key: cord-0990630-bgpu7r53 authors: Melchior, M.; Florence, A.-M.; Davisse-Paturet, C.; Falissard, B.; Galera, C.; Hazo, J.-B.; Vuillermoz, C.; WARSZAWSKI, J.; Dione, F.; Rouquette, A. title: Labor market position and depression during the COVID-19 epidemic among young adults (18 to 30 years): a nationally representative study in France. date: 2022-03-30 journal: nan DOI: 10.1101/2022.03.25.22272948 sha: 1847b7229878089a15a33c3a47bca3d18a41259a doc_id: 990630 cord_uid: bgpu7r53 Objective: To examine the relationship between young adults labor force participation and depression in the context of the COVID-19 pandemic. Design, Setting, Participants: Data come from the nationally-representative EPICOV cohort study set up in France, and were collected in 2020 and 2021 (3 waves of online or telephone interviews) among 2217 participants aged 18-30 years. Participants with prior mental health disorder (n=50) were excluded from the statistical analyses. Results: Using Generalized Estimating Equation (GEE) models controlled for participant socio-demographic and health characteristics and weighted to be nationally-representative, we found that compared to young adults who were employed, those who were studying or unemployed were significantly more likely to experience depression assessed using the PHQ-9 (multivariate ORs respectively: OR: 1.29, 95% CI 1.05-1.60 and OR: 1.50, 1.13-1.99). Stratifying the analyses by age, we observed than unemployment was more strongly associated with depression among participants 25-30 years than among those who were 18-24 years (multivariate ORs respectively 1.78, 95% CI 1.17-2.71 and 1.41, 95% CI 0.96-2.09). Being out of the labor force was, to the contrary, more significantly associated with depression among participants 18-24 years (multivariate OR: 1.71, 95% CI 1.04-2.82, vs. 1.00, 95% CI 0.53-1.87 among participants 25-30 years). Stratifying the analyses by sex, we found no significant differences in the relationships between labor market characteristics and depression (compared to participants who were employed, multivariate ORs associated with being a student: men: 1.33, 95% CI 1.01-1.76; women: 1.19, 95% CI 0.85-1.67, multivariate ORs associated with being unemployed: men: 1.60, 95% CI 1.04-2.45; women: 1.47, 95% CI 1.01-2.15). Conclusions and relevance: Our study shows that in addition to students, young adults who are unemployed also experience elevated levels of depression in the context of the COVID-19 pandemic. These two groups should be the focus of specific attention in terms of prevention and mental health treatment. Supporting employment could also be a propitious way of reducing the burden of the Covid-19 pandemic on the mental health of young adults. The COVID-19 pandemic has had a major impact on worldwide population mental health, still unresolved today (1). In particular, there is evidence of significant increases in the prevalence of major depression (up to approximately 27%) and anxiety disorders (up to approximately 25%), respectively leading to an estimated 50 and 76 million new cases globally (1). In the general population, factors associated with a deterioration in mental health include female sex, age under 35 years, loneliness and low income (2) . In addition, several studies have found that university students experienced especially large increases in the prevalence of depression and anxiety disorders, with approximately a third suffering from either condition (3) (4) (5) . However, most of these studies were conducted in China (3) (4) (5) where the spread of COVID-19 was limited after 2019 and accompanied by strict preventive measures, or in the United States (3) where to the contrary, the toll of the COVID-19 epidemic has been high and distance teaching was implemented for extended periods of time. Data from additional countries can help better understand the extent to which students are a high-risk group in terms of mental health during in the context of the current sanitary crisis. Moreover, young people are disproportionately likely to experience job instability and unemployment, which increased during the course of the COVID-19 epidemic and have been identified as risk factors of depression and anxiety (6) (7) (8) (9) (10) . The association between unemployment and mental health has previously been observed and is considered to be partly bidirectional (11), indicating that it is essential to consider preexisting mental health difficulties in assessing changes in psychological distress during the course of the COVID-19 epidemic. France is characterized by high levels of youth unemployment compared to other OECD countries: 18.5% among [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] year-olds in 2021, as compared with 11.5% on average (12) . Additionally, a high proportion of youths are Not in Employment, Education or Training (NEET): 19% of 20-24 year-olds as compared to 15.5% on average (13) . Youth All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Moreover, France is among countries which were hit especially hard by the COVID-19 pandemic, particularly in 2020 and 2021. By March 2022, the COVID-19 epidemic has caused the death of over 140 000 persons (https://coronavirus.jhu.edu/region/france) and resulted in a strict national lockdown from March to May 2020, followed by two mitigated lockdowns from November 2020 to January 2021 and from April to June 2021. Between these periods, a curfew was implemented, and social and educational activities were restricted. According to a nationally representative survey of the French population conducted by the French Public Health Agency, the prevalence of symptoms of depression among 18-24 year-olds was estimated to be 16% in the Spring 2020 and 22% in January 2022, while respectively 33% and 43% suffered from symptoms of anxiety, that is double the rates observed prior to the COVID-19 epidemic (15) . Similarly, a study conducted by the French Ministry of Health estimated, that in May and November 2020, respectively 22 and 19 % of 15-24 years old experienced symptoms of depression while the prevalence prior to the COVID-19 pandemic was estimated to be 10% (16) . The aim of our study was to test the association between labor market participation and depression among young adults during the course of the COVID-19 epidemic. In particular, we compared young adults who were employed to those who were studying or unemployed, while accounting for sociodemographic and health factors potentially associated with depressive symptoms, including history of mental health problems prior to the COVID-19 epidemic. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Data come from the EPICOV study, a French nationally-representative cohort designed to assess the main characteristics of the COVID-19 pandemic on the population of France as well as its impact in terms of sociodemographic and health outcomes (17) . As previously described, participants (≥ 15 years of age, residing in mainland France or three out of five overseas territories) were randomly selected from the national tax database (FIDELI). FIDELI covers 96.4% of the population living in France, providing postal addresses for all individuals, and an e-mail address or telephone number for 83%. Sampling was stratified on two characteristics: residential area ('département' -equivalent to a county), and area-level poverty defined as 60% of the median household income per capita. This sampling frame was designed to ensure overrepresentation of less densely populated and more socioeconomically disadvantaged areas. Individuals living in residential care or prison were not invited to participate in the study. All potential participants were contacted by post, e-mail and text messages (SMS), with up to seven reminders. Self-computer-assisted-web interviews (CAWI) were conducted with 80% of study participants. The remaining 20% were randomly assigned to CAWI or computerassisted-telephone interviews (CATI). The first wave of data collection took place in the Spring 2020 (02/05/2020 -12/06/2020) at the end of the first national COVID-19-related lockdown; the second in the Fall 2020 (26/10/2020 -14/12/2020), during the second national COVID-19-related lockdown; the third in the Summer 2021 (24/06/2021 -09/08/2021) after the end of the third national COVID-19related lockdown. Initially, 371,000 persons were randomly selected and 134,391 completed the first EPICOV study questionnaire (36.4% participation). Measures of mental health were exclusively included in an extended version of the study questionnaire completed by 10% of the study All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 March 30, 2022. ; https://doi.org/10.1101/2022.03.25.22272948 doi: medRxiv preprint sample followed-up longitudinally (n= 14,237 persons in the first study wave, n=12,519 in the second and n=10,780 in the third study wave). Since our aim was to study determinants of depression among young adults, statistical analyses were restricted to participants aged 18-30 years who had at least one measure of depression during the course of follow-up (n= 2,271 persons in the first, n=1,966 in the second and n=1,615 in the third study wave). Participants' aged 15-17 years at the time of the study were excluded because 95,57% were students, therefore there was no variability in labor force participation in this group. Additionally, to limit the influence of preexisting mental health difficulties on our results, we excluded study participants who had a prior history of depression or anxiety which limited their daily activities (n=50) (Figure 1) . In our study sample, 74.2% of participants completed the study questionnaire by CAWI and 25.8% by CATI. The primary outcome in our study was depression over the two weeks preceding the assessment, measured by the Patient Health Questionnaire -9 items (PHQ-9) score (18). Following previously published guidelines, participants with a score >= 10 were considered to have moderate to severe depression and were considered as cases. Participants' labor market participation was assessed at each study wave and treated as a time-varying exposure. We distinguished participants who reported that they were employed, studying, unemployed or out of the labor force (that is neither employed, unemployed, nor studying). Participants who reported studying and working were considered to be primarily students. Based on prior evidence, we selected several socio-demographic and health characteristics as possible confounding variables of the association between labor market participation and All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Paris and suburbs), lone living (yes vs. no), having children (yes vs. no), and presence of a chronic disease (yes vs. no, the most frequent diseases reported were asthma -4.6%, physical disability -1.1%, gastro-intestinal disorder -0.9%, musculo-skeletal disorder -0.9%, respiratory disorder -0.6%, diabetes -0.5%). Other characteristics were studied as timedependent variables: perceived financial situation (ascertained by the following question: 'Do you consider your financial situation to be -comfortable, suitable, tight, difficult?'), romantic relationship (yes vs. no), outings in preceding week (5 times), and experience of COVID-19 symptoms (yes vs. no, one or more symptoms among cough, fever, dyspnea, or ageusia, dysgeusia or anosmia). Reweighting and marginal calibrations were used to correct for non-participating bias and to take into account the effect of the means of data collection (Internet or telephone). To test the association between young adults' labor market participation and depression during the COVID-19 epidemic we proceeded as follows. First, associations were tested in bivariate Generalized Estimating Equations (GEE) (19) regression models weighted for study weights. Second, a multivariate GEE regression model was implemented including the study wave and all covariates associated with the study outcome with a p-value <0.20 in bivariate models. In subsequent analyses, we repeated the analyses stratified on age (18-24 / 25-30 years), 25 being the age when youths in France cannot anymore be claimed as dependents by their parents. Additionally, we tested statistical interactions between participants' labor market participation and sex. Missing data on covariates (maximum 37%) were imputed using multiple imputations under a missing at random hypothesis (20) . Data management and descriptive analyses were performed using R 4.0; weighted GEE models were implemented using SAS V9.4. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 March 30, 2022. ; 1 0 The EPICOV study received approval from an ethics committee (Comité de Protection des Personnes Sud Méditerranée III 2020-A01191-38) and from France's National Data Agency (Commission Nationale Informatique et Libertés, CNIL, MLD/MFI/AR205138). Across the three study waves, the weighted prevalence of depression was respectively 17.1%, 14.9% and 13.8%. As shown in (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In a longitudinal nationally representative study of 18-30-year-olds, we found that labor market participation is associated with depression during the course of the COVID-19 epidemic in France. Young adults are a known high-risk group in terms of mental health; our study indicates that students and persons who are unemployed are especially vulnerable. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. These results are robust to control for other characteristics associated with depression, highlighting the psychological needs of youths who are outside of the labor force and who should be the focus of psychological screening and intervention, particularly in the context of the COVID-19 pandemic. Our study has limitations which need to be acknowledged prior to interpreting the data. First, the EPICOV study started in the Spring of 2020 and we have only retrospective data on participants' history of pre-existing mental health problems. Nevertheless, we were able to exclude from the statistical analyses participants who had the most severe forms of psychopathology. While future studies should examine prospective changes in mental health among cohorts of young people followed since prior to the COVID-19 pandemic, our results do not appear to be due to pre-existing severe mental health difficulties. Second, the EPICOV cohort suffered attrition, with participants belonging to socioeconomically disadvantaged groups least likely to participate over the long term (17) . This lead us to conduct repeated measures analyses, which make it possible to maximize the use of information provided by study participants and limit possible biases due to attrition. Moreover, our statistical analyses are weighted to render the study sample representative of young adults in France. Third, in the EPICOV cohort, depression was ascertained with the PHQ-9 questionnaire, which is self-reported. While clinical diagnoses may yield a more accurate picture of severe depression, the PHQ-9 has satisfactory validity vs. clinical diagnosis (18) and we specifically studied a level of symptoms consistent with moderate to severe depression. Our study also has strengths which we would like to highlight. First, EPICOV is a nationallyrepresentative sample of the population residing in France and the results are generalizable to the whole country (15) . Second, to our knowledge, ours is one of few investigations in the context of the COVID-19 pandemic to directly compare young people in different labor market positions, including those who are out of employment and who are generally more All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 March 30, 2022. ; https://doi.org/10.1101/2022.03.25.22272948 doi: medRxiv preprint difficult to include in health surveys and cohort studies. Third, EPICOV participants were followed-up from the Spring 2020 to the Fall 2021, making it possible to study participants' mental health through different phases of the COVID-19 pandemic. Fourth, the use of online as well as telephone interviews and multiple call-backs helped include participants more reluctant to take part in health surveys in the study sample. Our results are in line with other studies which found elevated levels of depression and anxiety among university students (3) (4) (5) 21) . First of all, students are vulnerable because of high levels of financial difficulties as well social isolation, which increased during the course of the epidemic (21, 22) . Additionally, there is evidence that students have experienced specific issues during the course of the COVID-19 pandemic, such as worries related to their studies, difficulties finding motivation to study, missed learning opportunities (23) as well as increased conflicts with others and limited ability to study (24) , which could fuel fears about their future and negative mood. There is suggestion that distance learning is especially complex for students from disadvantaged backgrounds and negatively contributes to their stress levels (25) . Additionally, students are one of the groups -along with adolescents, in which use of screen-based media and devices increased particularly during the course of the pandemic (22). While this helped them stay in contact with friends and family and pursue coursework, inability to socialize and meet others probably hampered the ability to share worries and anxieties and seek social support. Mental health of young adults who are unemployed Young adults are especially likely to experience job instability which we found to be associated with depression, consistently with other studies (6, 7, 26) . Importantly, while persons who experience mental health difficulties may be especially likely to become unemployed (27) , there is also evidence that the experience of unemployment predicts later symptoms of depression and anxiety (28) . The COVID-19 epidemic, which had a major All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. impact on employment irrespective of individuals' characteristics, is an interesting set-up to examine the relationship between unemployment and mental health, as pre-existing psychological difficulties are likely to play a lesser role than prior to the pandemic. Evidence showing that employment loss during the course of the sanitary crisis predicts psychological difficulties suggests that labor market characteristics could directly impact individuals' mental health (10). Mechanisms that could contribute to elevated levels of depression among young adults who are unemployed include financial difficulties, stress associated with uncertainty about labor market prospects, as well as social isolation (8, 10) . One of the issues that remains unresolved at this point is whether depression which occurred among young adults during the COVID-19 pandemic will persist or predict other mental health difficulties over the longer-term. The COVID-19 pandemic which had massive effects on population mental health, highlighted the need to test and make available interventions to prevent and treat persons who experience psychological distress and depression. Young adults have elevated risks of psychological distress, yet they are also one of the groups with lowest levels of access to mental healthcare (29) . A recent umbrella review reported that psychosocial interventions as well as combined psychological and educational interventions lead by general practitioners are effective in preventing depression among youths (30) . In addition, there is evidence that psychological counselling provided in the university context can effectively alleviate symptoms of depression among students (31) . Moreover, mobile apps addressing signs of stress and psychological distress, which are effective and cheap, seem particularly wellsuited to young adults who widely use screen-based media (32, 33). Mobile-based interventions and telemedicine may also be effective to address the mental health needs of young adults who are unemployed, although it may be more difficult to identify and target them. Additionally, efforts aiming to favor employment among young adults contribute to good mental health (34) . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Young adults experience high levels of depression in the context of the COVID-19 pandemic, students and the unemployed being at higher risk than young adults who are employed. Our findings imply that young adults who are in training or looking for work should be the focus of attention in terms of prevention and mental health treatment. In particular, efforts should be deployed to improve access to effective psychosocial interventions and mental health treatment for persons who are unemployed. Moreover, strategies that enhance young people's chances of finding employment could have benefits in terms of mental health. Anonymous aggregated data for the first wave of the EPICOV study are available online. The EPICOV dataset is available for research purposes concerning the first round, and will be available by the end of March 2022 for the second wave for research purpose on CASD (https://www.casd.eu/), after approval from French Ethics and Regulatory Committee (Comité du Secret Statistique, CESREES and CNIL). Access to anonymized individual data may become available before the expected period on request from the corresponding author, conditional on approval from an ethical and regulatory body. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic Factors associated with psychological distress during the coronavirus disease 2019 (COVID-19) pandemic on the predominantly general population: A systematic review and meta-analysis Post-secondary student mental health during COVID-19: a meta-analysis Anxiety and depression in Chinese students during the COVID-19 pandemic: a meta-analysis. Frontiers in public health Impact of the COVID-19 pandemic on the mental health of college students: a systematic review and meta-analysis Estimating influences of unemployment and underemployment on mental health during the COVID-19 pandemic: who suffers the most? Public Health The psychological impact of the COVID-19 crisis is higher among young Swiss men with a lower socioeconomic status: Evidence from a cohort study Characteristics of those most vulnerable to employment changes during the COVID-19 pandemic: a nationally representative crosssectional study in Wales The mental health of young people who are not in education, employment, or training: a systematic review and metaanalysis Youth unemployment rate (indicator).2022 Accessed on Youth not in employment, education or training (NEET) (indicator) Accessed on The impact of COVID-19 on employment and jobs2021 CoviPrev : une enquête pour suivre l'évolution des comportements et de la santé mentale pendant l Une dégradation de la santé mentale chez les jeunes en 2020 Prevalence of SARS-Cov-2 antibodies and living conditions: the French national random population-based EPICOV cohort The PHQ-9: validity of a brief depression severity measure An overview of methods for the analysis of longitudinal data Multiple imputation and posterior simulation for multivariate missing data in longitudinal studies Understanding mental burden and factors associated with study worries among undergraduate medical students during the COVID-19 pandemic Stress and associated factors among French university students under the COVID-19 lockdown: The results of the PIMS-CoV 19 study Why lockdown and distance learning during the COVID-19 pandemic are likely to increase the social class achievement gap Psychological distress during the COVID-19 pandemic in France: a national assessment of at-risk populations. General psychiatry Unemployment among young people and mental health: A systematic review Lost in transition: a systematic review of the association between unemployment and mental health E-mental health care among young adults and help-seeking behaviors: a transversal study in a community sample Primary prevention of depression: An umbrella review of controlled interventions Psychological interventions for depression in Chinese university students: A systematic review and meta-analysis Effectiveness of mobile app-based psychological interventions for college students: a systematic review of the literature Digital mental health interventions for depression, anxiety, and enhancement of psychological wellbeing among college students: systematic review No reuse allowed without permission. (which was not certified by peer review) is the author/funder The authors want to thank EPICOV study participants and members of the EPICOV study group, who all had a significant contribution to this study and made it possible.