Prevalence and predictors of psychological distress before, during, and after a COVID-19 pandemic wave in Switzerland, 2021

There are concerns about acute and long-term mental health effects of the COVID-19 pandemic. This study examined the prevalence and predictors of psychological distress before, during, and after a pandemic wave in Switzerland, 2021. Prevalence of psychological distress was estimated in adults aged 35–96 years using the General Health Questionnaire-12 administered in June 2021 (Specchio-COVID19 cohort, N = 3965), and compared to values from 2003 to 2006 (CoLaus|PsyCoLaus cohort, N = 5667). Anxiety and depression were assessed from February to June 2021 using the Generalised Anxiety Disorder scale-2 and the Patient Health Questionnaire-2, respectively. Prevalence of psychological distress in June 2021, after the pandemic wave (16.0% [95% CI, 14.6%–17.4%]) was comparable to pre-pandemic levels (15.1% [14.0%–16.2%]). Anxiety and depression were highest at the start of the pandemic wave in February 2021, and declined from February to June with the relaxation of measures. Predictors of psychological distress included being younger, female, a single parent, unemployed, a change in working hours or job loss in the past 6 months, greater perceived severity and contagiousness of COVID-19, and self-reported post COVID-19. By June 2021, following a pandemic wave, prevalence of psychological distress in Switzerland was closer to pre-pandemic levels. These findings highlight the need for additional mental health support during times of stricter government policies relating to COVID-19; yet they also suggest that individuals can adapt relatively quickly to the changing context.


Introduction
There are concerns about acute and long-term mental health effects of the novel coronavirus  pandemic. Research in Europe, the US, and China demonstrated that mental health deteriorated in the early stages of the pandemic (Fiorillo et al., 2020;McGinty et al., 2020;Pierce et al., 2020;C. Wang et al., 2020). For example, prevalence of psychological distress in the UK rose from 18.9% in 2018-2019 to 27.3% in April 2020, one month into lockdown; a stronger increase than previous upward trends (Pierce et al., 2020).
Acute distress during the COVID-19 pandemic may lead to long-term mental health problems. Measures of psychological distress predict mental illness (Goldberg et al., 1997), and experiences during a pandemic can have long-term mental health effects (Brooks et al., 2020). For example, elevated levels of anxiety, depression, and post-traumatic stress symptoms among SARS patients, quarantined people, and healthcare workers during the 2003 SARS epidemic persisted years later (Lee et al., 2007;Liu et al., 2012;Maunder et al., 2006). Research suggests that COVID-19 can be considered a traumatic stressor event capable of eliciting traumatic stress reactions (Bridgland et al., 2021).
Research tracking changes in mental health during the early stages of the COVID-19 pandemic suggests mental health improves as measures are eased (Robinson et al., 2022). In England, anxiety and depression declined across the first 20 weeks following lockdown (March-August 2020), with the fastest decreases during lockdown easing measures (Fancourt et al., 2021). A similar trend was seen in Germany (March--June 2020; Bendau et al., 2021). In Switzerland, 2020 was marked by the first and second pandemic waves. A considerable proportion of Swiss participants reported that their mental health had been adversely affected during the first pandemic wave (Hernandez et al., 2021). Prevalence of distress was reported to increase further during the second wave of autumn and winter 2020, which coincided with a retightening of restrictions to contain the pandemic (de Quervain et al., 2020). Just one known previous study examined trajectories of mental health at a later stage of the pandemic in Switzerland (August 2020-May 2021). In a sample of 732 people living in the Ticino region of Switzerland, depression, anxiety, and stress increased from August 2020 to May 2021, with the greatest increases during the second wave of the pandemic, and little differences after the second wave (Piumatti et al., 2022). The third pandemic wave in early 2021 (February to May) was less severe than the first and second in terms of hospitalisations and deaths (Federal Office of Public Health, 2021). By June 2021, there were few cases, and government measures were less stringent than earlier in the year (University of Oxford, 2021). Mental health may have been generally better at this stage of the pandemic, although this has not yet been examined. As the situation has evolved, monitoring mental health throughout the pandemic, with comparisons to pre-pandemic data, is needed to inform future public health measures.
Pre-pandemic factors, such as demographics and life circumstances, likely remain important predictors of psychological distress throughout the pandemic. Being female and younger predicted psychological distress, and greater increases in distress, during the early stages of the pandemic (Fancourt et al., 2021;Kwong et al., 2021;Pierce et al., 2020). COVID-19-specific factors, such as a diagnosis of COVID-19 (Taquet et al., 2021), changes in work circumstances and unpaid care at home (McDowell et al., 2021;Xue and McMunn, 2021), high perceived risk of dying from COVID-19, and social distancing (Kämpfen et al., 2020), also contributed to people's mental health during the early stages of the pandemic.
The main aim of this study was to examine the prevalence of general psychological distress, as well as symptoms of anxiety and depression, during and following the third wave of the COVID-19 pandemic in Switzerland (February-June 2021), with comparisons to pre-pandemic levels of distress. A second aim was to examine a range of prepandemic and COVID-19-specific predictors of psychological distress at this stage of the pandemic.

Participants
Information on psychological distress and potential predictors during the pandemic was from Specchio-COVID19, a population-based digital study launched in December 2020 to follow up serosurvey participants in Geneva, Switzerland (Baysson et al., 2022). Serosurvey participants were randomly selected from the Bus Santé population-based study, and from Geneva registries (Stringhini et al., 2020;Stringhini et al., 2021). Adult serosurvey participants were invited to take part in the Specchio-COVID19 study after a baseline serologic test. From the 10 616 adult serosurvey participants invited, 6394 enrolled in Specchio-COVID19 (participation rate 60%, not taking into account participants unreachable owing to false email addresses). Non-specific psychological distress was assessed via questionnaire in June 2021. Anxiety and depression were assessed through monthly questionnaires administered from February to June 2021. Of the 6394 participants enrolled in Specchio-COVID19, 4636 individuals (73%; mean age = 51 years (range = 18-96 years)) completed the measure of psychological distress in June 2021 (see Supplementary Fig. S1).
This study was carried out in accordance with the latest version of the Declaration of Helsinki. The Specchio-COVID19 study was approved by the Cantonal Research Ethics Commission of Geneva (CCER project ID 2020-00881); the physical and psychiatric investigations of CoLaus| PsyCoLaus were approved by the Institutional Ethics Committee of the University of Lausanne (project reference numbers: 16/03, 33/09, 26/ 14 and 134/05, 239/09, respectively), which later became the Ethics Commission of Canton Vaud. All participants provided written informed consent.

Measures
Upon registration, Specchio-COVID19 participants completed a questionnaire assessing sociodemographic factors and pre-existing chronic disease. Self-reported SARS-CoV-2 infections, perceived COVID-19 severity and contagiousness, COVID-19-related quarantine, changes in work circumstances, and loneliness, were assessed through monthly questionnaires. Self-reported post COVID-19, need for psychological support, suicidal ideation, and psychosocial factors including perceived stress and social support were assessed by questionnaire in June 2021.

Mental health outcomes
In Specchio-COVID19 and CoLaus|PsyCoLaus, the 12-item General Health Questionnaire (GHQ-12) measured non-specific psychological distress (Goldberg et al., 1997;Lundin et al., 2016). Respondents rate the extent to which they have experienced each symptom in the past two weeks compared to their usual state. Responses for each symptom were dichotomised (0-0-1-1) and summed into an index (range 0-12). A score of 4 or more indicates clinically significant levels of psychological distress (Pierce et al., 2020).
In Specchio-COVID19, the General Anxiety Disorder-2 scale (GAD-2) and the Patient Health Questionnaire-2 (PHQ-2) assessed anxiety and depression. Respondents indicate how often they experienced each symptom over the past 2 weeks. Scores are summed (range 0-6); a total score of 3 or higher indicates probable depression or anxiety disorder (Kroenke et al., 2003(Kroenke et al., , 2007. The PHQ-2 and GAD-2 have acceptable sensitivity and specificity for any depressive and anxiety disorder, respectively, at a threshold of ≥3 (Plummer et al., 2016;Staples et al., 2019). In CoLaus|PsyCoLaus, depression and anxiety disorder were assessed using the Diagnostic Interview for Genetic Studies (Nurnberger et al., 1994). Depressive disorder included diagnosed current major depressive disorder or dysthymia. Anxiety disorder included diagnosed current generalised anxiety disorder, panic disorder, agoraphobia, social phobia, specific phobias, posttraumatic stress disorder, or obsessive compulsive disorder.
In Specchio-COVID19, participants were asked whether they had felt a new need for psychological help or to contact a listening and psychological support unit since the beginning of the pandemic (yes/no). Suicidal ideation over the past month was assessed using the following question from the Columbia-Suicide Severity Rating Scale (Posner et al., 2011): 'In the past month, have you wished you were dead or wished you could go to sleep and not wake up?' (yes/no).

Sociodemographic factors and living circumstances
Sociodemographic factors included biological sex (male, female), age (years), and ethnicity (white European or other). Education level was primary (none or compulsory education), secondary (high school diploma or vocational training), or tertiary (university level qualification). Employment status was employed, self-employed, retired, unemployed, or other economically inactive (not working and not looking for work, such as students, and people unable to work for health reasons or disability). Household income was categorised according to household composition, using information from the Cantonal Office of Statistics of Geneva for 2015-2017 (low (below the first quartile of the income distribution), medium (between the first and third quartiles), or high (above the third quartile)). Living arrangement was living alone, as a single parent, as a couple with children, as a couple without children, or with other adults. Household density was the ratio of people to bedrooms, and categorised as overcrowded (>2), or not overcrowded (≤2). Housing conditions were house with outdoor space, appartment with outdoor space, or appartment without outdoor space. Urbanicity was rural, urban outskirts, or urban.

Work conditions
Each month (January-June 2021), participants reported their employment situation, and whether their working conditions had changed. This information was used to create three binary variables: 1) change to telework in the last 6 months, 2) change in working hours (increased or decreased) in the last 6 months, 3) job loss in the last 6 months. For each variable, employees whose working conditions had not changed was the reference category.

Health and psychosocial factors
Participants were asked if they had any long-standing chronic illness, and to select their illness(es) from a list, which included mental health conditions. These responses were used to create a variable indicating presence or absence of a mental health condition. Participants were also asked if they were receiving any psychological support from a health professional (psychologist, psychiatrist, or psychotherapist) prior to the pandemic, and whether this support was maintained during the pandemic. This information was combined to create a variable with three categories: no psychological support prior to pandemic, psychological support prior to and during pandemic, and psychological support prior to but not during pandemic.
Loneliness, perceived stress, and social support were assessed using the 3-item Revised UCLA Loneliness Scale (Hughes et al., 2004), Perceived Stress Scale (Cohen, 1988), and 3-item Oslo Social Support Scale (Bøen et al., 2012), respectively. For each scale, higher scores indicate greater loneliness, perceived stress, and social support. Internal consistency of the scales was high (Cronbach's α range = 0.6-0.9).
Adverse life events in the past 6 months (yes/no) included death of a loved one, illness of a loved one, separation/divorce/family problems, violence at home, violence outside the home, job loss (own or partner), bankruptcy and/or financial difficulties, new or worsened diseases and accidents, and other events considered negative.

COVID-19-specific factors
To examine potential short-and longer-term effects of COVID-19 on mental health, previous SARS-CoV-2 infection (natural) was measured as: 1) a positive PCR test result in the last 6 months (January-June 2021), and 2) a positive serology test result since March 2020 (Stringhini et al., 2020;Stringhini et al., 2021a;Stringhini et al., 2021b). Other COVID-19-specific factors included self-reported post COVID ('are you currently or have you suffered from long-term COVID, that is, long-term manifestations of COVID-19 (such as symptoms that persist beyond three weeks)?' Yes/no), having to quarantine due to contact with a person who tested positive for SARS-CoV-2 in the last 6 months (yes/no), perceived severity of COVID-19, and perceived contagiousness of COVID-19 (1 = not at all, 5 = extremely).

Statistical analysis
Statistical analyses were conducted using R version 4.1.0 (R Foundation for Statistical Computing, Vienna, Austria). Proportions and 95% confidence intervals (CIs) were calculated for the measure of psychological distress in the Specchio-COVID19 and CoLaus|PsyCoLaus cohorts separately. Analyses of both data sets incorporated survey sampling weights (for age, sex, and education using statistics from the Swiss Federal Statistical Office) to generate nationally representative estimates of psychological distress. Mean scores, with standard deviation bounds, on the GAD-2, PHQ-2, and UCLA-Loneliness Scale were reported from February to June 2021 in the Specchio-COVID19 cohort. Supplementary analysis examined prevalence (proportions and 95% CIs) of anxiety and depression in the Specchio-COVID19 and CoLaus| PsyCoLaus cohorts. A second supplementary analysis examined prevalence of psychological distress, anxiety, and depression in CoLaus|Psy-CoLaus in winter and summer months separately.
Logistic regression was used to examine pre-pandemic and COVID-19-specific predictors of psychological distress (GHQ-12 score ≥4) in the total Specchio-COVID19 cohort (N = 4636). Univariable and minimised multivariable regression models were ran to identify groups at risk for psychological distress, which could be further examined in future research, for example in a structural equation model with more specific confounding structures. Age, sex, education level, and preexisting mental health condition were included as covariates in the multivariable regression models. Cluster robust standard errors were used to account for potentially nested data within households. Of the total sample, 1544 (33%) shared a household with another participant. The majority of these participants (90%) indicated that they were living as a couple, and were therefore biologically unrelated. A p value of <0.05 was considered statistically significant.

Sample characteristics
Characteristics of adults aged 35 years and above in the Specchio-COVID19 and CoLaus|PsyCoLaus cohorts are shown in Table 1. The cohort samples were similar in terms of age and sex distribution, employment status, and most were white Europeans. A greater proportion of Specchio-COVID19 participants were educated to tertiary level than CoLaus|PsyCoLaus participants (66% vs. 21%). Characteristics of the total Specchio-COVID19 sample, including potential predictors of psychological distress, are shown in Table 2. Participants were on average 51 years old (range 18-96 years), 57% were women, and 65% were educated to tertiary level. In comparison with the general population of Switzerland, there was an over-representation of 45-55 year olds, and those with tertiary level education (see Table S2). Compared with non-responders, participants who completed the measure of psychological distress in June 2021 were older (mean age 51 years vs. 43 years), more highly educated (4% vs. 7% had no formal education), had a higher income (18% vs. 23% had low income), a greater proportion were white Europeans (92% vs. 86%), a greater proportion were retired (22% vs 8%), and a smaller proportion were living as a couple with children (43% vs. 53%) (see Table S3). Depression and anxiety symptoms, as well as loneliness, declined from February-June 2021 in the total sample (see Fig. 2). Including all age groups, the weighted prevalence of probable depressive disorder in February 2021 was 13.5% (11.9%-15.2%; N = 3082); the weighted prevalence of probable anxiety disorder was 15.9% (14.2%-17.7%; N = 3082).

Prevalence of psychological distress
In June 2021, 12.6% (477/3771) reported they had felt a new need for psychological help or to contact a counseling and psychological support unit since the start of the pandemic. 5.3% (200/3771) reported having had suicidal thoughts in the past month; an additional 2.9% (110/3771) did not wish to respond.

Predictors of psychological distress in the Specchio-COVID19 cohort
Predictors of psychological distress in the total Specchio-COVID19 sample are shown in Table 3.
In the univariable models, female sex, non-white ethnicity, unemployment status or other economic inactivity, living in an apartment with or without access to outdoor space, and living alone, as a single parent, as a couple with children, or with others (not children) were demographic factors associated with greater risk of psychological distress (OR range = 1.20-2.54). Older age and retirement was associated with reduced risk of psychological distress. Pre-existing mental health condition, pre-pandemic psychological support, experiencing adverse life events in the past 6 months, greater loneliness, and greater perceived stress were health and psychosocial factors associated with greater risk of psychological distress (OR range = 1.27-2.81). Greater social support was associated with reduced risk of psychological distress. Experiencing a change in working hours in the past 6 months, experiencing a job loss in the past 6 months, greater perceived contagiousness of COVID-19, and self-reported post COVID-19 were COVID-19-specific factors associated with greater risk of psychological distress (OR range = 1.17-2.08).
In the multivariable models, all of the above, apart from non-white ethnicity, living in an apartment with or without access to outdoor space, living alone or as a couple with children, and stopping psychological support during the pandemic, remained statistically significant predictors of psychological distress.

Discussion
Overall prevalence of psychological distress among adults aged 35 years and above in June 2021, following a COVID-19 pandemic wave in Switzerland, was comparable to pre-pandemic levels in 2003-2006. Depression, anxiety, and loneliness declined from February to June 2021, along with the progressive relaxation of containment measures in Switzerland. Within the sampled period (February-June 2021), the highest prevalence of probable depression and anxiety disorder was in February 2021, at the start of the wave. Mental health was closer to prepandemic levels by the end of the wave in June 2021, but a fair proportion of participants reported a new need for psychological support since the start of the pandemic. Pre-pandemic and COVID-19-specific factors predicted psychological distress. These findings highlight the need for continued monitoring of mental health as the pandemic evolves, alongside efforts to support vulnerable groups.

Prevalence of psychological distress
In Europe, the US, and China, prevalence of psychological distress during the early stages of the pandemic was higher than before the pandemic (McGinty et al., 2020;Pierce et al., 2020;C. Wang et al., 2020). The uncertainty and sudden change to everyday life, reduced social contact, and health concerns explained this early increase in distress (Robinson and Daly, 2021). The comparatively lower rates of psychological distress seen in June 2021 in the present study, alongside the decline in depression, anxiety, and loneliness, from February to June, coincides with the end of the third wave (Federal Office of Public Health), and a relaxation of government policies in Switzerland (Oxford COVID-19 Government Response Tracker). By the end of the third wave, 66% of the Geneva population had developed SARS-CoV-2 antibodies after vaccination and/or infection (Stringhini et al., 2021b), there was some resumption to pre-pandemic life, and perceived health-related threats may have subsided. These findings concur with research  during the early stages of the COVID-19 pandemic, which found mental health improved as measures were eased (Robinson et al., 2022), and suggests overall there has been resilience in mental health. Before and during the pandemic, prevalence of psychological distress was highest among women, and lowest among older adults (≥65 years). These findings concur with those of the latest Swiss Health Survey in 2017 (Schuler et al., 2020). In the Specchio-COVID19 cohort, 30% of 18-34 year olds reported psychological distress in June 2021. In the Swiss Health Survey, 22% of 15-34 year olds in Geneva reported psychological distress (Zufferey, 2020). Although it is not possible to draw firm conclusions due to differences in the age groups, and in the measures used, the higher prevalence of distress among younger participants in Specchio-COVID19 suggests younger people may still be disproportionately affected by the pandemic (Pierce et al., 2020).

Predictors of psychological distress in the Specchio-COVID19 cohort
Additional pre-pandemic predictors of psychological distress in June 2021 included unemployment, being a single parent, having a preexisting mental health condition, experiencing adverse life events in the past 6 months, greater loneliness, and greater perceived stress. Retirement and greater social support were associated with reduced risk of psychological distress. These findings corroborate and extend research on psychological distress during the early stages of the pandemic (Fancourt et al., 2021;Kwong et al., 2021;Pierce et al., 2020), and highlight the persistence of social inequalities in mental health. Ensuring safe access to essential services (such as childcare), financial security (for example through government investment in debt respite), free or heavily subsidised mental health services (including online services, where appropriate), and mental health promotion within the community are ways in which vulnerable groups can be supported during and following the pandemic (Aknin et al., 2022;McDaid, 2021). Mental health care providers also play a key role in responding to their patients concerns related to COVID-19 (Ghebreyesus, 2020).
COVID-19-specific predictors of psychological distress in June 2021 included a change in working hours or job loss in the past 6 months, greater perceived severity and contagiousness of COVID-19, and selfreported post COVID-19. Previous research found job loss, but not switching to remote working in April 2020, predicted poorer mental health in US adults (McDowell et al., 2021). The present study corroborates these findings during a pandemic wave in 2021, and confirms that a change in working hours (increase or decrease) may be detrimental for mental health. The impact of changing work circumstances on mental health likely depends on the duration of change, as well as the individual's personal circumstances, and requires further investigation. Some families may benefit from the flexibility of home working, while others may feel isolated (Birimoglu Okuyan and Begen, 2022).
Worries about the risk of infection and dying from COVID-19 have previously been associated with poorer mental health (Kämpfen et al., 2020). Worries about COVID-19 may worsen mental health, and poorer mental health may lead to rumination and worries about COVID-19. Self-reported post COVID-19, but not previous SARS-CoV-2 infection, predicted psychological distress in the present study. A large-scale US study including 62 ′ 354 COVID-19 cases found that a COVID-19 diagnosis was associated with increased risk of psychiatric disorder during the first 14-90 days after diagnosis (Taquet et al., 2021). The present study included 258 cases with a positive PCR/rapid antigen test in the last 6 months, and may not have been sufficiently powered to detect an association, other than in more severe cases with self-reported post COVID-19. Associations between COVID-19 and psychological distress are bidirectional, and could be mediated by biological and behavioural factors (Mazza et al., 2020;Taquet et al., 2021;Wu et al., 2020).
There was no association between quarantine (due to contact with a person who tested positive for SARS-CoV-2) in the past 6 months and psychological distress. This finding contrasts with previously reported adverse effects of quarantine during the COVID-19 pandemic (Shi et al., 2020;Yunhe Wang et al., 2021). However, null associations between quarantine and mental health have been reported (Yongguang Wang et al., 2011). The conditions and duration of quarantine, individual circumstances, and COVID-19-related attitudes play an important role (Brooks et al., 2020;Shi et al., 2020;Yunhe Wang et al., 2021). For example, quarantine is most strongly associated with psychological distress among vulnerable groups (Yunhe Wang et al., 2021); and centralised quarantine in an unfamiliar environment is more strongly associated with adverse outcomes (Shi et al., 2020). Further research is needed to assess the longer-term effects of quarantine during the COVID-19 pandemic.

Strengths and limitations
Study strengths include the population-level, longitudinal data from two cohorts, and range of potential predictors of psychological distress. The GHQ-12, GAD-2, and PHQ-2 are validated self-report measures, and correlate strongly with clinically diagnosed psychiatric disorder. However, they are not clinical assessments and do not capture all forms of mental distress. The comparison of psychological distress in the pre- pandemic and Specchio-COVID19 cohorts was limited to adults aged 35 years and above, and cannot be generalised to younger age groups, who may be particularly impacted by the pandemic (Pierce et al., 2020). The pre-pandemic measure of psychological distress was assessed in 2003-2006, many years before the pandemic. Although various aspects of society have changed since this time, some studies found no secular change in the prevalence of anxiety and depression from the early 2000s until prior to the pandemic (e.g. Baxter et al., 2014;Bebbington and McManus, 2020). Other studies have reported an increase in the prevalence of anxiety and depression, especially in young adults aged 18-35 years (Gagné et al., 2022;Goodwin et al., 2020). Given that previous research suggests anxiety and depression may have remained stable or increased over time prior to the pandemic, we can be confident that the prevalence of psychological distress that we measured one year into the pandemic was not higher than pre-pandemic estimates. The decline in anxiety and depression in Specchio-COVID19 was observed from winter to summer. However, the results suggest effects greater than usual seasonal variations (Winthorst et al., 2011); and supplementary analysis in CoLaus|PsyCoLaus showed no difference in the prevalence of psychological distress, anxiety, and depression in winter and summer months. Specchio-COVID19 participants had a higher education level than CoLaus|PsyCoLaus participants. Although weights were applied to make the sample representative of the Swiss population, it is still possible that there was some selection bias due to other factors associated with survey participation not accounted for by weighting (e.g. topic interest). The data were from French-speaking regions in Switzerland, and vulnerable individuals with poor mental health are less likely to  participate in surveys; therefore, the findings may not generalise to all regions in Switzerland, or to more vulnerable groups. During the development of the Specchio-COVID19 study, post COVID-19 was defined as symptoms lasting longer than 3 weeks, but it is currently defined as usually occurring 3 months from the onset of COVID-19, with symptoms lasting for at least 2 months (Soriano et al., 2021). Further research is therefore needed to examine associations between post-COVID19 and mental health. Since June 2021, there have been further pandemic waves, containment measures, and the recent lifting of measures in March 2022. It will be important to examine mental health during this time, as well as potential longer-term effects in vulnerable groups.

Practical implications and conclusions
Acute distress during the COVID-19 pandemic may lead to long-term mental health problems, therefore it is important to monitor how mental health, including symptoms of psychological distress evolve during the pandemic, as well as predictors of distress. This study suggests that psychological distress was highest at the start of the third pandemic wave in Switzerland, but by the end of the wave in June 2021, overall prevalence was comparable with pre-pandemic levels. These findings highlight the need for additional mental health support during times of stricter government policies relating to COVID-19; yet these data also suggest that individuals can adapt relatively quickly to the changing context. This study also identified a range of pre-pandemic and COVID-19-specific predictors of distress at the end of the third wave, highlighting the persistence of inequalities in mental health. These findings therefore highlight the need for ongoing monitoring of mental health as the pandemic evolves, with efforts from the government and healthcare practitioners to support vulnerable groups.

Data availability statement
The data are available on request from the corresponding author. The ethics protocols under which the data were collected do not permit public data deposition.

Role of the funding source
The Specchio-COVID19 study is funded by the Swiss Federal Office of Public Health, the General Directorate of Health of the Department of Safety, Employment and Health of the canton of Geneva, the Private Foundation of the Geneva University Hospitals, the Swiss School of Public Health (Corona Immunitas Research Program), and the Fondation des Grangettes. The CoLaus|PsyCoLaus study is supported by research grants from GlaxoSmithKline, the Faculty of Biology and Medicine of Lausanne, the Swiss National Science Foundation (grants 3200B0-105993, 3200B0-118308, 33CSCO-122661, 33CS30-139468, 33CS30-148401, 33CS30_177535 and 3247730_204523), and the Swiss Personalized Health Network (project: Swiss Aging Citizen Reference). The funders had no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the article for publication.

Declaration of competing interest
None.