Psychological Responses and Strategies Towards the COVID-19 Pandemic Among Higher Education Students in Portugal and Switzerland: A Mixed-Methods Study

Background The COVID-19 pandemic has caused overwhelming changes in individual and community daily-life, resulting from the public health measures implemented to contain it, and also from its psychological and socio-economic consequences. These shifts and consequences impacted the entire population, but some groups are more likely to be affected by these changes, including higher education students. Objectives a) to investigate mental health status and its determinants among higher-education students in Portugal and Switzerland; and b) to explore adjustment patterns used by these students to overcome the impact of the COVID-19 pandemic. Methods A cross-sectional study with a mixed-methods sequential explanatory design was conducted in two phases. First, an online survey was conducted among higher education students in Portugal and Switzerland, in Portuguese and French respectively. A convenience sampling method was used. Second, some participants from the first phase were invited to participate in four online focus group discussions (two in each country) using a maximum variation sampling method. Results The survey was answered by 1,880 students. Portuguese students revealed higher levels of stress and anxiety, but lower depression symptoms and less resilient coping compared to Swiss respondents. Hope was identified as an explanatory variable for mental health symptoms in students from both countries. In the focus groups (n = 27), 13 adjustment strategies were found, which were subdivided into three spheres: personal, social, and contextual. Conclusions The results suggest that the COVID-19 pandemic had a mild to moderate impact on most of the evaluated mental health variables. Nevertheless, the students reacted and mobilized positive short-term strategies, which need to be reinforced in order to prevent long-term psychological harm. In addition, our results can inform psychosocial interventions to minimize psychological impact, anxiety, depression, and stress due to sanitary crises or other population-wide problems or disasters.


INTRODUCTION
COVID-19 has brought physical, psychological, and social hardship to populations around the world. Arguably, the pandemic represents a "perfect storm" for mental health due to its prolonged and unpredictable character, based on an unfamiliar and invisible danger, and its real threat to lives and livelihoods (1).
Among the "at-risk" population, higher-education students have the dual characteristic of being vulnerable to stress and anxiety, and of potentially being able to embrace new technologies or new situations more readily than other groups. The former is evident from studies that pre-date COVID-19. Before the pandemic, "many students across the globe experienced high levels of anxiety, depressive moods, lack of self-esteem, psychosomatic problems, substance abuse, and suicidality" (2). A recent meta-analysis revealed that the prevalence of depression (39%) or anxiety (36%) among college students greatly increased during the COVID-19 pandemic (3). Another systematic review reported a general prevalence of anxiety in Europe as high as 51%, above that in Asia (33%), but lower than in the USA (56%) (4). In various studies, female gender, social isolation, student status, and low quality of social relations were identified as risk factors for lower mental health (5)(6)(7).
Although several studies have focused on the mental health of higher education students, there is a scarcity of studies documenting a holistic portrayal of associations between mental health and protective variables, such as hope and resilient coping. Students seem vulnerable to the negative effects of the pandemic and few studies find significant signs of resilience, such as the use of effective coping strategies (6). Moreover, no study used mixed methods to gain a better understanding of how COVID-19 affects mental health and what coping strategies students may employ in response.
The transactional model of stress by Lazarus and Folkman (8) served as the theoretical underpinning of the current study. It posits mutual interactions between people and their environment and suggests that the stress response is highly influenced by individual appraisal processes (9). This rationale underlies the interplay between physical, psychological, cultural, and social elements and their relationships with mental health, psychological well-being, and social functioning (10). As situations vary between countries, an investigation of the mental health and coping abilities of students under COVID-19 seemed of particular interest, if spanning culturally different settings. This study focuses on students in two middle-sized European countries, Portugal and Switzerland, which in December 2021 had a similar number of registered cases of COVID-19: around 1.3 million, for total populations of 10.3 and 8.6 million inhabitants, respectively (11).
The pragmatic aims of this study were: a) to investigate mental health status and its determinants among highereducation students in Portugal and Switzerland; and b) to explore adjustment patterns used by students to overcome the impact of the COVID-19 pandemic.

METHODS
The study followed a mixed-methods approach with a crosssectional explanatory sequential design, beginning with a quantitative phase that supplied context and participants for the subsequent qualitative phase (see Figure 1). Afterwards, the findings of the quantitative and qualitative phases of the study were combined (12).
The two parts of this international study were as follows: a) an online survey (quantitative) assessing the mental health status and psychological responses of higher-education students during the COVID-19 pandemic; b) a Focus Group study (qualitative) discussing the strategies adopted by students to face the COVID-19 pandemic.
Ethical approval to conduct this study was obtained from the Ethics committee in Portugal (CE/IPLEIRIA/22/2020) and Switzerland (swissethics/CER-VD-2020-02889). Informed consent was obtained from all individual participants included in the study.

Data Collection
This cross-sectional study involved students from Portuguese and Swiss higher-education institutions. Adults enrolled in education levels above high school, including undergraduate and graduate programs, were considered eligible. The age composition of students in this type of programs is very heterogenous, reflecting actual aging societies and lifelong learning concept. For that reason, no age limit was imposed. Students participating in Erasmus or other mobility programs were excluded. Participants were recruited using convenience sampling. Email contacts of interested participants were recorded in phase I, so they might be invited to participate in the qualitative phase. The total target population of Portuguese and Swiss higher-education students (4 Portuguese institutions and 4 Swiss institutions) was estimated at 61.000 individuals (13). The minimum sample size (n = 385) was calculated considering the most conservative scenario (a proportion of 50%), a level of confidence of 95%, and an error margin of 5%. Data were collected between April 2020 and June 2021, via an online survey, sent by the institutions who participated in the study, including: a) Sociodemographic and health information: gender, age; marital status; level of studies; study area; employment (if applicable); household size; perceived health status (using a scale ranging from 0 [worst] to 100 [best] respondents rated their own physical/mental and overall health); and history of chronic disease. b) The Depression, Anxiety and Stress Scale-21 (DASS-21) (14), in Portuguese (15) and French (16), to assess psychological distress experienced in the preceding week, using recommended cut-off points (17). Internal consistency (Cronbach's alpha) in this study for depression, anxiety, and stress was 0.92, 0.90, and 0.92, respectively, indicating very good reliability. c) The Herth Hope Index (HHI), with 12 items designed to measure hope in adults, in Portuguese (18) and French (19).
Internal consistency (Cronbach's alpha) in this study was 0.97, indicating excellent reliability. d) The Brief Resilient Coping Scale (BRCS), a 4-point measure to assess an individual's tendency to cope adaptively (20). The Portuguese version was validated by Pais-Ribeiro and Morais (21), and the French version was translated by Ionescu (22). Cronbach's alpha for this study was 0.79. e) The Impact of Event Scale-Revised (IES-R) (23), with 22 items describing distress experienced in response to traumatic events. This reliable and valid measure was used in Portuguese (24) and French (25). Internal consistency in the present study was excellent (Cronbach's α = 0.95).

Statistical Analysis
Data were analyzed using SPSS Version 28. Descriptive data analysis was performed. Associations and/or differences between variables were determined with univariate and multivariate analyses. A linear regression model was used to evaluate the predictive capabilities of the independent variables with respect to the corresponding decision variables (anxiety, stress, and depression). A 0.05 level of significance was used.

Data Collection
Phase II of this study was qualitative research to obtain additional insights into the online survey responses by carrying out Focus Group Discussions (FGDs). Before conducting the FGDs, the authors developed questions based on previous research (13,26) enabling the acquisition of information regarding the strategies adopted by students to promote better psychological adjustment during the pandemic crisis (13). Participants who completed the anonymous online survey had the option to provide their contact information (e.g., email) if they were interested in participating in the FGDs. Eligible students were selected for the FGDs using a maximum-variation sampling method, creating a heterogeneous purposive sample. Four FGDs (two in each country) were conducted using Microsoft Teams and transcribed in full. In each country, the same non-directive moderating style was used to promote interactions between participants. Each FGD took approximately 60-90 min.

Data Analysis
The data were coded and inductively analyzed using a thematic analysis approach (27). WebQDA software was used for qualitative data management (28). The research teams from each country examined qualitative data separately. Swiss and Portuguese data were then merged and analyzed together for common themes and patterns. Themes were identified inductively by two researchers from each country and discussed with the entire research team until consensus was reached. Verbatim quotes from participants were translated from the original into English by a bilingual researcher and used as examples when presenting results. Themes and subthemes were represented using coding trees to illustrate conceptual relationships between the emerging codes. Example extracts from participants were numbered according to FG (1)(2)(3)(4), with nationality (PT/CH) in brackets.

Validity
The validity of a mixed-method study depends on both the quantitative and qualitative phases. Potential validity threats were minimized by using a large sample size in the quantitative phase, by selecting maximal variety (based on results obtained in phase I) between the participants selected for the FGDs, and by focusing equally on the validity of the two phases (12). To ensure trustworthiness, the researchers were deeply involved in data handling and maintaining rigor during data analysis. The research team adopted a pragmatic stance, integrating the rationale of different methodological approaches (12), and including nurses involved in academic teaching and with practical nursing expertise (AQ, MAD, CL, FSM, TCN, and ZC), as well as a psychologist (DA) and a public health researcher (DC).

Sociodemographic and Health-Related Characteristics of the Study Sample
Of the 1,880 participants in the survey, 1,522 were from Portugal and 358 from Switzerland (see Table 1). The mean age of participants was 23 years (SD = 6.5) (range 18-59).
Regarding the level of studies, 81.3% (n = 1,528) were currently undergraduate students (Bachelor's degree), 38.4% in the field of health area, and 27.6% (n = 518) declared being employed. The majority declared no chronic illness, but 13.4% (Switzerland) and 26.3% (Portugal) indicated they lived with elderly or chronically ill people.
Regarding sociodemographic and health-related variables, statistically significant differences were found between the participants from both countries of origin, except for marital status (p = 0.123). A comparison of perceived health status, psychological impact, hope, and coping between the two countries is shown in Table 2. Portuguese respondents reported higher levels of perceived health status compared to Swiss students. However, the severity of psychological impact appeared to be worse in Portuguese students compared to Swiss students (p < 0.001). The overall sample had a mean depression score of 7.0 ± 5.6, mean anxiety of 5.2 ± 5.3, and mean stress of 8.2 ± 5.6. Students from Portugal reported higher levels of stress and anxiety, but lower depression levels compared to Swiss respondents (p < 0.001; with negligible Cohen's effect size).
The IES-R global mean score was 31.7 ± 19.3 (range, 0-88). Statistically significant differences between the two groups emerged for this scale, with the Portuguese sample reporting higher psychological impact compared with the Swiss sample (p < 0.001; with a small Cohen's effect size).

Prevalence and Severity of Mental Health Symptoms and Psychological Impact
The results of the IES-R were as follows (see Table 3): no impact in 37.6%, mild psychological impact in 16.3%, moderate in 5.6%, and severe in 40.5%. Compared to the Swiss sample, Portuguese students had significantly higher scores on the IES-R scale with p-values < 0.001.
According to DASS-21, students with clinically significant anxiety were 44% of the total sample. The sum of clinically significant stress was 34.1%. Those with clinically significant depression were 31.7%. Among Swiss participants, respectively 31.2, 21.6, and 45.6% had symptoms of stress, anxiety, and depression above the normal range. More than half of the Portuguese students (64.3%) had normal scores on the stress  subscale, but 29.7% were in the mild range, and 6% could be classified as moderate. Regarding anxiety, 63.8% of the total sample had symptoms in the normal range; 12.5% were mild; 15.4% were moderate; 6.6% were severe; and 1.6% were extremely severe. A normal level of depressive symptoms was present in 71.5%, mild in 11.4%, moderate in 16.2%, and severe in 1% of the total sample. As indicated in Table 3, in almost all the variables under study (except for stress), the COVID-19 pandemic had a more negative impact on Portuguese students than on Swiss students; these differences were statistically significant (p < 0.001).

Factors Affecting Depression, Anxiety, and Stress Among Students
As shown in Table 4, the perception of mental health state and hope were negatively correlated with stress in both samples. Similarly, perception of mental health state and hope were negatively correlated with depression. In Portuguese students, resilience (β = 0.161, p < 0.001) played a protective role relative to depression. Anxiety, perception of mental health (β = −0.98, p < 0.001), and hope (β = −0.182, p < 0.001) were negatively correlated with depression in the Portuguese sample. In Swiss students, hope and overall state of health were negatively Frontiers in Psychiatry | www.frontiersin.org associated with anxiety, while resilience (β = 0.166, p = 0.035) was a positive predictor of good mental health. Hope was found to be a significant explanatory variable for all mental health symptoms, in students of both countries.
In the Swiss sample, the model showed a predictive capacity ranging from 25.1% for anxiety to 49% for depression symptoms. In the Portuguese sample, explained variance ranged from 25.4% for anxiety to 44.2% for depression. Both models proved satisfactory (29) presenting F values with high statistical significance (p < 0.001); and all t-test values were significant at p<0.001.

Qualitative Phase
The four FGDs relied on a total of 27 participants (Switzerland: 15 participants for 2 FGDs; Portugal: 12 participants for 2 FGDs). Using thematic analysis, 13 subthemes were extracted and classified into three main themes -or spheresconcerning the adjustment strategies adopted by students (see Table 5).
The first theme was related to the personal sphere, where students developed the most adjustment strategies in both countries. They referred to an opportunity for personal growth and adaptability. Portuguese students talked about a "window of opportunity" to be involved in academic and professional projects, although students in Switzerland did not mention this point. Focusing on work-life balance was a strategy mentioned in both countries, mostly by organizing and separating work from private life. Self-care through the adoption of healthy lifestyles was also important and was mostly related to physical activity and/or eating habits. Being compassionate with others emerged as a strategy that helped put things into perspective and remain aware of one's role in helping and protecting others, either directly through volunteer work or indirectly by staying at home and avoiding contacts. Students reported feeling overwhelmed by the quantity of information from different media and felt the need to select the most useful sources. Compliance with sanitary measures was an essential strategy that emerged in both countries with some differences, especially concerning the different waves of the pandemic.
In the social sphere, students evoked new ways of communicating via digital networking to compensate for the lack of physical proximity and stay in safe contact with friends and relatives. Portuguese students mentioned new ways of being physically close, such as communicating through a window or from a balcony, while students in Switzerland found alternative ways of socializing by going for physically distanced walks with friends. A strategy that was greatly discussed in all FGs was the importance of being with family and friends. Students felt more aware of the importance of cultivating these relationships and giving value to time passed with loved ones.
In the contextual sphere, students of both countries talked about the importance of adapting the "teaching/learning" environment. They found themselves in a new situation and had to adapt almost constantly to new ways of studying and learning online. Even in difficult situations, advantages were perceived by some students, such as not having to commute to the university. Tailored teaching support was a significant strategy, especially in maintaining their motivation. Portuguese students brought up that school support services helped them deal with certain difficulties, but this was not mentioned by students in Switzerland.

DISCUSSION
This study set out to investigate the psychological and mental health responses of higher education students during the COVID-19 pandemic, and the coping strategies patterns they used to manage or deal with the pandemic. We found that 31.7, 44.0, and 34.1% of the participants had depression, anxiety, and stress symptoms, respectively. These rates were lower than in previous studies. For example, a meta-analysis (30) found a higher prevalence of anxiety and depression in pandemicaffected college students. In Switzerland, a survey conducted in April-May 2020 found that 85.8% of surveyed undergraduate students reported symptoms of anxiety, although in most cases the symptoms were mild (63.3%) (31). Another study, with 1,075 Portuguese and Spanish undergraduate students in April-May 2020, revealed high levels of perceived stress (51.9%) (32). These differences may be related to when data was collected. Higher levels of depression in Switzerland and of anxiety and stress in Portugal suggest timing of data collection is relevant when investigating the prevalence of mental health symptoms. Recent findings seem to indicate that feelings of overwhelming, stress and anxiety were higher in the early months of the pandemic (33,34). This evidence confirms the relevance of our results from students in Portugal, where data collection started 1 month after the outbreak of the pandemic. However, depressive symptoms, which include feelings of sadness, loss of interest and pleasure in activities, as well as disruptions of daily life, may remain high or even increase during a pandemic due to social isolation and physical distancing (35). This may explain why the Swiss students in our study had higher depressive symptoms: their data collection began 12 months after the start of pandemic.
Our results show that the COVID-19 pandemic had a worse impact on Portuguese than Swiss students. We hypothesize that this may be related to cultural differences. The Portuguese population is predominantly Catholic, with a close-knit family ethic, and experienced the pandemic as markedly disruptive, breaking family ties due to social restrictions, and highlighting rooted fatalistic values (36). Although Switzerland generally scores highly on individualism, Swiss society is also characterized by adherence to rules -a characteristic linked to federalism (a defining attribute of the Swiss political system) and a collective mind-set (37). In such a cultural setting, individuals may attribute high value to the group, thus reducing the personal negative impact of external adverse events. Finally, trust toward scientists is higher in Portugal than in Switzerland (37); such trust may not have been an advantage in the initial phases of a pandemic that caught most scientists by surprise.
Resilience and hope feature prominently in our study, but to our knowledge they have rarely been used when investigating student reactions to the COVID-19 pandemic. Our research stresses the protective role of hope in the mental health status of students. Similar results have highlighted the protective benefits of hope on cognitive abilities in uncertain times (38,39) and that hope can be a predictor of active coping styles adopted by students (40). Nevertheless, most of our participants had low levels of resilient coping. A possible reason is that the pandemic led to increases in anxiety and depression, but also undermined personal resilience (41).
Our qualitative findings in both countries showed that students developed adjustment strategies mostly in the personal sphere. Confinement and isolation may have made students more introspective, leading to solitude and, in turn, to the development of such strategies. As shown by Höglinger and Heiniger (42), feelings of solitude have been reported as being difficult to manage through the different pandemic waves. The adjustment strategies reported in our study are consistent with coping strategies reported by authors such as Shanahan et al. (43) and Burton-Jeangros et al. (26), who found that frequent use of strategies like keeping a daily routine, reappraisal/reframing, engaging in hobbies and activities and staying connected with others were associated with reduced distress (26,43). Such approaches are needed to boost resilience factors protecting the individual against psychological distress (6). it's mainly the key words of adaptation, change, fear and awareness of our environment, but also of ourselves. And then from my point of view, it also allowed us to take time for ourselves..." "Window of opportunity" to be involved in academic and professional projects FGD1 (PT) "[...] it was also a window of opportunity, also because I managed to get involved in other projects that I would never achieve in terms of geographical distance." Focus on Work-life balance FGD1 (PT) "Structuring a timetable is very important […]. Sometimes we lose a little bit of the notion that we're working, the brain is working and it's tiring and that we're not taking any time for ourselves." Compliance with the sanitary measures FGD2 (CH) "I think that during the first wave there were a lot more people who paid attention, who were more inclined to respect the rules [  Students reported using social media and technology to maintain contact with family members and find information, an important coping strategy to deal with the pandemic lockdown found on other studies (34,42,44). This fits well within the concept of digital natives (45): many of today's students feel comfortable with technology from an early age and consider it an integral and necessary part of their lives. Our study supports this concept, especially the positive aspects of social media exposition, such as social connectedness, normalization of behaviors, compliance with government directives and access to support groups. Unsurprisingly, students mainly had a positive attitude toward online teaching and learning activities (46). Nevertheless, some scholars (45,47) point out negative consequences such as loneliness, psychological distress, sleep deprivation, emotional anxiety, the spread of misinformation, and difficulty to discern facts from fiction.
The findings of this study suggest that the COVID-19 pandemic had a mild to moderate impact concerning most of the evaluated mental health variables. Nevertheless, the students reacted and mobilized positive short-term strategies that should be reinforced in order to protect them from long-term psychological harm. Furthermore, these strategies help students to stay mentally healthy and mitigate delayed-onset posttraumatic stress disorder during the COVID-19 pandemic (48).

Implications for Practice, Research, and Policy
At a practical level, we suggest that higher education institutions reorganize teaching methods that provide greater consideration for student psychosocial functioning. Goal 3 of the 2030 Agenda for Sustainable Development (49) underlines the importance of new learning methodologies to promote self-care and personal knowledge for good mental health. Other recommendations include increasing social support (e.g., peer support system and professional mediation), which have the potential to alleviate the mental health burden. We also suggest designing or tailoring university-based interventions (e.g. psychoeducation, emotional self-regulation, positive mental health promotion) to address student biopsychosocial needs (50).
At a research level, it is necessary to anticipate future crises and prepare for the transition to the post-COVID-19 world. We need to understand the implications of pandemics on the development of the academic, professional, and personal paths of students. Generally, our understanding of student health must integrate knowledge of their pre-existing health conditions. Future research would benefit from a prospective study with a comparison group that would help understand the longer-term impact of epidemics on the development of psychopathological symptoms. Also, studies investigating longitudinal data on risk and protective factors (e.g., substance use, coping strategies, satisfaction with life, social support, family and peer relationships dynamics) would be helpful.
At a policy level, we believe countries should introduce new forms of mental health support, including information materials (mostly online) and mental health support phone lines. It may be worthwhile to shift some mental health services to a telemedicine format and increase investment in mental health (51). Moreover, the COVID-19 has provided opportunities to build on positive innovations: the flexibility of community-based care, digital healthcare, and the connection of physical and mental health (52).

Strengths and Limitations
Our results provide a theoretical basis for a broad audience of educators and researchers and may add to a better understanding of the impact of COVID-19 on higher education. Another strength is the mixed-methods research design and the use of valid and reliable instruments. The multicentric online survey provided broad ecological validity, while the focus groups brought explorative depth and consistency, and raised factors that could potentially mitigate some of the negative impacts. Despite these strengths, this study has some limitations. First, data collection did not occur at the same time in each country, because each was submitted to separate ethics committees that did not operate under the same rules or at the same speed. Second, there is an absence of information on preexisting mental health conditions, which we suggest should be included in future research on the topic. Third, the use of nonprobability sampling procedures and self-reporting measures limit the representativeness of the study. Fourth, the sample lacks homogeneous criteria since most of the participants were female with Portuguese nationality. Lastly, our conclusions are based on cross-sectional data, so we are not able to report on changes over time and establish causality.

CONCLUSION
Our findings support the idea that the COVID-19 pandemic has a significant impact on the mental health of higher education students. In the future, it would be desirable to focus on positive ways of coping and the potential of hope in this regard. Based on our results, positive coping and hope should be integrated into the standard training of students across all study areas. Indeed, a greater focus on protective factors may be a promising way to prepare students -and others -for future hardships, at individual, community, and societal levels.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Ethics Committee in Portugal (CE/IPLEIRIA/22/2020) and Switzerland (swissethics/CER-VD-2020-02889). The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
The study was conceived and designed by AQ, MD, and CL in Portugal and by FS-M and TC-N in Switzerland. Jointly, all five set up an international collaboration, completed the study plan and obtained funding. Data collection and data analysis were carried out by DA, FS-M, and TC-N in Switzerland and by AQ, CL, MD, and ZC in Portugal. DC and CL drafted the first version of the manuscript. All authors contributed to the article and approved the submitted version.