key: cord-1043564-fz1yfhme
authors: Munoz-Navarro, R.; Cano-Vindel, A.; Schmitz, F.; Cabello, R.; Fernandez-Berrocal, P.
title: Emotional distress and associated sociodemographic risk factors during the COVID-19 outbreak in Spain
date: 2020-05-30
journal: nan
DOI: 10.1101/2020.05.30.20117457
sha: c9003f70329b4a57bc99e6a4383b16708caf59d9
doc_id: 1043564
cord_uid: fz1yfhme

Background Spain is one of the countries that has been most affected by COVID-19 disease. However, the emotional distress on the Spanish population remains poorly understood. The aim of this study was to determine the presence of emotional disorders and related symptoms and to assess the influence of sociodemographic characteristics on this population. Methods We conducted a cross-sectional survey using an online platform. Participation was completely voluntary. Sociodemographic variables were collected and symptoms of anxiety, depression, and panic were assessed through three questionnaires: Generalized Anxiety Disorder-7 (GAD-7), Patient Health Questionnaire-9 (PHQ-9), and the Patient Health Questionnaire-Panic Disorder (PHQ-PD). Chi-squared statistics were applied to determine the influence of sociodemographic variables on symptom severity and diagnosis. Results Most respondents (n=1753) were female (76.8%), with a mean (SD) age of 40.4 (12.9) years; 39.1% were married and 39.5% held a high school degree. Severe and moderate symptoms of anxiety and depression were high (6.5% and 14.3%; 9.9% and 12.9%, respectively) and more than one in four participants (25.7%) experienced a panic attack. A high proportion of participants met diagnostic criteria for emotional disorders: 15.3% for GAD; 12.2% for MDD; and 17.2% for PD. Significant risk factors were as follows: female, young adult, single, unemployed, and low income. Conclusion Due to limitations related to the study design (convenience sample), the findings of these study may not be applicable to the general population. Nonetheless, the high prevalence of emotional symptoms and disorders in this sample suggests that mental health interventions are urgently needed in Spain.

Emotional distress during the COVID-19 outbreak in Spain experiencing symptoms on most days as specified by one of the first two questions, plus symptoms on four other items. This algorithm was shown to have optimal sensitivity and specificity (.88 and .80) for the diagnosis of MDD in a Spanish primary care sample (18) and presented also excelent psychometric properties (19) .

The PHQ-PD is the module that assesses DSM-IV-based panic disorder (PD). We used the validated Spanish-language version of this instrument (20), which can be used to screen for panic attacks in the last two weeks, and includes an algorithm for the diagnosis of PD. In this validation study, the screening question yielded a sensitivity score of .83 (specificity = .66) and the modified algorithm of the original version yielded a sensitivity of .77 (specificity = .72) which makes the PHQ-PD a very useful tool to asses for panic attacks and panic disorder.

Data stratified by the level of symptoms of depression, anxiety, and panic attacks as well as for the diagnosis (GAD, MDD and PD) are presented as numbers and percentages. For symptom severity and diagnoses, chi-squared tests were applied to determine whether sociodemographic variables (gender, age, marital status, educational level, level of income, and employment situation) were differentially associated with levels of symptom severity and the proportion of diagnoses. The SPSS statistical software version 26.0 (IBM Corp) was used to obtain z-values and standardized residuals, and post-hoc tests were performed using chi-squared tests on the squared residuals (df=1). Bonferroni correction of the z-values was applied for value greater than +/-1.96 (α = .05). Results were reported in the text as: (z-value; p-value).

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. . https://doi.org/10.1101/2020.05.30.20117457 doi: medRxiv preprint Emotional distress during the COVID-19 outbreak in Spain

Participation was completely voluntary, the survey was anonymous, and confidentiality of all information provided was assured. Before starting the survey, all participants were required to read the instructions and provide informed consent and could abandon the survey at any time, for any reason. The study was approved by the clinical research ethics committee of Hospital University La Fe of Valencia. The confidentiality of personal data was protected under the Spanish Data Protection Law. Table 1 ).

---Insert Table 1 here --- p<.05), the 40-59 year age group (-2.7; p<.001) and finally by the group >=60 years of age (-3.2, p<.01) (see Table 2 ).

Marital status was significant for depression (p<.001) and panic attacks (p<.01). The proportion of married people with inexistent depressive symptoms (5.7; p<.001) was significantly higher than in singles (-5.7; p<.001), with a lower proportion of moderate depressive symptoms (-5,6; p<.001), and less severe depressive symptoms (-2.1; p<.001) than singles (2.1; p<.001). The proportion of married people who experience a panic attach (-3.8; p<.001] was significantly less than that observed in the group of people living with an unmarried partner (3.6; p<.01) (see Table 3 ).

---insert Table 3 here ---Educational level was significant for anxiety (p<.001), depression (p<.001) and panic attacks (p<.01). The proportion of people with a postgraduate degree with inexistent anxiety symptoms (2.6; p<.001) was higher than observed in university graduates (-2.2; p<.05). People with a high school degree presented low levels of mild symptoms (-3.1; p<.001]. Postgraduate participants presented a lower proportion of moderate anxiety symptoms (-2.4; p<.01) compared to people with a high school degree (2.5; p<.01).

Finally, the highest proportion of severe anxiety symptoms was observed among . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table 3 ).

---insert Table 4 here ---Income levels were significant for anxiety (p<.001), depression (p<.001) and panic attacks (p<.001). The proportion of people earning more than €60,000 presenting inexistent anxiety symptoms (3.2; p<.001) and mild anxiety symptoms (-2.3; p<.05) was significantly less than people earning between €36.000-60,000 (2.8; p<.01). The latter also presented a lower proportion of moderate anxiety symptoms (-3.9; p<.001) compared to people earning less than €12,000 (2.8; p<.01). Finally, people earning €12,000-24,000 presented the highest proportion of severe anxiety symptoms (2.1; p<.001) compared to the lowest proportion observed in those earning €24,000-36,000 (2.2; p<.001). For depressive symptoms, the proportion of people earning < €12,000

with inexistent depressive symptoms (-2.7; p<.001) was less than in those earning €12,000-24,000 (-2.1; p<.01), people earning €36,000-60,000 (3.9; p<.001) and people earning more than €60,000 (2.6; p<.001). Further, people earning < €12,000 presented the highest proportion of severe depressive symptoms (3.1; p<.001) compared to those earning between €36,000-60,000 (-2; p<.05) and more than €60,000 (-2.6; p<.001).

People earning < €12,000 presented the highest proportion of panic attacks (3.2; p<.01)

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. . https://doi.org/10.1101/2020.05.30.20117457 doi: medRxiv preprint Emotional distress during the COVID-19 outbreak in Spain compared to those earning between €12,000-24,000 (2.7; p<.05), between €24,000-36,000 (-2.1; p<.05) and more than €60,000 (-3; p<.001) (see Table 4 ).

---insert Table 5 here ---Employment status was significant for anxiety (p<.001), depression (p<.001) and panic attacks (p<.01). Retired participants presented the highest proportion of inexistent anxiety symptoms (2.6; p<.001) while individuals who had been laid off presented the highest proportion of mild anxiety symptoms (2.5; p<.05). Part-time workers (2.2; p<.05) and unemployed not searching for work (2.5; p<.05) presented the highest proportion of moderate anxiety symptoms compared to full-time workers (-2.9; p<.01). No results were found in people that were laid-off (see Table 5 ).

proportion of people with mental health issues. The prevalence of anxiety and depressive symptoms and panic attacks was high, with 20.8% of participants presenting moderate to severe anxiety symptoms, 22.8% with moderate to severe depressive symptoms and a 25.7% presenting panic attacks. The prevalence of emotional disorders was also high, with 15.2% meeting diagnostic criteria for GAD, 12.2% for MDD, and 17.2% for PD. As said, a very recent study conducted in a large sample (n=3480) in Spain, reported symptoms of depression, anxiety and PTSD in a 18.7%, 21.6% and 15.8%, respectively (12). This prevalence was somehow higher that in our study, probably after the use of very brief tools like the GAD-2 and the PHQ-2, with only two items, which can lead to some false positives (21) . In our study, the use of levels of symptoms and diagnostics algorithms may offer a more accurate prevalence.

In this line, the National Health Survey, conducted by the Spanish Ministry of Health, reported a prevalence of near 6.7% for anxiety and depression; a 9.1% and 9.2% for female and a 4% and 4.3% for male, respectively (22) . Previous research on the prevalence of emotional disorders in the general population in Spain has shown that the prevalence of anxiety and mood disorders is 6.2% and 4.4%, respectively (23,24) For panic disorders, one study found a prevalence of 0.8% in Europe and 0.6% in Spain (24). The prevalence rates observed in the current study are much higher than previous reports, probably because the prevalence in our study is based on screening tools rather than clinical interviews. A more recent study conducted by Navarro-Mateu et al. (25) in the Spanish region of Murcia presented prevalence rates for anxiety and mood disorders of 9.7% and 6.6%, respectively. Those authors argued that the increase of prevalence was due to stressors such as socioeconomic risks. Other studies have shown that the prevalence of emotional disorders rises in response to natural disasters (earthquakes)

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In terms of the influence of sociodemographic factors, consistent with other published reports (28), we found that several factors were associated with an increased likelihood of presenting symptoms and/or being diagnosed with an emotional disorder: female sex, young age (18-25 years), single, low educational level, unemployed, and low income.

Our findings show that females appear to be more affected by the current pandemic than males, with a higher prevalence of GAD and PD, a finding that is consistent with previous reports (29). Nearly one-third of females (29.2%) in our survey reported experiencing a panic attacks, while moderate symptoms of anxiety and depression were also very high (15.8% and 14.6%, respectively) in among females. While no differences in MDD were observed between males and females, a surprisingly high percentage of female-19.8% and 16.5%-presented PD or GAD, respectively. Navarro-Mateu et al.

(2017) reported similar higher prevalence rates for women; although the prevalence in our study was considerably larger.

We also found that young adults (age18-25) and adults (age 26-39) presented more emotional disorders (GAD, MDD and PD) than older people. Among young adults, the proportion presenting panic attacks (32.8%), moderate to severe depressive symptoms (36.5%), and moderate to severe anxiety symptoms (29%) were surprisingly high.

Although the increased prevalence of these disorders in young adults has been previously reported (30), the prevalence rate of emotional disorders in our sample was substantially higher. This finding suggests that it is imperative to implement preventive . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. . https://doi.org/10.1101/2020.05.30.20117457 doi: medRxiv preprint Emotional distress during the COVID-19 outbreak in Spain measures in young adults (31). By contrast, elderly and retired people were the least affected of all the subgroups, perhaps due their experience with previous crises, which may have provided them with adequate coping mechanisms.

Severe depressive symptoms and PD were more prevalent in singles than in married people and those living with a non-married partner. Interestingly, people living with a non-married partner presented higher rates of PD than married people. This could be explained by the fact than being married provides some security in affective and socioeconomic areas. For instance, a study conducted in a large Spanish sample (>10,000 participants) examined the influence of gender and partner/marital status with respect to social instability, finding that a poorer mental health status was associated with poor stability among cohabiting women but not among married ones (32). This suggests that marriage can function as a protective factor for stability, especially in times of crisis, which may also explain why single people may be more affected.

For level of education, our results indicate that a higher level of education was a protective factor against emotional disorders. We found that university graduates presented higher rates of anxiety than those with a postgraduate educational level, while a higher proportion of people with basic education presented depression and panic disorder compared to postgraduate participants. These findings are consistent with previous research showing that a low educational level may be a risk factor for mental health problems (28). However, it is important to consider that the present study, based on an online survey, was comprised of a large proportion of well-educated people who are also more skilled with new technologies. Consequently, our findings may underestimate the true proportion of people in the general population currently suffering from psychological problems.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. . https://doi.org/10.1101/2020.05.30.20117457 doi: medRxiv preprint

We also found that low income was a significant risk factor. A family income level of less than €24,000 per year was a risk factor for PD, GAD, and MDD. Indeed, severe depressive symptoms were more prevalent in people earning less than €12,000, accounting for 15.2% of that group. Even more shocking was the high rate of panic attacks among people earning less than €12.000 (nearly 34%) or less than €24,000 Finally, employment status was a predictor for emotional disorders, a finding that is related to income level. Full-time workers and retired people were less affected by psychological problems than unemployed people, something consistent also with previous research (28). Interestingly, no results were found in people that were laid-off, something that could have been expected.

The main limitation of this study was the convenience sample, which was l comprised of volunteers through an online survey. Overall, the sample was younger and more highly educated than the general population. Consequently, the sample is not representative of the general population. In addition, people who were more affected by the crisis may have been more willing to participate in the study. Given the likely . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. . https://doi.org/10.1101/2020.05.30.20117457 doi: medRxiv preprint Emotional distress during the COVID-19 outbreak in Spain presence of self-selection effects, it is possible that prevalence rates are biased upwards.

However, given the enormity of the potential health and socioeconomic threat posed by the virus, the high prevalence rates observed in this study certainly seem plausible, a finding that is further supported by previous research showing that quarantines can produce negative psychological consequences, including PTSD, confusion, and anger (7). Potential stressors during quarantine include quarantine duration, fear of infection, frustration, and boredom, among others. Future research should investigate these risk factors and the psychological resources like emotion regulation strategies that may be protective against the onset of emotional disorders (33).

The findings of this study represent a call for action in Spain and worldwide. Our data show a major impact of this global health crisis on mental health and it seems probable that the resulting economic crisis may be even more harmful. As previous research has shown, the 2008 economic crisis had a severe negative impact on mental health in Spain (25, 34, 35) . Clearly, there is a need to implement preventive and treatment strategies as well as to reinforce health care services in times of crisis. Indeed, primary care services should expect a rapid and significant increase in demand due to the increased prevalence of common mental health problems (34,36). Unfortunately, the availability of evidence-based psychological treatments for emotional disorders in the primary care setting in Spain and globally is scant (37). For this reason, we believe that is essential to reinforce primary care services to help patients with emotional disorders (38). Also, preventive strategies (31) should be implemented worldwide in the general population to help address the mental health crisis currently facing the world (27).

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. We thank all the collaborators who kindly helped in the sample recruitment process.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. . More than €60,000 125 7.1 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 30, 2020. . 

Differences in the determinants of posttraumatic stress disorder and depression after a mass traumatic event

Post-traumatic stress disorder following disasters: A systematic review

The psychological distress and coping styles in the early stages of the 2019 coronavirus disease (COVID-19) epidemic in the general mainland Chinese population: a web-based survey

Mental health Emotional distress during the COVID-19 outbreak in Spain primary care setting in Spain

A computerized version of the Patient Health Questionnaire-4 as an ultra-brief screening tool to detect emotional disorders in primary care

Resumen metodológico. Encuesta Nac Salud España

Use of mental health services in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project