key: cord-0913154-a1grf71v
authors: Pompili, Maurizio; Innamorati, Marco; Sampogna, Gaia; Albert, Umberto; Carmassi, Claudia; Carrà, Giuseppe; Cirulli, Francesca; Erbuto, Denise; Luciano, Mario; Nanni, Maria Giulia; Sani, Gabriele; Tortorella, Alfonso; Viganò, Caterina; Volpe, Umberto; Fiorillo, Andrea
title: The impact of Covid-19 on unemployment across Italy: consequences for those affected by psychiatric conditions
date: 2021-09-17
journal: J Affect Disord
DOI: 10.1016/j.jad.2021.09.035
sha: b4d91ccae3b89f0289e6efa6e262783f9b5fdf16
doc_id: 913154
cord_uid: a1grf71v

BACKGROUND: : Severe psychological and psychosocial consequences of the COVID-19 pandemic are expected, especially for people already vulnerable to biological or psychosocial stressors, including those with mental health problems. The study aimed to investigate factors associated with the loss of jobs and unemployment during the COVID-19 pandemic. In particular, we investigated whether mental illness was associated with a higher risk of losing one's job because of the COVID-19 pandemic. METHODS: : Nineteen thousand four hundred ninety-six adults living in Italy were administered an online protocol including a sociodemographic checklist and questionnaires investigating suicide ideation and risk, mental health status and general distress (stress, anxiety, and depression), resilience, and perceived support. RESULTS: : One thousand two hundred seventy-four reported having lost their job because of the COVID-19 pandemic, and 5.4% of the sample reported a mental illness (mostly a depressive disorder). Unemployment was independently associated with mental illness, poor mental health, and depression. Mental illness was associated with the risk of losing one's job because of the COVID-19 pandemic, but not at the multivariate analyses. Those who lost their job because of the COVID-19 pandemic (compared to others) reported worse mental health and depression. LIMITATIONS: : The presence of mental illness was self-reported by respondents and the administered measures were self-reported questionnaires affected by social desirability and other response bias. CONCLUSIONS: : The COVID-19 pandemic and social isolation measures and lockdown used to contain its spread among the Italian population were associated with occupational insecurity, especially among the more vulnerable social categories.

In a recent statement, the WHO (2021) indicated that the COVID-19 pandemic has presented an unprecedented challenge to public health and the world of work, with nearly half of the world's global workforce being at risk of losing their livelihoods. In this situation, the effects on the psychological health of the contagion, the fear of being infected, or of the measures of quarantine and social isolation can be evident, as supported by studies investigating past pandemics (e.g., SARS and Ebola) (Barbisch et al., 2015) . More specifically, Fiorillo and Gorwood (2020) suggested that severe psychological and psychosocial consequences are expected, especially for people already vulnerable to biological or psychosocial stressors, including those with mental health problems.

Literature has indicated that poor health, mainly as a consequence of mental disorder, could be associated with higher unemployment (Chatterji et al., 2007; Claussen, 1999; Goldberg et al., 2001; Heponiemi et al., 2007; Leino-Arjas et al., 1999; Thomas et al., 2005; Zhang et al., 2009) and this, in turn, could be associated with higher risk of suicide (Elbogen et al., 2020; Lundin et al., 2012) . Especially during times of economic hardship, people with mental illness could be the most disadvantaged in unemployment and financial insecurity (Evans-Lacko et al., 2013; Viinamaki et al., 2000) .

The COVID-19 pandemic harmed the global economy, including an increase in the unemployment rate (International Labour Organization, 2020; International Monetary Fund, 2020) and great challenges in the labour market (Brenner and Bhugra, 2020; Crayne, 2020; Kaur et al., 2020; Ksinan Jiskrova et al., 2021) . Furthermore, the COVID-19 pandemic was also associated with other factors influencing mental health and fears of increase in suicide risk, such as social distancing and quarantine measures (Brooks et al., 2020; Daly et al., 2021; Li et al., 2020; McIntyre and Lee, 2020; Pompili, 2021; Unutzer et al., 2020) . Now, according to several researchers, all these factors could lead to an epidemic of suicide ideation and behaviors (Kawohl and Nordt, 2020; McIntyre and Lee, 2020; Samson and Sherry, 2020) . For example, in a recent publication Iob, Steptoe, and Fancourt (2020) investigated abuse, self-harm, and thoughts of suicide/self-harm in the UK during the first month of the COVID-19 pandemic using data from the and reported that 18% of respondents experienced thoughts of suicide or self-harm in the first month of lockdown and 5% reported harming themselves at least once since the start of the lockdown.

Thus, the study aimed to investigate factors associated with the loss of one's job and unemployment during the COVID-19 pandemic. In particular, we investigated whether mental illness was associated with a higher risk of losing one's job because of the COVID-19 pandemic and whether geographic variations were visible (i.e., northern Italian regions vs.

central Italian regions vs. southern regions and major islands). We hypothesized that people with poor health (i.e., those with psychiatric illnesses) have a higher risk of being unemployed and losing their job because of the COVID-19 pandemic. We also hypothesized that unemployment and job loss could be associated with a higher risk of suicide ideation, higher levels of general distress (i.e., depression, anxiety and distress), and lower general health levels.

This is a cross-sectional study part of the Covid Mental Health Trial (COMET), a national trial coordinated by the University of Campania "Luigi Vanvitelli" (Naples) in collaboration with nine Italian universities. Information about the design of the COMET collaborative network study can be found in Giallonardo et al. (2020) and . The research design of the study has been approved by the Ethical Review Board of the University of Campania "L. Vanvitelli".

The participants were 19,496 adults (14,017 women and 5,479 men) living in Italy.

Mean age of the sample was 39.0 years (SD=13.2; age range=18/83 years).

Sociodemographic characteristics are reported in Table 1 . Participants were recruited according to a snowball sampling method within an online survey. Inclusion criterion was being 18 years old and older. No exclusion criteria were defined for the recruitment of the sample. The full methodology is described in detail in Giallonardo et al. (2020) .

All the participants were administered an online protocol described elsewhere . For this study, we analyzed sociodemographic information (sex, age, marital status, school attainment, working status, and area of residence) and clinical information (positivity for the COVID-19 virus, presence of physical and mental illness), epidemiological data (number of positives of COVID-19 in the Italian regions), and data from the following psychological tests: the General Health Questionnaire -12 items version (GHQ) (Goldberg et al., 1997) , the Depression Anxiety Stress Scales (DASS-21) (Lovibond and Lovibond, 1995) , the Connor Davidson Resilience Scale (CD-RISC) (Connor and Davidson, 2003) , the Multidimensional Scale of Perceived Social Support (MSPPS) (Zimet et al., 1990) , and the Suicidal Ideation Attributes Scale (SIDAS) (van Spijker et al., 2014) . Two questions assessed the working status of the participants. People were asked whether they were unemployed at the time of the assessment and in people who responded positively it was asked whether they loss their job because of the spread of the COVID-19 pandemic. The presence of mental and physical health problems was self-reported by participants.

The GHQ (Goldberg et al., 1997 ) is a 12-item questionnaire assessing mental health status. Each item is rated on a 4-point Likert-type scale (0-3). A total score is calculated, and higher scores are indicative of worse mental health status. The Italian version of the GHQ proved to be a reliable scale with a Cronbach's alpha of 0.81 (Politi et al., 1994) . In the present sample Cronbach's alpha was 0.73.

The DASS-21 (Lovibond and Lovibond, 1995) is a 21-item questionnaire assessing three dimensions of psychopathology: depression, anxiety, and stress. Each item is rated on a 4-level Likert-type scale (0-3). We calculated a total score as an index of general distress, and scores for the three dimensions (i.e., depression, anxiety, and stress). The Italian version of the DASS-21 demonstrated good psychometric properties (e.g., internal consistency and convergent validity) (Bottesi et al., 2015) . In the present sample Cronbach's alpha was 0.82.

The CD-RISC (Connor and Davidson, 2003) is a 10-item scale assessing resilience.

Each item is rated on a 6-level Likert-type scale, and higher scores indicate higher levels of controllability, closeness to attempt, and level of distress associated with suicidal ideation.

Participants are asked to respond on a 10-level Likert-type scale. Higher scores are indicative of higher suicide risk. The SIDAS demonstrated good internal consistency and convergent validity in Australian (van Spijker et al., 2014) and Chinese (Han et al., 2017) adults. We dichotomized responses to the item no. 1 (i.e., how often have thoughts about suicide) and

reported frequencies and percentages of those who endorsed a frequency >1 indicating the presence of suicide ideation. In the present sample Cronbach's alpha was 0.82.

All the analyses were performed with the statistical package for social sciences (SPSS) 19.0. A series of one-way Fisher exact tests, chi-squared tests, and independent sample t-tests were used to assess differences between groups at the bivariate level (people who were unemployed at the time of the assessment vs. other respondents, and those who lose one's job because the COVID-19 pandemic vs. others). Bonferroni correction was used for controlling multitesting. Cohen's d, Cramer's v, and phi coefficients were reported as measures of effect sizes. Small effect sizes were indicated by d>0.2 or v|phi > 0.1, medium effect sizes by d>0.5 or v|phi > 0.3, and large effect sizes by d>0.8 or v|phi>0.5.

Variables significant at the bivariate analyses were included as independent variables in two generalized linear models (with a robust estimator) used to fit a binary logistic regression (binomial distribution and logit link function were used). Job status (unemployed vs. others) and loss of one's job because the COVID-19 pandemic (those who lost their jobs vs. others) were included in the analyses as a criterion. Odds ratios (OR) and their 95% confidence intervals (95% CI) were reported as measures of association. ORs > 1 indicate higher risk in the index group when compared to the reference group. ORs < 1 indicate lower risk in the index group when compared to the reference group (i.e., higher risk in the reference group compared to the index group). Positivity of the COVID-19 virus was not included in the multivariate analyses because of the low number of subjects included in some categories. All tests are significant at p<0.05.

Three thousand sixty-five participants reported being unemployed at the time of the assessment (Table 1) Only 1.5% of the respondents reported having been tested positive for COVID-19.

Around 13% reported having a physical illness, and 5.4% reported a mental illness (mostly depressive disorders). Fourteen percent of the sample reported a higher risk for suicide ideation.

Differences between groups are reported in Tables 1 and 2 

Differences between groups are reported in 

In our sample, people who reported having a mental disorder reported more frequently to be unemployed than other respondents. Furthermore, among those with a job before the COVID-19 pandemic, people who reported having a mental disorder also reported having lost their job during the COVID-19 pandemic (despite being not significant at the multivariate analyses). This supports the idea that people with mental illness could be less resilient and more disadvantaged in unemployment during an economic crisis (Evans-Lacko et al., 2013; Viinamaki et al., 2000) . Furthermore, social integration could be problematic in people with mental illness even without considering the possible consequences of social distancing and quarantine measures adopted with the COVID-19 pandemic (Mueller et al., 2006) . Such measures could have affected those with mental illness more than other people and have limited their social network and their ability to receive emotional support to cope with stress deriving from fear of being infected or losing loved ones because of COVID-19, or even fear of losing their job. Social networks' disintegration and poor social support are considered to be potent risk factors for suicide behaviors and ideation (Heikkinen et al., 1993; Innamorati et al., 2008) .

In a recent study (Iob et al., 2020) investigating self-harm and thoughts of suicide/self-harm in the UK during the first month of the COVID-19 pandemic, 18% of respondents experienced thoughts of suicide or self-harm in the first month of lockdown, and 5% reported harming themselves at least once since the start of the lockdown. In our study, 14% of the sample reported a higher risk for suicide ideation (a score > 1 at the first item of the SIDAS), and no differences were significant between participants who reported to be unemployed and those employed, or between respondents who reported having lost their job because the COVID-19 pandemic and those who did not. No differences were present even when limiting these analyses among respondents with mental illness (unemployed vs. employed: 14.9% vs. 14.2%, p=0.41; people with mental illness who lost their job because of the COVID-19 pandemic vs. others: 14.1% vs. 18.5%, p=0.16). This is not in line with our hypothesis and the general view that the COVID-19 pandemic could be associated with an epidemic of suicide ideation and behaviors (Kawohl and Nordt, 2020; McIntyre and Lee, 2020; Samson and Sherry, 2020) .

Although the groups did not differ for suicide ideation, people who reported to be unemployed (compared to other respondents) or who reported having lost their job (compared to other respondents) had worse mental health status and higher distress, depression, and anxiety. General distress and depression could be significant predictors of suicide behaviors and ideation (Overholser et al., 2012) , but this association could be mediated by other psychological factors (Campos et al., 2017) . In our sample, people who reported losing their job despite reporting worse mental health status and distress still reported being resilient and having social support not dissimilar from other respondents.

These results could explain why those who lost their job did not report an increase in suicide risk despite higher distress and worse mental health status. This finding is in line with recent results from the international literature which highlighted that suicide rates were overall stable or sometimes decreased during the first year of the pandemic (Pirkis et al., 2021; Pompili, 2021) .

Our results also pointed to other psychosocial factors of vulnerability when facing the COVID-19 pandemic-being female and having lower school attainment were independently associated with a higher risk of losing one's job because of the COVID-19 pandemic (Garrouste et al., 2010; Newell, 2020) . For example, Garrouste, Kozovska, and Perez (2010) suggested that an individual's probability of being in long-term unemployment could decrease with his/her educational level (Eurostat, 2019) . Furthermore, Italy is ranked amongst the countries in the EU with the lowest gender equality (Rosselli, 2014) , especially in the work and labor market. Higher vulnerability in women could be associated with the fact that women (compared to men) have more frequently low-paid or temporary jobs (Newell, 2020) . Furthermore, teleworking demands could be more difficult in women to conciliate with family responsibilities (Eurofound, 2020) .

Furthermore, geographic variations were evident with a higher risk of losing one's job because the COVID-19 pandemic in the Italian southern regions and major islands than in the Italian northern and central regions, reflecting the economic and social imbalance among the Italian regions. For example, southern Italian regions are still disadvantaged in terms of life expectancy and access to care and quality of health services (Ferré et al., 2014) .

Our findings have some limitations to their generalizability. First, the presence of mental illness was self-reported by respondents. Second, the administered measures were self-reported questionnaires affected by social desirability and other response bias. For example, the presence of suicide risk was evaluated with a single item of the SIDAS and not with clinical interviews assisted by questionnaires such as the Columbia Suicide Severity

Rating Scale (Posner et al., 2011) . Third, the data was gathered between March and May 2020 during the first national lockdown, and thus, it could not represent the situation during the months following the first lockdown characterized by local lockdowns and several changes in the Italian national policies regarding the emergency of the COVID-19 pandemic.

However, this study has several strengths. For example, the online survey allowed us to reach a large portion of the Italian population in a pandemic situation when face-to-face contacts were not possible. The methods used allowed us to recruit a large sample representative of the Italian population.

The COVID-19 pandemic and social isolation measures and lockdown used to contain its spread among the Italian population were associated with occupational insecurity, especially among the more vulnerable social categories (for example, people with mental illness and women). Clinicians should assess the presence of life events, such as the risk of losing one's job because of the COVID-19 pandemic in psychiatry patients. Unemployment and loss of one's job were associated with worse mental health and general distress, and higher suicide risk. Thus, supportive psychosocial interventions are needed for the general population in order to help people at risk to overcome stress and anxiety due to the pandemic.

Moreover, targeted interventions for job employment or support should be developed and provided to people with pre-existing mental disorders who have a higher risk of losing job. 

Is There a Case for Quarantine? Perspectives from SARS to Ebola

The Italian version of the Depression Anxiety Stress Scales-21: Factor structure and psychometric properties on community and clinical samples

Acceleration of Anxiety, Depression, and Suicide: Secondary Effects of Economic Disruption Related to COVID-19

The psychological impact of quarantine and how to reduce it: rapid review of the evidence

Does psychache mediate the relationship between general distress and suicide ideation?

Psychiatric disorders and labor market outcomes: evidence from the National Latino and Asian American Study

Health and re-employment in a five-year follow-up of long-term unemployed

Development of a new resilience scale: the Connor-Davidson Resilience Scale (CD-RISC)

The traumatic impact of job loss and job search in the aftermath of COVID-19

Associations between periods of COVID-19 quarantine and mental health in Canada

Psychometric evaluation of the Multidimensional Scale of Perceived Social Support (MSPSS) in people with chronic diseases

Misurare il supporto sociale percepito: Proprietà psicometriche della Multidimensional Scale of Perceived Social Support (MSPSS) in un campione di studenti universitari. Risorsa Uomo. Rivista di Psicologia del Lavoro e dell

Connor-Davidson Resilience Scale: Proprietà psicometriche della versione italiana

Multidimensional Scale of Perceived Social Support (MSPSS): un contributo alla validazione italiana [Multidimensional Scale of Perceived Social Support (MSPSS): A contribution to Italian validation

Financial Strain and Suicide Attempts in a Nationally Representative Sample of US Adults

Living, working and COVID-19. Publications Office of the European Union

The higher the education, the lower the unemployment rate

The mental health consequences of the recession: economic hardship and employment of people with mental health problems in 27 European countries

Italy: Health System Review

The consequences of the COVID-19 pandemic on mental health and implications for clinical practice

Effects of the lockdown on the mental health of the general population during the COVID-19 pandemic in Italy: Results from the COMET collaborative network

Education and Long-Term Unemployment

The Impact of Quarantine and Physical Distancing Following COVID-19 on Mental Health: Study Protocol of a Multicentric Italian Population Trial

The validity of two versions of the GHQ in the WHO study of mental illness in general health care

Correlates of long-term unemployment among inner-city adults with serious and persistent mental illness

Translation and validation of the Chinese versions of the Suicidal Ideation Attributes Scale, Stigma of Suicide Scale, and Literacy of Suicide Scale

Life events and social support in suicide

Relationship between unemployment and health among health care professionals: health selection or health effect?

Completed versus attempted suicide in psychiatric patients: a psychological autopsy study

Almost 25 million jobs could be lost worldwide as a result of COVID-19, says ILO

World economic outlook: the great lockdown International Monetary Fund

Abuse, self-harm and suicidal ideation in the UK during the COVID-19 pandemic

Individual, interpersonal and economic challenges of underemployment in the wake of COVID-19

COVID-19, unemployment, and suicide

Job loss and lower healthcare utilisation due to COVID-19 among older adults across 27 European countries

Predictors and consequences of unemployment among construction workers: prospective cohort study

Anxiety and depression among general population in China at the peak of the COVID-19 epidemic

Manual for the Depression Anxiety & Stress Scales

Unemployment and suicide in the Stockholm population: a register-based study on 771,068 men and women

Projected increases in suicide in Canada as a consequence of COVID-19

Social support modifies perceived stigmatization in the first years of mental illness: a longitudinal approach

Gender inequality in Italy at the time of the pandemic

Understanding suicide risk: identification of high-risk groups during high-risk times

Reliability, validity and factor structure of the 12-item General Health Questionnaire among young males in Italy

Can we expect a rise in suicide rates after the Covid-19 pandemic outbreak?

The Columbia-Suicide Severity Rating Scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults

The Policy on Gender Equality in Italy. Policy Department for Citizens' Rights and Constitutional Affairs

Projected increases in suicide in Canada as a consequence of COVID-19 revisited

Employment transitions and mental health: an analysis from the British household panel survey

Psychiatry in the age of COVID-19

The suicidal ideation attributes scale (SIDAS): Communitybased validation study of a new scale for the measurement of suicidal ideation

Mental health at population level during an economic recession in Finland

Impact of COVID-19 on people's livelihoods, their health and our food systems Joint statement by ILO, FAO, IFAD and WHO 13

Chronic diseases and labour force participation in Australia

Psychometric characteristics of the Multidimensional Scale of Perceived Social Support

Reference group=lower depression or anxiety or stress. In bold significant tests

None