On January 30, 2020, the World Health Organization (WHO) announced the appearance of an illness derived from a novel coronavirus (COVID-19), declaring the outbreak of infection as a Public Health Emergency of International Importance and projecting to the world the highest level of warning (World Health Organization, 2020). In order to try to avoid the spread of infection, historically, lockdown has been used as a preventative measure that reduces the risk of contagion. This exceptional measure was previously imposed during epidemics such as the Severe Acute Respiratory Syndrome (SARS) of 2003 in parts of China and Canada, and during the Ebola outbreak that overran entire villages in Africa in 2014. In Spain, the COVID-19 lockdown was imposed during a period of at least 50 days across the majority of the territory. In Spain, a state of emergency was decreed to establish a COVID-19 lockdown that entailed the domestic confinement of the population during a period of 50 days in the majority of the territory.

Lockdown involves social isolation that can have a negative impact at the psychological level, sometimes extending to months or years afterwards (Rubin & Wessely, 2020). In the context of situation, there have recently been several reviews of work related to the psychological impact of lockdown during the principal infectious disease outbreaks of the 21st century (Brooks et al., 2020; Molero et al., 2020). In their results, the majority of the studies reviewed were shown to have reported on general psychological symptoms, emotional alterations, distress, irritability, insomnia, lack of concentration, etc., including symptoms of posttraumatic stress. In the case of respiratory syndrome, studies indicate that the effect of lockdown was a predictor of depression and posttraumatic stress symptoms in healthcare professionals, continuing for as long as 3 years afterwards (Liu et al., 2012).

In addition to the situation of confinement, we must take into account the high infectious potential of COVID-19, the lack of preparation for prevention and treatment of the disease, as well as its mortality rate, all aspects that may cause this situation to leave deep psychological scars in many people. The study carried out in China by Horesh & Brown (2020) argues that this situation can and should be viewed from the perspective of trauma. However, in this situation there are several unique characteristics: on the one hand, in contrast with other traumatic events, the emergency situation and obligatory social isolation have been prolonged in duration, constituting a continuous trauma for which there was no subsequent period of safety; this differs from what happens with respect to other types of emergency, such as earthquakes and floods, which consist of a single, isolated traumatic event followed by a period of relative calm and security (Jarero & Artigas, 2018). On the other hand, this is a new type of mass trauma whose scale is global and whose impact extends to all aspects of society: the economy, work, education, social relations, way of life, leisure, etc. (Pan, 2020). Owing to all of this, we can consider a psychological impact upon the population to be highly probable and of high intensity.

In China, one of the first studies directed at understanding the psychological impact of COVID-19 (Wang et al., 2020) included a survey of 1,210 people, in which 53% evaluated the impact of the situation as moderate-serious, 16% reported moderate to serious depressive symptoms, 28% moderate or serious symptoms of anxiety, and 8% reported moderate or serious levels of stress. In Spain, various studies have also sought to examine such effects. Ozamiz-Etxebarria et al. (2020) carried out an investigation into the levels of stress, anxiety, and depression in the first phase of the outbreak and detected higher averages following March 14, when the confinement began. In a study comprising the participation of 6,789 people from different Autonomous Communities, Balluerka et al. (2020) analyzed the psychological effects of the pandemic and confinement, noting that 77.5% of the participants had experienced an increase in their level of uncertainty faced with the present situation, 75.5% reported having been afraid of losing a loved one, 67.9% reported worrying about having or contracting a serious illness, whether coronavirus or another, and 52.8% reported having suffered sleep problems. The same study found that, in general, participants had experienced some increase in anxiety or panic attacks, and found an inverse relationship by age: among the older age groups, the worsening of psychological discomfort was less severe, while the younger groups (18-25 years of age) of the study population presented higher averages in stress, anxiety, and depression by comparison with the 26-60 years group and the over-60s group. In the same vein, Justo-Alonso et al. (2020), in analyzing the role of age in the early psychological responses to the pandemic, also concluded that the younger population, between 18 and 25, had been more affected, slept worse, presented more claustrophobia and somatizations, and greater difficulties in maintaining routines. In a sample of vulnerable children and adolescents, Vallejo-Slocker et al. (2020) detected worse results in psychological wellbeing than those published for official national reference in Spain in 2017, before the outbreak of COVID-19. These data point to the idea that exposure to trauma is not always sufficient to explain the development of posttraumatic stress (Brewin et al., 2000), and that individual factors, including age, may play an important role in this condition.

Since the pandemic arose, young people have been the focus of studies by several researchers. Liang et al. (2020) evaluated mental health since the COVID-19 outbreak in a cross-sectional study, finding that mental health problems were serious in the majority of young people during the emergency: 40.4% had a tendency to have psychological problems, while 14.4% of young people in the sample showed symptoms of posttraumatic stress. Similarly, the study undertaken by Pan (2020) on the family life of college students amid the virus notes that 82.98% of students reported feeling fear in the face of this emerging public health event. The prevalence of depressive symptoms, anxiety symptoms, and a combination of depressive and anxiety symptoms was 43.7%, 37.4%, and 31.3%, respectively, among Chinese high school students during the COVID-19 outbreak (Zhou et al., 2020). In Mexico, results obtained from a sample of students aged from 10 to more than 30 years (M=21.95) showed, in line with results from Spanish studies, that during confinement the highest levels of anxiety, depression, stress, psychological symptoms, sleeping difficulties, and social dysfunction in daily activity were presented in the youngest groups (18-25 years) (González Jaimes et al., 2020).

A further aspect to consider is that the impact of exposure to trauma is cumulative in nature and that the risk of posttraumatic stress disorder (PTSD) increases in people exposed to multiple stressors (Lowe et al., 2015). In the emergency situation provoked by COVID-19, Brooks et al. (2020) identified, among the studies reviewed, a set of stressors related with mental health during compulsory confinement: first, the duration of lockdown, whereby the longer it extended the more it was associated with symptoms of posttraumatic stress, avoidance behavior, and anger; fear of infection or of infecting family members; frustration and boredom associated with loss of routine and reduction in social contact; the fact of having insufficient supplies or the impossibility of accessing medical care; and lack of knowledge of the situation due to inadequate information.

In the present situation, it is possible to believe that the stressors described by Brooks et al. (2020) in respect of SARS or Ebola have also been present during the lockdown brought about by COVID-19. In particular, we consider it necessary to know the magnitude with which these stressors have been present in the young people that have been confined in order to be able to identify the psychological impact upon this group. In another regard, we do not know with any precision the changes that successive weeks of lockdown might introduce among these stressors, taking into consideration that they can range from the uncertainty of the first week, to exhaustion or relief at the end of confinement in the fourth. And, finally, we consider it especially useful to acquire a knowledge of how these stressors may be affected by the presence or absence of the virus, whether in the participant or in their environment.

In light of the above, the present study has the objective of understanding the psychological effects on college students of the lockdown brought about by COVID-19. More concretely, our objectives are to analyze the following in respect of the PTSD psychological responses: 1) whether the frequency and intensity of the stressors are high and show similarity in their levels; 2) whether the frequency and intensity of stress responses increase as the period of confinement increases; and 3) whether the frequency and intensity of the stress responses are higher in students who have been in contact with the virus, compared to those who have not.

Method

Participants

The study comprised the voluntary participation of 270 students, of whom 89.3% were women and the average age was 21.21 years (DT= 2.951). In regard to the demographic characteristics during the lockdown, 73% lived with their family, 16% with housemates, and the rest were in other situations (with their partner, alone, etc.). The number of inhabitants in the areas in which lockdown was spent ranged from less than 500 (8.5%) to more than 50,000 (46.3%).

Procedure

One month after the beginning of the lockdown, the selected sample responded to a questionnaire (Annex I) on four occasions, once per week, with the last instance coinciding with the end of the confinement. Concerning the construction of the questionnaire, we started, on the one hand, from the stressors identified by Brooks et al. (2020): duration of the lockdown (D), fear of infection (MI), insufficient supplies (SI), inadequate information (I), frustration (F), and boredom (A). On the other hand, we took into account the symptomatology associated with PTSD (DSM IV and V): recurrent and intrusive thoughts about lockdown, persistent avoidance of stimuli associated with the situation, dulling of the individual's general activity, and persistent symptoms of increased arousal. From the combination of stressors and PTSD symptoms (DSM IV and V), we selected 18 items. For each item, the participants had to respond to a frequency scale from 1 (never) to 5 (very frequently) and to an intensity scale from 1 (with no intensity) to 5 (with great intensity).

In addition, for the purpose of understanding potential contagion, we devised two items referring to the students themselves and their environment. In relation to the item referring to the participant, we categorized the degree of presence of the virus: I have not had symptoms or tested positive for Coronavirus and Yes, I have had symptoms or I have tested positive for Coronavirus. And in the assessment of the environment, we dichotomized the degree of presence of the virus: No one in my environment has had symptoms or has tested positive for coronavirus and At least one person in my environment has had symptoms or has tested positive for coronavirus. From the responses to these items we generated two groups. The first, named group in contact with the virus, made up of students who had had symptoms or for whom a positive diagnostic test had been received on their own part or on the part of someone in their environment. The second group, named group without contact with the virus, made up of those who had not had symptoms or for whom no positive diagnostic test had been received on their own part nor on the part of anyone in their environment.

Data collection was carried out using the Google Forms application. The responses collected were transferred to an IBM-SPSS v25 matrix for the accomplishment of the relevant statistical analyses.

Results

In Table 1, we display the descriptive data for the variables of interest.

As our first objective, we sought to understand whether the frequency and intensity presented by the PTSD psychological responses (recurrent and intrusive thoughts, avoidance or dulling, and hyperarousal) associated with the stressors described by Brooks et al. (2020) are high in college students and whether the stressors have an equal effect. To do this, we carried out two GLM repeated measures procedures, the intra factors being the set of five stressors; taking as a dependent variable for the first procedure the frequency of the symptoms associated with the stressors and for the second their intensity. The results indicated that there were effects on the frequency (F (4, 265) = 228.22; p ≤ 0.001) and the intensity (F (4, 266) = 183.15; p ≤ 0.001) of the symptoms associated with the stressors.

The highest frequency of responses corresponded to the stressors Frustration and boredom (M= 3.15; ET= 0.06) and Inadequate information (M= 3.16; ET= 0.06) (Figure 1a). The next stressors for frequency of symptoms were Duration (M= 2.36; ET= 0.06) and Insufficient supplies (M= 2.03; ET= 0.06). The stressor that presented a less frequent response was Fear of contagion (M= 1.82; ET= 0.04).

The PTSD symptoms that manifested with the greatest intensity corresponded to the stressors Frustration and boredom (M= 3.11; ET= 0.06) and Inadequate information (M= 2.95; ET= 0.06) (Figure 1b). The next stressor in respect of intensity of symptoms was Duration (M= 2.37; ET= 0.06). Meanwhile, the symptoms associated with less intense stressors were Insufficient supplies (M= 1.89; ET= 0.06) and Fear of contagion (M= 1.74; ET= 0.04).

With regard to the differences between scores for the 5 stressors in terms of their two measures, the Bonferroni a posteriori tests, both in frequency and in intensity of the stress response associated with the stressors, indicated that the stressors Frustration and boredom and Inadequate information did not differ from each other, but that the rest of the stressors did differ. In turn, the stressors with the lowest scores, Duration, Fear of contagion and Insufficient supplies, differed from each other (Figures 1a and 1b). In no case were values indicating a high frequency and intensity of symptoms associated with the stressors.

As our second objective, we aimed to understand whether the frequency and intensity of stress responses associated with the stressors would increase as the period of confinement increased. To do this, we undertook a GLM repeated measures procedure, with the intra factor being the set of stressors and the inter factor the weeks, taking as the dependent variable the frequency of symptoms associated with the stressors. The results of the multivariate test indicated that there were effects due to stressors (F (4, 262) = 211.89; p ≤ 0.000) but that there was no interaction effect between stressors and weeks. Regarding the inter factor, it indicated that there were differences between the weeks (F (3, 265) = 6.05; p ≤ 0.001). Specifically, the Bonferroni tests indicated that the first and second weeks differed from the fourth (d14 = .46; p ≤ 0.001 and d24 = .39; p ≤ 0.011) the latter week being when the symptoms associated with the stressors appeared less frequently.

The results collected in Figure 2a showed significant differences over the course of the weeks of confinement in the frequency with which the stress responses associated with the stressors appeared. Specifically, in relation to the Duration stressor, the frequency of symptoms decreased significantly when comparing the first week (M = 2.49; ET = 0.94) with the fourth (M = 2.01; ET = 0.14). The frequency of symptoms associated with the stressor Insufficient supplies also diminished over the lockdown period, when comparing the first week (M = 2.33; ET = 0.92) with the third (M = 1.87; ET = 0.11) and with the fourth week (M = 1.55; ET = 0.13).

Following this, we again performed a GLM repeated measures procedure, with the intra factor being the set of stressors, and the inter factor the weeks, taking as the dependent variable the intensity of the symptoms related to the stressors. The results of the multivariate test indicated that there were effects due to stressors (F (4, 263) = 168.56; p ≤ 0.000) but that there was no interaction effect between stressors and weeks. Regarding the inter factor, the results indicated that there were effects due to weeks (F (3, 266) = 4.56; p ≤ 0.004). Specifically, the Bonferroni tests indicated that the first and second weeks differed from the fourth (d14 = .39; p ≤ 0.006 and d24 = .38; p ≤ 0.022) the latter week being when the symptoms associated with the stressors appeared with less intensity.

The results collected in Figure 2b showed significant differences over the course of the weeks of confinement in the intensity with which the stress responses associated with the stressors appeared. Specifically, we only found significant differences in the intensity of the symptoms related to the stressor Insufficient supplies, with the scores decreasing from the first (M = 2.09; ET = 0.89) to the fourth week (M = 1.54; ET = 0.13). As a group, all the stressors, both in the frequency and the intensity of the symptoms related to PTSD, showed a lower magnitude in the final week.

Finally, in regard to the third objective, it interested us to find out if the frequency and intensity of the stress response associated with stressors were higher in students who had been in contact with the virus (themselves or through someone from their environment), when compared to those who had not had any contact.

A variable of interest was to discover how many participants had had symptoms of COVID-19 themselves or had had someone in their environment who had, since this could influence the result of the frequency or intensity with which they manifested the presence of associated PTSD symptoms (Table 2). A very low percentage of students presented symptoms of Coronavirus or tested positive (11.5%). For the 88.5% of students who had no symptoms or tested positive, 57.3% of those around them had either had no symptoms or had not had a positive test. In short, the group in contact with the virus was composed of 137 students (50.7%).

Following this, we performed a GLM repeated measures procedure, wherein the intra factor was the set of stressors and the inter factor was the degree of student contact with the virus, taking as the dependent variable the frequency of the stress response. The results indicated that there were differences between the stressors (F (4, 264) = 225.75; p ≤ 0.000) but that no interaction effects were found (F (4, 264) = 0.43; p ≤ 0.789). The inter effects test indicated that there were effects due to symptoms (F (1, 267) = 16.48; p ≤ 0.000) (Figure 3a). This showed that, in the two groups, the stressors behaved in the same way, although the frequency of PTSD symptoms was lower in the group without contact with the virus.

Subsequently, we performed a GLM repeated measures procedure, wherein the intra factor was the set of stressors and the inter factor was the degree of student contact with the virus, taking as the dependent variable the intensity of the stress response associated with the stressors. The results indicated that there were differences among the stressors (F (4, 265) = 186.18; p ≤ 0.000) and among the interaction effects (F (4, 265) = 2.52; p ≤ 0.041). The inter effects test indicated that there were effects due to symptoms (F (1, 268) = 14.83; p ≤ 0.000) (Figure 3b). As in the case of the psychological symptoms associated with frequency, the group with contact with the virus scored with greater intensity in response to the stressors, with the exception of Fear of Contagion, for which there were no significant differences between the two groups.

Discussion and Conclusions

In respect of the stated objectives, our results indicate that the participants show a moderate or even low magnitude both in the frequency and in the intensity of PTSD symptoms associated with the proposed stressors. As the period of confinement elapsed, both the frequency and the intensity decreased. With regard to the difference between the participants who had had contact with the virus and those who had not, in general, parallel behavior with the stress response is observed, its intensity and frequency being lower in the case of students who did not have symptoms of the virus themselves and did not have anyone in their environment who did.

Regarding the first objective, the scores provided by our students suggest that responses corresponding to PTSD symptoms associated with the stressors, analyzed during lockdown in our population of college students, have been neither frequent nor intense. Participants only manifested, because of the boredom and the frustration of the situation, a certain degree of intrusive thoughts, avoidance, and hyperarousal; in addition, the quality of the information they received also caused some nervousness and concern. Surprisingly, the fear of becoming infected or infecting others was not a source of stress for them. This may be due, in part, to the influence of some psychological variables related to youth development, such as the belief in invulnerability regarding various sources of risk to the health of young people. This belief could also be reinforced by the objective information that the effects of COVID-19 were less intense and less prevalent in this age group.

In relation to our second objective, the weekly evolution of stressors during confinement indicates that the frequency and intensity of symptoms associated with the stressors in the first week do not differ from the second, but that as confinement extends and nears the end of the lockdown, worries, avoidance, or nervousness in respect of these stressors practically disappear. Not all stressors behaved in the same way, both in intensity and frequency of their psychological impact, over the course of the weeks: isolation time ceased to be a stressor in the final week, when the end of confinement was already reported; furthermore, the scant concern about not having sufficient sanitary supplies was even more diminished, with the difference being significant from the third week onward, both in frequency and in intensity.

The third objective of the study is focused on analyzing the role of the presence of the virus in the frequency and intensity of psychological symptoms. Contrary to what might be expected, it has not been a variable with significant weight in the experience of stressors. The group that had been in contact with the virus scored slightly higher for all stressors. However, these differences, although significant, do not indicate the presence of distress among the students. This is more evident in regard to Fear of contagion, where the two groups do not differ in intensity. The belief of invulnerability with respect to the lightness of their symptomatology may explain why they are not excessively concerned about the latter and, therefore, it is logical that they do not fear contagion from the virus.

Our results, on the one hand, diverge from those obtained by studies undertaken in various countries in previous health crises for groups with a similar age range. These studies identified the presence of stressors and symptoms of emotional discomfort, anxiety, depression, or PTSD (Brooks et al., 2020; Shuang - Jiang Zhou et al, 2020; Pan, 2020, Liang et al 2020). On the other hand, our results are also very different from those obtained by recent studies in the Spanish population, where they found symptoms of psychological distress (Balluerka et al., 2020; Justo-Alonso et al., 2020). One possible explanation in regard to these differences may lie in the fact that these authors have focused on assessing the psychological response, while in our work we have focused on quantifying the intensity and frequency with which the PTSD-associated (DSM IV and V) symptomatology is found to be directly associated with stressors.

Drawing on all the above and taking into account the current circumstances, it can be deduced that the absence of risk perception by young people, during the lockdown in the first wave of the virus, continues months later. The low adherence to the public health regulations imposed by the authorities, especially the participation in massive parties, reflects that our young people continue to experience no fear, clearly exposing themselves to contagion and contributing to the maintenance of this critical situation. In consequence, as regards the applications of our study, we consider it to be important to carry out an awareness campaign aimed at young people, in order to raise awareness, to increase perception of the risk factors, and to promote protective factors related to the high contagiousness of COVID-19. In this work, it would be necessary to combine lines of intervention of various types: on the one hand, to display the harshest reality of this experience, with realistic and striking messages and images of the physical, psychological, and socioeconomic consequences of the pandemic; and on the other hand, in order to avoid inoculation against this line of information, as occurs with traffic campaigns, there might be an alternation with other information that lowers the drama, using irony or black humor to show the consequences of not carrying through the protection measures. It is true that fear can paralyze and block our system of response to a traumatic situation, such as COVID-19, and, therefore, not help us to face it adequately, but the absence of any kind of response of concern or anxiety in the current moment keeps us in this situation of real danger.

Finally, as the study participants were college students who voluntarily responded through the virtual teaching support platform, our sample was small, exposing our results to certain biases. Although further application of our questionnaire to larger samples might help to explain the differences between our results and those obtained by other studies, it is also necessary to take into account that subsequent confinement conditions will be very different from those experienced during the first wave of the pandemic.

References

Notes

^* Research article. None conflict of interest was declared.

Author notes

^a Autor de correspondencia. Correo electrónico: avc@usal.es

Additional information

How to cite: Valentín, A., López Pérez, E., González-Tablas, M. del. M., Mateos García, P. M., & Gordillo León, F. (2023). Did college students suffer stress during the covid-19 lockdown? Universitas Psychologica, 22, 1-11. https://doi.org/10.11144/Javeriana.upsy22.dcss

Descargar