Participants and procedures

Cross-sectional study carried out with undergraduate students from 33 in-class courses at the Western Parana State University (UNIOESTE) from 5 campuses (Cascavel, Foz do Iguaçu, Francisco Beltrão, Marechal Candido Rondon and Toledo). Data collection occurred from September 14 to October 19, 2020.

The non-probabilistic convenience sampling was adopted, and the inclusion criterion was that the undergraduate must be enrolled in a classroom course at UNIOESTE and be 18 years old or older. Invitations to participate in the study were sent to undergraduate students enrolled in 2020, through the e-mails registered at the university (institutional and personal) and the dissemination of the study’s link in students’ digital media and on the university website. From the total of 8,977 enrolled undergraduates, 1,579 students responded the online research instruments and fulfilled the inclusion criteria, and 355 were excluded because did not answer the 4 scales (158 answered only the questionnaire, 19 answered only 1 scale, 123 answered only 2 scales, 55 answered only 3 scales), resulting in the final sample of 1,224 undergraduate students (13.6% of those enrolled), that responded all the questions of the applied instruments. Thus, there was no missing data to be computed.

Measures

Data collection was performed digitally, through a link to access the online instruments on the Surveymonkey^® platform, containing a questionnaire prepared for the study and 4 self-applied scales.

Statistical analysis

Descriptive and inferential statistics were performed using the Minitab^® statistical software version 19.2020.1, considering a significance level of α<0.05. Microsoft Excel^® software was used for tabulation and coding the data. The reliability and consistency were verified by the Cronbach’s Alpha coefficient, classified as: weak (α<0.6), moderate (0.6≤α<0.7), good (0.7≤α<0.8), very good (0.8≤α<0.9) or excellent (α≥0.9) [60].

All data from scales and subscales presented non-normal distribution in the Kolmogorov-Smirnov normality test (p-value<0.010) and the Anderson-Darling normality test (p-value<0.005). The correlations between the scores on the scales were verified using the non-parametric Spearman’s correlation coefficient, and the following intervals were classified as significant: weak (0.1–0.3), moderate (0.4–0.6), strong (0.7–0.9) [61].

Univariate logistic regression was performed for the presence of symptoms of depression, anxiety and stress (binary dependent variables: normal versus symptomatic) in relation to the independent variables: 23 categorical variables (dichotomized) from the questionnaire and 21 continuous variables (SWLS, 6 PWB subscales and 14 BriefCOPE subscales). The independent variables that obtained p≤0.20 in the univariate logistic regression were included for testing in the multivariate model.

The predictor variables were determined by multivariate logistic regression (forward method, logit function, 95% confidence interval, α<0.05). The regression model’s adjustment analysis was based on the Hosmer-Lemeshow test (p>0.05 indicating that the model present good fit), and the model’s sensitivity and specificity by the analysis of the area under the ROC (Receiver Operating Characteristic) curve, considering the discrimination as: acceptable 0.7≤ROC<0.8; excellent 0.8≤ROC<0.9; outstanding ROC≥0.9 [62]. The multicollinearity was verified by VIF (Variance Inflation Factor), considering absence VIF = 1 and acceptable VIF<5 [63].

Ethics procedures

Research Project approved by the Ethics Committee on Human Research of the Western Parana State University (judgement 4,184,430 of July 31^th, 2020, CAAE 35209620.8.0000.0107) and with authorization from the person responsible for the study fields. The research invitation letter clarified the optional nature of participation and the guaranteed anonymity, and those interested gave digital acceptance in the Informed Consent Form before filling out the instruments.

Prevalence of depression, anxiety and stress symptoms

The results indicate that the majority of undergraduates surveyed presented mild to severe symptoms of depression (60.5%), anxiety (52.5%) and stress (57.5%). Compared with other studies during the COVID-19 pandemic, using the same instrument, the percentages of the present study were higher than those found in the Brazilian adult population for depression (21.5%, 46.4% and 54.5%), anxiety (19.4%, 39.7% and 41.1%) and stress (21.5%, 42.2% and 45.7%) [7, 9, 10], and in the population of 8 countries (China, Spain, Italy, Iran, USA, Turkey, Nepal, Denmark) for depression (14.6% to 48.3%) and anxiety (6.3% to 50.9%) [5].

Regarding stress, the result of the surveyed students was higher than those of undergraduates from Bangladesh [28], Egypt [26], Ethiopia [24, 30], Lebanon [25] and Turkey [29], whose percentages ranged from 12.7% to 47.8%. For anxiety, the percentage was lower than those in Egypt (53.6%) and higher than the others (27.7% to 48.6%). For depression, the percentage was lower than those of Egypt and Turkey (70.5% and 64.6%) and higher than others (21.2% to 46.9%). Another study in undergraduates from Bangladesh [27] found higher values than all mentioned studies for depression (76.1%), anxiety (71.5%) and stress (70.1%).

The higher percentages of depression, anxiety and stress symptoms in surveyed undergraduates in relation to the Brazilian population may be related to the fact that the data collection of the present research occurred in a more advanced period of the pandemic in 2020 than the data collection of the cited studies [7, 9, 10].

Most undergraduates reported having had at least one life aspect negatively affected by the COVID-19 pandemic (97.3%), notably the studies (80.6%). During the period of data collection, most respondents had restarted academic activities remotely (79.8%) about 1 or 2 months ago (August 17, 2020), after the interruption of the school year. Thus, the high prevalence of depression, anxiety and stress symptoms also suggests the additional psychological impact of the pandemic on students due to repercussions on education, such as adaptation to remote classes, as assessed in other studies [10, 15, 16, 25].

Prevalence of satisfaction with life and psychological well-being

In this study, undergraduates obtained average scores for satisfaction with life (20.1) lower than that found in technology students in Poland during the pandemic (22.0) [48] and in medical students from the state of Pernambuco in Brazil, before the pandemic (22.3) [40].

Most of the surveyed undergraduates presented average/high scores in the 6 dimensions of the psychological well-being scale (54.6% to 98.4%) and the lowest frequency of average/high scores were observed in environmental mastery dimension (54.6%). A longitudinal study with Spanish university students, which divided the study participations into 3 groups (ordinary week, pre-confinement and confinement in the pandemic), observed a decrease in the environmental mastery between the ordinary week and the pre-confinement [64]. In the context of the COVID-19 pandemic, with restrictions imposed by public health protocols, the low scores for environmental mastery in the surveyed undergraduates may be influenced by the difficulties in choosing and creating appropriate contexts to meet their personal needs as a result of isolation and social distancing.

A study suggests that subjective well-being, involving satisfaction with life and negative and positive affects, tends to be strongly influenced by emotional experiences, which vary over time and context, and psychological well-being tends to be based on more stable abilities [65]. Thus, in the present research, the lower scores for satisfaction with life suggest that these levels are being more affected by the pandemic crisis compared to levels of psychological well-being among undergraduates surveyed.

Use of coping strategies

The frequency of average/high scores for the adaptive coping active coping (82.6%) and planning (90.1%) stands out in the sample, indicating that most of the surveyed undergraduate students tend to think about the best way to deal with the problem when facing a stressful situation and try to take actions to eliminate, minimize or circumvent it.

The surveyed undergraduates had a low frequency of average/high scores (39.0%) regarding the adaptive coping humor and it had no significant correlation with symptoms of depression, anxiety and stress, diverging from another study in the pandemic with university students in France that found negative correlations of humor with anxiety and depression [66].

Correlations between the studied dimensions

The results of the correlation analysis showed that the higher levels of satisfaction with life and the 6 dimensions of psychological well-being, in addition to higher scores of 3 adaptive coping (active coping, planning, positive reframing) may act as protective factors against symptoms of depression, anxiety and stress, as well as higher scores of 5 maladaptive copings (self-distraction, denial, substance use, behavioral disengagement, self-blame) may be risk factors for such symptoms. These results corroborate the data of other studies with students during the COVID-19 pandemic: in China, it was found that adaptive coping may, along with resilience and social support, mitigate the association between pandemic stressful events and stress symptoms [49]; in France, depression symptoms were positively correlated with 4 maladaptive copings (substance use, denial, behavioral disengagement, self-blame) and negatively correlated with the 3 adaptive copings (active coping, planning and positive reframing) [66]; in Poland, in technology students, anxiety was positively correlated with emotion-oriented copings (self-blame, irritation, getting tense, worrying, fantasizing) and negatively correlated with satisfaction with life [48].

Studies prior to the pandemic, using the same instruments, found: in Australian university students, negative correlations between all dimensions of psychological well-being and depression [39]; in the Portuguese population, positive correlations of all dimensions of psychological well-being with the adaptive coping active coping and negative correlations with maladaptive copings self-distraction, denial, substance use and behavioral disengagement [45].

Predictors of depression, anxiety and stress symptoms

In the present study, being female, aged between 18 and 24 years old, having a chronic disease, presenting lower scores in the dimensions of psychological well-being (positive relations with others and self-acceptance), and higher scores of maladaptive copings substance use and self-blame were common predictors of depression, anxiety and stress symptoms in undergraduate students during the COVID-19 pandemic.

Studies in the Brazilian population during the pandemic suggest that being a woman, younger and having a chronic disease were among the predictors of severe anxiety symptoms [10] and were associated with depression and anxiety symptoms [36]. Being female and younger were among the predictors for severe symptoms of depression and stress [10] and were associated with greater severity of symptoms of depression, anxiety and stress [8].

A review study of the population mental health during the pandemic, including 8 countries (China, Spain, Italy, Iran, USA, Turkey, Nepal and Denmark) suggested that being a woman, up to 40 years old and having chronic or psychiatric illness were among the risk factors associated with stress, in addition to other variables such as being a student, being unemployed and frequently exposed to social media and news about COVID-19 [5]. In the Italian population, a study showed that being a woman was among the predictors for depression, anxiety and stress; being young was among the predictors of stress; and having a history of medical problems was among the predictors of depression and anxiety [67]. In the Chinese population, a study found that being a woman, a student, having specific physical symptoms, and evaluating oneself as being in poor health were associated with high levels of depression, anxiety and stress [68].

In Egyptian students, female gender and chronic illness, in addition to having a family member or acquaintance infected with COVID-19 and lack of psychological support, increased the risk of depression, anxiety and stress symptoms [26]. In Ethiopian students, female gender and history of medical illness, in addition to being at home and having little social support, increased the risk of depression symptoms [24].

A review study before the pandemic identified that female gender was among the sociodemographic characteristics most associated with psychological distress in undergraduates [12]. During the pandemic, this variable was associated with symptoms of depression, anxiety and stress in several studies in students [24, 26, 27, 29] and in populations in Brazil [7–10] and other countries [5, 67, 68]. Such a condition points to the need for further studies aiming to identify possible common causes of female susceptibility to stress during a pandemic crisis, regardless of age and culture.

In addition to the vulnerabilities to stress already identified in undergraduates in periods prior the pandemic [12] and the stressors that emerged in the pandemic crisis [16], younger students (18 to 24 years old) tend to suffer greater impact with social distancing, as it hinders the natural movement of this population to seek greater closeness with their peers [15]. Thus, it is possible to think that adults tend to have developed adaptation and coping mechanisms when facing stressful situations, being able to better preserve their own mental health.

The pandemic crisis has impacted the chronically ill population´s mental health [5, 9, 10, 19, 24, 26, 36, 67] and can be explained by the uncertainties involving the risks of worsening and mortality by COVID-19 resulting from pre-existing diseases.

Lower scores for self-acceptance (dissatisfaction with yourself, disappointment with past events, concern with certain personal qualities and desire to be different from what you are [69]) and positive relations with others (few stable and trusting relationships, difficulty in being warm, open and concerned about others [69]) are opposite conditions to some protective factors for psychological distress in students found in the scientific literature, such as self-esteem, talking to friends, communication skills and social engagement [12]. In a pandemic context, restrictions on face-to-face contacts may increase lonely moments, especially in those who tend to isolation and have few satisfying empathic interactions, which can intensify stress and associated disorders in those with difficulties in feeling good about themselves.

The maladaptive coping substance use was a common predictor of depression, anxiety and stress. A study in university students in Ethiopia also found the use of substances, such as tobacco and alcohol, in the last 3 months during the pandemic, among the predictors of anxiety [30]. In the present study, there was a low percentage of substance use (15.5%), diverging from what was found in a study carried out in the Brazilian population, in which 40.8% reported having increased the use of drugs, medications, tobacco or food during the pandemic [9].

Most of the surveyed students (82.9%) showed average/high use of the coping self-blame (criticizing oneself for responsibility in the situation [43]) and this coping was a predictor of depression, anxiety and stress symptoms. Self-blame often occurs after a stressful event that can generate negative outcomes, to which the responsibility, causality and/or intentionality is attributed to oneself, originated frequently from a distorted cognition [70]. It is possible to think that the pandemic crisis may have intensified the tendency to blame oneself for adverse events in those predisposed to use such maladaptive coping. A study in United States undergraduates, before the pandemic, found that a high level of so-called proactive coping strategies (stressors perceived as potential challenges rather than threats) were predictors of low levels of self-blame and stress, and the authors suggest the implementation of actions to assist undergraduates in using proactive coping to deal with stressful situations [70].

In the present study, stress and anxiety symptoms had as a common predictor lower scores of environmental mastery, which can be understood as difficulty in managing daily issues and inability to see opportunities to change or improve the context [56, 71]. It can be thought that the pandemic crisis and the resulting restrictive measures tend to make actions impossible and restrict opportunities.

Another common predictor of stress and anxiety symptoms was higher scores of the coping venting, which can be understood as the exposure or the communication of emotional stress to others [59], being consistent with the evasion characteristic attributed to this coping, as keeping the focus on annoyance distracts oneself from the efforts required to eliminate or reduce stress [59].

Symptoms of depression and anxiety had as predictors cessation or reduction of family income during the pandemic. The result is consistent with other studies in the Brazilian population that found the decrease in family income among the predictors of severe symptoms of depression [10] and associated with higher percentages of depression and anxiety symptoms [36]. A review study of previous pandemics points out that financial losses are problems during and after the quarantine period and are risk factors for psychological disorders symptoms [2], which warns that the pandemic economic effects tend to last and be a risk factor for psychological disorders for students and the population.

Another predictor of depression and anxiety was higher scores of coping denial (refusal to believe that the stressor exists or trying to act as if the stressor is not real [59]), which reinforces the argument that denial can create additional problems because it allows the stressful event to become more serious, making it more difficult to cope with [59], especially in a pandemic context, in which effective actions are necessary for COVID-19 prevention, as well as to overcome the possible educational, social and economic difficulties resulting from the pandemic crisis.

In addition, stress symptoms had as predictors the condition of having paralyzed or reduced leisure activities during the pandemic and higher scores of personal growth (sense of continuous development with the personal potentials realization) [56, 71].

Reduction of leisure was among the predictors of severe symptoms of depression, anxiety and stress in a study in the Brazilian population [10]. In Bangladeshi students, limited or lack of recreational activities were associated with stress, anxiety and depression [28]. These results suggest the importance of relaxation produced by leisure in preserving mental health. A pre-pandemic study in Chinese students suggested that effectively reducing academic stress and engaging in leisure activities are both important in promoting and improving emotional well-being [72].

In relation to the higher personal growth scores, it is possible to think that undergraduates who value continuous growth may have had an increase in stress levels due to the impossibilities of the pandemic moment, especially in relation to studies and professional training.

The predictors of depressive symptoms were not working in addition to doing internship, lower scores of satisfaction with life and the coping religion and higher scores of active coping and behavioral disengagement.

In this study, 49.6% of the surveyed undergraduates who do not work reported having ceased or reduced family income during the pandemic and of these, 90.7% do not live alone, suggesting loss of employment or need for help in the family, which may explain the favoring of depressive symptoms, because unemployment was found among the predictors of depression in pandemic studies in several countries [5].

The presence of higher scores of active coping among the predictors of depression can be analyzed based on the questioning of Carver, Scheier and Weintraub (1989) about what would happen to a person who prefers to use active coping when the situation requires restriction or moderation [59]. The results suggest that the restrictions imposed by the pandemic crisis, which make it impossible or difficult to take actions to remove, circumvent or lessen the effects of the stressor, could cause depressive symptoms, such as loss of incentive and low perceived probability of achieving meaningful goals [52, 53], in those who tend to use active coping as a stress coping strategy.

Lower scores of satisfaction with life (poorer personal judgment about how satisfied someone is with your current life, regardless of objective assessments about quality of life) [54], higher scores for the coping behavioral disengagement (giving up goals related to the stressor element) [43, 59] and lower scores of the coping religion (low or no participation in religious activities) [43, 59] could favor the emergence of the following depressive symptoms: devaluation of life; lack of interest and involvement; hopelessness [52, 53]. Not practicing religion was one of the predictors of severe depression symptoms in the Brazilian population [10].

The results of the present study showed that satisfaction with life, dimensions of psychological well-being and coping strategies were associated and/or predictors of symptoms of depression, anxiety and/or stress during the pandemic crisis.

In general, the levels of personal growth and purpose in life tend to be higher in younger people, tending to decrease with age when there are no opportunities to engage in meaningful activities, and the maintenance of high levels throughout life have been associated with health benefits [37]. The majority of the surveyed undergraduates, composed mostly of young people aged 18 to 24 years old, presented high personal growth scores (98.4%) and purpose in life (80.2%), which raises questions about how to help them conserve such levels.

The psychological well-being and satisfaction with life, even though related to the personality and considered relatively stables, despite short-term fluctuations resulting from transitory events, the personality traits may change after specific experiences and interventions [73]. Similarly, coping strategies can be learned responses, initiated voluntarily and, with repetition, become involuntary [43]. Therefore, actions may be proposed to help undergraduates to obtain higher scores of adaptive coping and dimensions of psychological well-being.

Study limitations

The convenience sampling does not allow generalizing the results to the totality of Brazilian university students or the researched university. However, since the sample includes students from 5 campuses (in different municipalities) located in the western region of southern Brazil and from 33 undergraduate courses at the university, it can be considered broad and representative.

A cross-sectional study does not allow causality inferences and to evaluate the long-term impacts of the pandemic crisis, however, it presents relevant evidence that can foster future longitudinal studies considering the relations of the levels of depression, anxiety and stress symptoms with life satisfaction, psychological well-being and coping strategies.

The use of self-report measures is also a limitation of the present study. However, the fact that the four scales applied in this research were international and the versions used had been validated in Brazil may minimize this limitation. In addition, the calculation of the Cronbach’s alpha coefficient indicated good or very good reliability and consistency of the instruments used.