade de A Coruña 38 4.1 Universidade de Vigo 155 16.6 Others 78 8.4 Level of education Undergraduates 810 86.9 Postgraduates 85 9.1 Ph.D. students 37 4.0 Branch of knowledge Arts and humanities 161 17.3 Experimental sciences 111 11.9 Health sciences 391 42.0 Engineering and architecture 84 9.0 Social and legal sciences 183 19.7 Self-perceived social class Low 40 4.3 Low-middle 245 26.3 Middle 543 58.3 Middle-high 102 10.9 High 2 0.2 Diagnosed with a mental health problem before quarantine Yes 130 15.7 During quarantine Returned to parent’s home 504 56.0 Being high risk COVID-19 people Because of age 16 1.8 Due to previous diseases 102 11.3 Because being essential workers 20 2.2 No 770 85.6 Having lived with other high risk COVID-19 people Because of age 235 26.1 Due to previous diseases 307 34.1 Because they work in the medical field 109 12.1 Because being essential workers 317 35.2 No 290 32.2 Suffered COVID-19 Myself 8 0.9 Someone close to me 206 22.9 No 704 78.2 Northwest of Spain. They were classified into 5 groups according to the branch of knowledge to which their undergraduate, master or Ph.D. program belonged: Arts and Humanities, Experimental Sciences, Health Sciences, Engineering and Architecture, and Social and Legal Sciences. When the demographics of the sample are compared to the overall population in the main university of Galicia (USC) and in the whole Spanish university system (SUE; see Table 2), we find that our sample is composed of a relatively higher proportion of women: 75.5% versus 61.1% and 55.3.% in USC and SUE, respectively. With respect to the branches of knowledge, most of the participants in our study were enrolled in programs related to Health Sciences (42.0%), Social and Legal Sciences (19.7%), and Arts and Humanities (17.3%); these proportions are very similar in the USC system (see Table 2), but not totally in the SUE system, where a higher proportion of undergraduates enroll in Social and Legal Science programs (47.1%). With regards to educational levels, the high representation of undergraduate students in our study (86.9%) was very similar to the proportion in both the USC (81.6%) and SUE (80.2%). A large percentage of the students (56.0%) in the sample moved back to their parents’ home during the pandemic and reported having lived with high risk COVID-19 people (83.4%) or being part of the high-risk population themselves (15.1%). While most of the participants had not suffered the COVID-19, 22.8% had lived in an environment with an infected person close by. Variables and Instruments For the purposes of the present study, measures of psychological health, psychosocial stressors, and coping strategies were administered in the context of the broader COnVIDa-20 project. Psychological Health Four measures were used as indicators of psychological health. Specifically, instruments for the assessment of TABLE 2 | Percentage of students enrolled in university studies as a function of gender, level of education and branch of knowledge in our study and for the USC and SUE systems. % of students Variables Our study USC Statistics for all Spanish universities (SUE) Gender Women 75.5 61.6 55.3 Men 23.8 38.4 44.7 Others 0.6 – – Level of education Undergraduates 86.9 81.3 80.2 Postgraduates 9.1 7.9 14.3 PhD students 4.0 10.8 5.5 Branch of knowledge Arts and humanities 17.3 13.2 10.4 Experimental sciences 11.9 11.9 6.5 Health sciences 42.0 30.5 18.3 Engineering and architecture 9.0 7.0 17.7 Social and legal sciences 19.7 37.4 47.1 Data retrieved from the online resource “EDUCAbase”, created by the Spanish government. The statistics refer to the 2019–2020 academic year. Frontiers in Psychology | www.frontiersin.org 4January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 5 Padrón et al. The Psychological Wound of COVID-19 anxiety, depression, irritability, and self-perceived change in psychological health were administered. Anxiety Students were asked to respond to the 7-item Generalized Anxiety Disorder Scale (GAD-7; Spitzer et al., 2006). On the GAD-7 scale, symptoms of anxiety over the last 15 days (e.g., “Feeling nervous, anxious, or on edge”) were reported using a 4- point Likert rating scale ranging from 0 (not at all) to 3 (almost every day), and thus total scores ranged from 0 to 21. Scores of 5, 10, and 15 were taken as the cut-off points for mild, moderate, and severe anxiety, respectively. When used as a screening tool, scores of 10 or higher were taken as suggestive of a significant pattern of anxiety (García-Campayo et al., 2010). Cronbach’s αin this study was 0.89 (MIC = 0.54). Depression The Patient Health Questionnaire (PHQ-9; Kroenke et al., 2001) was administered as a measure of depression. Taking a 15- day period as a reference, participants were asked to rate the presence of depressive symptoms (e.g., “Feeling down, depressed, or hopeless”) using a 4-point Likert scale, from 0 (not at all) to 3 (nearly every day), and thus total scores ranged from 0 to 27 (Cronbach’s α= 0.87, MIC = 0.44). Scores of 5, 10, 15, and 20 were considered as cut-off points for mild, moderate, moderately severe, and severe symptoms, with scores of 10 or higher being indicative of possible depression. Irritability Since feelings of irritability have also been described as common effects of the isolation measures in health-related crises (Brooks et al., 2020), the Brief Irritability Test (BIT; Holtzman et al., 2015) was administered. This scale is composed of 5 items (e.g., “I have been feeling like I might snap”) with a 6-point Likert scale from 1 (never) to 6 (always). Cronbach’s αwas 0.92, MIC = 0.71. Self-perceived change in mental health In order to measure subjective changes associated with the COVID-19 situation specifically, we asked students whether they had perceived changes in their mental health during quarantine, using a Likert scale ranging from 1 (my mental health is much worse) to 5 (my mental health is much better); therefore, a high score in this scale indicates that the participants self-perceived an improvement, whereas a low score indicates that the participants perceived a deterioration in their mental health. Stressors Associated With the COVID-19 Context We developed 26 items to assess possible sources of stress during the quarantine. Participants had to indicate, on a Likert scale, the extent to which they had been disturbed by these during the quarantine, with 1 being “not at all” and 4 “a lot”. These items were subjected to Principal Components Analysis with Varimax Rotation. Both scree test and Kaiser’s criteria recommended a meaningful 5-factor solution, which grouped the stressors in the following domains: Academic stressors (e.g., “Not receiving the academic training that one expected”), Social distancing (e.g., “Being required to stay at home for so long”), Pandemic (e.g., “The risk that either you or people close to you might become infected by COVID-19”), General overload (e.g., “Lack of free time”) and Interpersonal conflicts (e.g., “The intensification of family conflicts”). Scales were created for the five domains by averaging the items aligned with each factor, with Cronbach’s α ranging from 0.69 (Interpersonal conflicts) to 0.80 (Academic stressors), and MIC ranging from 0.34 (Social distancing) to 0.40 (Academic stressors). The distribution of items across scales is shown as Supplementary Table 1. Coping Strategies in the COVID-19 Context A set of 14 items was administered to assess the strategies displayed by participants in dealing with difficulties encountered in the COVID-19 situation. The items were mainly based on the Brief COPE questionnaire (Carver, 1997), and encompassed strategies potentially relevant to students in the quarantine: for instance, emotional support (“Looking for understanding and emotional support from others”), trying to actively improve the situation (“Concentrating my efforts on looking for a solution that might resolve the difficulties I’m facing”), instrumental support (“Trying to get help and advice from other people”), self-distraction (“Doing something to distract me from the difficulties I’m facing”), substance use (“Consuming alcohol or other substances to feel better”), spiritual attitude (“Trying to find solace in my spiritual or religious beliefs”), venting (“Expressing my negative feelings”), humor (“Trying to laugh at the situation”), and acceptance (“Accepting the reality of the fact that this is happening and adapting myself to the situation”). We also developed additional items specifically suited to the quarantine situation: routine maintenance (“Trying to maintain routines and schedules”), self-care (“Taking care of my health (nutrition, exercise. . .) to be psychologically stronger”) and helping others (“Helping others with their own difficulties”). The items were rated on a 4-point Likert scale, from 0 (not used) to 3 (used a lot). Principal Component Analysis with Varimax rotation led to the identification of four domains, based on scree and Kaiser’s rules. The items were grouped by the analysis (factor loadings of at least 0.40) as follows: Other-oriented coping, which includes both seeking and providing help (e.g., “Trying to get help and advice from other people”; “Helping others with their own difficulties”), Reframing (e.g., “Looking for something good in what is happening”; “Concentrating efforts on looking for a solution to resolve difficulties”), Disengagement Activities (e.g., “Doing something to distract me from the difficulties I’m facing”; “Doing relaxing activities”), and Structure/Healthy Routines (“Trying to maintain routines and schedules”; “Taking care of my health (nutrition, exercise. . .) to be psychologically stronger”). According to this distribution, items were averaged as a means of composing four coping scales, with Cronbach’s αranging from 0.57 (Reframing) to 0.72 (Other-oriented), and MIC ranging from 0.31 (Reframing) to 0.40 (Other-oriented). The final scales are shown as Supplementary Table 2. Procedure The questionnaires were conducted using an internal web application, which was available online from April 27th to May 27th. Students were invited to participate mainly via WhatsApp and were encouraged to spread the link to other students using Frontiers in Psychology | www.frontiersin.org 5January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 6 Padrón et al. The Psychological Wound of COVID-19 the same platform, although other social media platforms like Twitter were also used. Hence a snowballing technique was used. This study was approved by the Bioethics Committee of the Universidade de Santiago de Compostela, and prior to beginning the questionnaire, participants were provided with the aims and requirements of the study. They were also asked to give their explicit agreement to participate in the study and were informed that participation was completely anonymous and voluntary. On average, the survey took 20 min to complete and there was no reward or compensation for participating. Data Analysis Firstly, descriptives for the stressors, coping domains, and indicators of psychological health were computed, with specific focus on percentages of participants scoring high in the measures that have cut-off points, i.e., anxiety and depression. Second, a multivariate path model was used to examine the relationships among sources of stress, coping strategies, and psychological health. Specifically, the significance of direct and indirect effects was tested through a path analysis in AMOS v24 using maximum likelihood bootstrapping techniques (5,000 bootstrap iterations) and bias-corrected 90% confidence intervals (Preacher and Hayes, 2008). Several fit indices were used to test the model fit, including the χ2statistic, comparative fit index (CFI), and root mean square error of approximation (RMSEA). Models with a CFI value of 0.90 or higher (Hu and Bentler, 1999) and a RMSEA value below 0.08 (Browne and Cudeck, 1993) were considered to have an acceptable fit. RESULTS Descriptives for the Studied Variables and Rates of Psychological Problems Table 3 sets out main descriptives for the measures of psychological health, sources of stress, and coping strategies. TABLE 3 | Descriptives of the main variables in the study. Variables Mean (SD) Range Psychological health Anxiety 10.82 (5.00) 0–21 Depression 12.98 (6.46) 0–27 Irritability 16.88 (6.23) 5–30 Changes in mental health 2.19 (0.77) 1–5 Stressors Academic 3.06 (0.72) 1–4 Social distancing 2.94 (0.66) 1–4 Pandemic 2.92 (0.59) 1–4 General overload 2.84 (0.78) 1–4 Interpersonal conflict 1.99 (0.74) 1–4 Coping Other-oriented 1.48 (0.68) 0–3 Reframing 1.65 (0.63) 0–3 Structure/healthy routines 1.39 (0.84) 0–3 Disengagement activities 1.80 (0.75) 0–3 The descriptives for the scales of psychological health (anxiety, depression, and irritability) are relatively high for community populations, indicating a high average level of emotional disturbance. Regarding the analysis of stressors, this shows that academic stressors were rated as the most disturbing, on average, whereas interpersonal conflicts were the least disturbing. As for coping strategies, the involvement in disengagement activities (distracting and relaxing) was the most widely used by students during the quarantine, whereas the structure/healthy routines strategy was reported to have been the least used. The descriptive statistics for each of the specific stressors and coping strategies are presented as Supplementary Tables 1 and 2. Among the stressors, the highest mean was achieved by “Uncertainty about the evaluation of the subjects you are taking” (mean = 3.45), followed by “The economic future of society as a consequence of the crisis” (mean = 3.40), “Uncertainty about the COVID-19 crisis” (mean = 3.36) and “Lack of face- to-face contact with loved ones” (mean = 3.30). Among the coping strategies, “Accepting the reality of the fact that this is happening and adapting myself to the situation” was the most used (mean = 2.02). Other strategies with high means were “Doing something to distract me from the difficulties I’m facing” (mean = 1.88), “Doing relaxing activities” (mean = 1.73), and “Helping others with their own difficulties” (mean = 1.72). Regarding the rates of psychological problems, when the cutoffs for anxiety are taken into account, 61.2% of participants scored equal to or higher than 10, i.e., the cut-off usually considered for identifying significant anxiety, according to the norms of the scale; specifically, 38.8% showed moderate anxiety, and 22.4% severe anxiety. As for the depression scale, 65.8% of participants scored equal to or higher than 10, which is the usual cut-off taken as a reference for depression screening (Manea et al., 2012): 23.4% showed symptoms that were moderate, 25.2% ones that were moderately severe, and 17.2% severe symptoms. When rates for anxiety and depression are compared across genders, significant differences are found. For anxiety, 63.8% of women and 52.8% of men scored above the cut-off (χ2= 7.79, 1 df, p<0.006). For depression, 68.0% of women and 58.8% of men surpassed the cut-off (χ2= 5.68, 1 df, p<0.02). The “others” gender could not be introduced into the comparisons due to the small size of the group. Differences were also found for the level of university studies (χ2= 12.02, 2 df, p<0.002); the rates for anxiety were 63.5% (undergraduates), 45.1% (postgraduates), and 46.7% (Ph.D. students). For depression, the rates were 68.3, 47.9, and 50%, respectively (χ2= 15.43, 2 df, p<0.001). No differences were found across branches of academic knowledge. The scores for self-perceived changes indicate that most participants felt that their mental health actually changed during the COVID-19 crisis, with a mean of 2.19 within a range from 1 (change to much worse) to 5 (change to much better). In terms of percentages, 14.7% perceived that they were much worse, 57.5% worse, 22.7% did not perceive any change, 4.2% perceived that they were better, with only 1% reporting that they felt much better. Additional analysis by gender revealed that the mean change in women was worse than for men (2.15 vs. 2.32; F[1,822] = 7.25, Frontiers in Psychology | www.frontiersin.org 6January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 7 Padrón et al. The Psychological Wound of COVID-19 TABLE 4 | Correlations between the constructs used in the path analysis. Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 (1) Academic 0.42*** 0.45*** 0.53*** 0.39*** 0.26*** −0.01 −0.09* 0.05 0.45*** 0.42*** 0.36*** −0.41*** (2) Social distancing 0.36*** 0.45*** 0.42*** 0.34*** −0.05 0.00 0.11** 0.43*** 0.40*** 0.38*** −0.42*** (3) Pandemic 0.30*** 0.29*** 0.33*** 0.09** 0.00* 0.12*** 0.34*** 0.24*** 0.24*** −0.22*** (4) General overload 0.35*** 0.26*** −0.02 −0.09* −0.05 0.47*** 0.43*** 0.37*** −0.39*** (5) Interpersonal conflict 0.13*** −0.04 −0.04 −0.03 0.45*** 0.43*** 0.48*** −0.35*** (6) Other oriented 0.27*** 0.23*** 0.31*** 0.26*** 0.15*** 0.17*** −0.18*** (7) Reframing 0.35*** 0.35*** −0.15*** −0.21*** −0.13*** 0.24*** (8) Structure/healthy routines 0.30*** −0.12*** −0.28*** −0.10** 0.20*** (9) Disengagement activities −0.06 −0.09** −0.03 0.09** (10) Anxiety 0.75*** 0.70*** −0.57*** (11) Depression 0.63*** −0.57*** (12) Irritability −0.50*** (13) Self-perceived change *p <0.05; **p <0.01, ***p <0.001; the correlations that remaining significant after Bonferroni’s correction (p <0.0006) are displayed in bold characters. p<0.008). No differences were found across university levels or branches of academic knowledge. Multivariate Path Model With the aim of analyzing how the different domains of stressors and coping strategies might have impacted psychological health, we then performed a path analysis. Gender and age were used as covariates in the model in order to control for effects on anxiety, depression, irritability, and self-perceived change. First, we tested a saturated model in which all the paths (both direct and indirect) were included. We then tested the fit of a reduced model, in which only the significant paths and covariances were retained. Table 4 presents the correlations between the constructs used in the path analysis. All sources of stress, as well as coping strategies, correlated significantly with anxiety, depression, irritability, and self-perceived change, except for disengagement activities. The highest correlation was found between interpersonal conflict and irritability (r= 0.48), and the lowest between disengagement activities and depression (r=−0.09). As expected, anxiety, depression, and irritability correlated strongly with each other (rs ranged from 0.63 to 0.75) and were negatively correlated to self-perceived change (rs ranged from −0.57 to −0.50). Figure 1 shows the final model in which only the significant paths and covariances were retained. The final model fits the data well, χ2= 237.40, df = 37, p<0.001, RMSEA = 0.08, CFI = 0.95, and predicted variation in health to an acceptable degree; explained variance ranged from 31% (irritability) to 38% (anxiety). Standardized regression (β) weights were used to examine the size and significance of the direct effects of the stressors specified within the model (Byrne, 2016). Tables 5,6present coefficients from the path analysis model, after controlling for gender and age. Results of the path analysis revealed that pandemic stressors showed a significant positive association with all the coping strategies. Social distancing was positively associated with other-oriented strategies, routines, and disengagement activities, but negatively associated with reframing. In contrast, general overload had a significant negative association with routines and disengagement activities, and a positive one with other-oriented strategies. Academic stressors only had a direct and significant negative effect on routines. Regarding the coping-psychological health pathway, reframing was associated with better psychological health consistently across measures, that is, less anxiety, less depression, less irritability, and fewer unfavorable changes in mental health. The negative association of routines with depression was particularly strong (β=−0.19, p<0.001). Conversely, otheroriented coping predicted higher levels of psychological ill-being: more anxiety (β= 0.08, p<0.001) and the perception of feeling worse, as indicated by the negative coefficient linking otheroriented and self-perceived health change (β=−0.07, p<0.05). The coping domain involved in psychological outcomes to the least extent was that of disengagement activities, which had no significant effect. Furthermore, once adjusted for confounding variables, decomposition of total effects (Table 7) showed that both the direct and indirect effects of sources of stress on psychological measures were statistically significant (p<0.05), providing a definite pattern: Higher general overload, social distancing, interpersonal conflict, and academic stressors were significantly associated with higher anxiety, depression, and irritability, and self-perceived change to a worse mental health. Pandemic stressors, in turn, only had a direct effect on anxiety, β= 0.092, p<0.001. The final model revealed 13 indirect pathways among sources of stress and psychological health measures throughout the model; however, the proportion of mediated effects (%) were weaker in magnitude, ranging from 7.34 to 18.84%. The indirect effects of social distancing (β=−0.002, 90% CI: −0.021, 0.016, p>0.10) and the pandemic (β=−0.005, 90% CI: −0.019, 0.009, p>0.10) on depression and anxiety, respectively, were no longer significant. Frontiers in Psychology | www.frontiersin.org 7January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 8 Padrón et al. The Psychological Wound of COVID-19 FIGURE 1 | Relationship between stressors, coping, and health-related indicators. All reported path coefficients (p <0.05) are standardized estimates. Terms of error, correlations, and covariances were omitted for clarity. TABLE 5 | Standardized and unstandardized regression weights, standard errors, z-values, and associated p-values for coping. Coping Stressor SRW URW SE C.R. p-value Other oriented ←Social distancing 0.20 0.20 0.03 5.58 0.000 Other oriented ←Pandemic 0.22 0.24 0.03 6.35 0.000 Other oriented ←General overload 0.10 0.09 0.03 3.01 0.003 Reframing ←Social distancing −0.10 −0.09 0.03 −2.77 0.005 Reframing ←Pandemic 0.13 0.13 0.03 3.54 0.000 Structure/healthy routines ←Academic −0.13 −0.19 0.04 −3.32 0.000 Structure/healthy routines ←Social distancing 0.08 0.10 0.05 2.06 0.039 Structure/healthy routines ←Pandemic 0.13 0.19 0.05 3.48 0.009 Structure/healthy routines ←General overload −0.09 −0.09 0.04 −2.30 0.021 Disengagement activities ←Social distancing 0.12 0.14 0.04 3.21 0.001 Disengagement activities ←Pandemic 0.11 0.14 0.04 2.96 0.003 Disengagement activities ←General overload −0.14 −0.13 0.03 −3.87 0.000 SRW, standardized regression weights; URW, unstandardized regression weights; SE, standard error. Frontiers in Psychology | www.frontiersin.org 8January 2021 | Volume 12 | Article 589927 fpsyg-12-589927 January 20, 2021 Time: 15:53 # 9 Padrón et al. The Psychological Wound of COVID-19 TABLE 6 | Standardized and unstandardized regression weights, standard errors, z-values, and associated p-values for health-related indicators. Health-related indicator Coping SRW URW SE C.R. p-value Anxiety ←Other oriented 0.08 0.61 0.17 3.63 0.000 Anxiety ←Reframing −0.14 −1.09 0.22 −4.84 0.000 Anxiety ←Disengagement activities −0.05 −0.32 0.14 −2.26 0.023 Depression ←Reframing −0.12 −1.24 0.29 −4.28 0.000 Depression ←Structure/Hhealthy routines −0.18 −1.37 0.17 −7.96 0.000 Irritability ←Reframing −0.10 −0.97 0.28 −3.42 0.000 Self-perceived change ←Other oriented −0.07 −0.08 0.03 −2.43 0.015 Self-perceived change ←Reframing 0.19 0.24 0.03 6.33 0.000 Self-perceived change ←Structure/healthy routines 0.10 0.09 0.02 3.63 0.000 SRW, standardized regression weights; URW, unstandardized regression weights; SE, standard error. TABLE 7 | Path analysis testing the indirect effects of the variables that entered into the model. Variables Total effects β(90%CI) Direct effect β(90%CI) Indirect effect β(90%CI) Anxiety General overload 0.21*** (0.15,0.27)0.20*** (0.14,0.26)0.01** (0.00,0.02) Pandemic 0.08** (0.04,0.12)0.09*** (0.05,0.13)−0.00 (−0.01,0.00) Social distancing 0.13*** (0.07,0.19)0.11** (0.05,0.17)0.02** (0.01,0.04) Interpersonal conflict 0.23*** (0.17,0.28)0.23*** (0.17,0.28)– Academic 0.14*** (0.07,0.20)0.14*** (0.07,0.20)– Depression General overload 0.19*** (0.13,0.25)0.17*** (0.11,0.23)0.01*(0.00,0.03) Pandemic −0.04*** (−0.06,−0.02)–−0.04*** (−0.06,−0.02) Social distancing 0.15*** (0.08,0.21)0.15*** (0.09,0.21)−0.00 (−0.02,0.01) Interpersonal conflict 0.23** (0.18,0.29)0.23** (0.18,0.29)– Academic 0.16*** (0.10,0.22)0.13*** (0.08,0.19)0.02** (0.01,0.03) Irritability General overload 0.12** (0.06,0.18)0.12** (0.06,0.18)– Pandemic −0.01*** (−0.02,−0.00)–−0.01*** (−0.02,−0.00) Social distancing 0.12*** (0.06,0.18)0.11** (0.05,0.17)0.01** (0.00,0.02) Interpersonal conflict 0.35*** (0.29,0.40)0.35*** (0.29,0.40)– Academic 0.09*(0.03,0.15)0.09*(0.03,0.15)– Self-perceived change General overload −0.14** (−0.20,−0.07)−0.12** (−0.18,−0.05)−0.01** (−0.03,−0.00) Pandemic 0.02*(0.00,0.04)– 0.02*(0.00,0.04) Social distancing −0.23*** (−0.30,−0.17)−0.20*** (−0.27,−0.14)−0.02*(−0.04,−0.00) Interpersonal conflict −0.13** (−0.19,−0.07)−0.13** (−0.19,−0.07)– Academic −0.18*** (−0.25,−0.17)−0.16*** (−0.23,−0.10)−0.01** (−0.02,−0.00) *p<0.05, **p<0.01, ***p<0.001. DISCUSSION A growing number of studies have addressed the psychological effects of the COVID-19 crisis, but very little is known about the impact on university students, even if students might be expected to be greatly affected by the pandemic conditions and by the policies implemented to curtail the spread of the disease. Emergent adulthood is itself a developmental time and is wellknown to be vulnerable to psychological difficulties (Schulenberg et al., 2004); in particular, high rates of mental problems have previously been reported in student populations (Auerbach et al., 2018). The COVID-19 crisis led to the closure of universities and forced students not only to change their general life conditions, but also to substantially adjust their daily academic work, long-term projects, and their expectations. Hence, the current study examined the psychological health of university students during the Spanish quarantine, considering the specific sources of stress perceived by these students during the crisis, as well as the strategies reported to have been used to cope with the situation of COVID-19. Mental Health During the COVID-19 Crisis Several indicators of mental health were analyzed in this study. On the one hand, we included measures of common psychological problems (anxiety and depression), using standardized measures widely employed in previous research for prevalence purposes (Quon et al., 2015), and also in recent research looking at the COVID-19 crisis (Zhu et al., 2020). On the other hand, we also included a measure of irritability (proneness to anger, annoyance, frustration, and aggressive Frontiers in Psychology | www.frontiersin.org 9January 2021 | Volume 12 | Article 589927