The impact of the prolonged COVID-19 pandemic on stress resilience and mental health: A critical review across waves

The global public health crisis caused by COVID-19 has lasted longer than many of us would have hoped and expected. With its high uncertainty and limited control, the COVID-19 pandemic has undoubtedly asked a lot from all of us. One important central question is: how resilient have we proved in face of the unprecedented and prolonged coronavirus pandemic? There is a vast and rapidly growing literature that has examined the impact of the pandemic on mental health both on the shorter (2020) and longer (2021) term. This not only concerns pandemic-related effects on resilience in the general population, but also how the pandemic has challenged stress resilience and mental health outcomes across more specific vulnerable population groups: patients with a psychiatric disorder, COVID-19 diagnosed patients, health care workers, children and adolescents, pregnant women, and elderly people. It is challenging to keep up to date with, and interpret, this rapidly increasing scientific literature. In this review, we provide a critical overview on how the COVID-19 pandemic has impacted mental health and how human stress resilience has been shaped by the pandemic on the shorter and longer term. The vast literature is dominated by a wealth of data which are, however, not always of the highest quality and heavily depend on online and self-report surveys. Nevertheless, it appears that we have proven surprisingly resilient over time, with fast recovery from COVID-19 measures. Still, vulnerable groups such as adolescents and health care personnel that have been severely impacted by the COVID-19 pandemic do exist. Large interindividual differences exist, and for future pandemics there is a clear need to comprehensively and integratively assess resilience from the start to provide personalized help and interventions tailored to the specific needs for vulnerable groups.

• The early stages of the COVID-19 pandemic were often associated with increased levels of distress and depressive and anxiety symptoms in the general population. • A substantial group of individuals has been either largely unaffected or is even doing better during the pandemic. • Longitudinal follow up showed remarkable signs of resilience. • Health care workers appear to be at an increased risk of stress-related psychological symptoms. • The mental health of children, adolescents, and students has been particularly affected by the pandemic. • Elderly people are more vulnerable to the physical effects of COVID-19, but also report lower psychopathology during the pandemic.
• Individuals with an existing psychiatric disorder are experiencing detrimental impact on their mental health from the COVID-19 pandemic, but do not seem to have further increased symptom severity compared with their pre-pandemic levels. • A high risk exists for psychiatric sequelae following a COVID-19 infection. • Many methodological shortcomings occur in the current literature which is often cross-sectional and relies on self-report, and it is moreover hard to directly compare results across many of the studies. • There is an urgent need for a personalized approach when it comes to identifying individuals at risk or resilient for the stressful effects of the COVID-19 pandemic. • The effects of stress and the resilience capacity are dependent on (neuro)biological, psychological, and environmental factors and also are heavily dependent on an individual's unique context.

Introduction
In 2020 many of us had hoped that the COVID-19 pandemic would be over in 2021, and that normal life would have resumed. The situation is clearly different: the pandemic is still ongoing with novel and more contagious variants leading to increased infection rates across the globe, with consequently more stringent restrictions in social interactions and more lockdowns. By the spring of 2021, over 156 million confirmed cases and more than 3.2 million deaths of COVID-19 have been reported ( https://covid19.who.int/ ), with health care systems worldwide being overburdened at certain times. As in 2020, summer was expected to be associated to a reduced impact of the pandemic. However, new variants of the virus emerged, such as the Delta variant and there is still concern for what will occur during fall and winter. The pandemic and the accompanying measures have led to changes in people's daily routines, limited social interactions, as well as formed tensions among families in lockdown together, and fear of getting ill and/or spreading the virus. At the same time, the prospect of mass vaccination efforts has given rise to hope. Undoubtedly, the pandemic has asked a lot from all of us given the high uncertainty and limited control over the situation. For mental health professionals, the key questions are: what are the effects of the COVID-19 pandemic on mental health, and what have we learnt from this unprecedented and prolonged pandemic regarding resilience at the individual and societal level? During the first wave of the pandemic, our Thematic Working Group on Resilience from the European College for Neuropsychopharmacology (ECNP) wrote an overview of stress resilience during the early stages of the pandemic ( Vinkers et al., 2020 ). Now, almost a year later, there has been a second wave and, in many countries, a third wave. Numerous studies have attempted to identify how the stressful pandemic has impacted mental health in the shorter and longer term across a wide range of populations. In fact, by April 2021, a staggering 120,000 publications on COVID-19 had appeared, with over 5,000 dealing with the impact of the pandemic on mental health and how stress resilience is shaped during the prolonged COVID-19 pandemic. Given that the pandemic is continuing, and there is the risk of future outbreaks, it is timely to consider its impact on mental health and factors that are linked to resilience against mental illness to guide the ongoing response to it. In view of this, we aim to provide a critical overview of how the pandemic has affected mental health in general, and how human stress resilience has shaped its impact on the shorter and longer term. Moreover, we aim to summarize whether there are specific effects of the COVID-19 pandemic on stress resilience across groups that may be more vulnerable (such as health care workers and adolescents), and what we can learn for possible future pandemics. In this selective review, we did not apply a systematic approach but rather used a targeted Medline search strategy related to COVID-19 topics complemented with a thorough search of references in key publications. A general premise should be made in the interpretation of the results of this review as several caveats impact on the interpretation of the evidence here summarized. First, most of the studies were conducted in the first wave of the pandemic, often with limited duration of follow-up, and are based on cross-sectional as-sessments and online surveys. Secondly, and possibly more importantly, the impact of the pandemic (as measured in terms of infections and death rates) as well as lockdown measures varied significantly from country to country. However, although these methodological limitations could reduce the robustness of the findings as well as their comparability among different countries, we believe this data synthesis might guide the reader in interpreting the impact of the pandemic on mental health and the modulating role of resilience.

The general population
The pandemic has affected almost every individual directly or indirectly, either due to (or fear of) COVID-19 infection, or because of the effects of far-reaching measures and their economic and social impact. Consequently, the impact of the pandemic on mental health outcomes has been frequently examined in the general population. In a study conducted in April 2020, using a probability sample ( N = 1468) and the Kessler-6 psychological distress scale (0-24 with 13 as a cutoff for serious distress), 13.6% of US adults reported symptoms of serious psychological distress, relative to 3.9% in 2018 ( McGinty et al., 2020 ). In another study among 9565 individuals from 78 countries, during the height of the lockdown (April -June 2020), the pandemic was experienced as at least moderately stressful for most people, and 11% reported the highest levels of stress. Symptoms of depression were also high, including 25% of the sample indicating that the things they did were not reinforcing, 33% reporting high levels of boredom, and nearly 50% indicating they wasted a lot of time ( Gloster et al., 2020 ). In a similar study which also used the Kessler-6 psychological distress scale ( N = 2555), distress levels did not increase, with equal numbers of US adults experienced serious psychological distress in February 2019 (prior to the pandemic) as in May 2020 ( Breslau et al., 2021 ). In the UK Household Longitudinal Study (UKHLS,N = 17,452), mental health was assessed with the 12-item General Health Questionnaire (GHQ-12) before and during the pandemic ( Pierce et al., 2020 ). In this study, the population prevalence of clinically significant levels of mental distress rose from 19% in 2018-19 to 27% in April 2020 (1 month into the UK lockdown). In a later report on the UKHLS, most individuals had either consistently good (39% of the participants) or consistently very good (38%) mental health across the first 6 months of the pandemic ( Pierce et al., 2021 ). A recovering group (12%) showed worsened mental health during the initial shock of the pandemic and then returned to pre-pandemic levels of mental health. The two remaining groups were characterized by poor mental health throughout the observation period, either with initial but sustained worsening in mental health (4%) or a steady and sustained decline in mental health over time (7%). Concerning major affective disorders, a nationally representative survey study of US adults (March -April 2020, N = 1,441) showed that the prevalence of depressive symptoms was more than three-fold higher during the COVID-19 pandemic when compared to the pre-pandemic prevalence (2017 -2018) ( Ettman et al., 2020 ). Indeed, a recent quantitative data synthesis, conducted by the Global Burden of Disease (GBD) Resource center, has shown that the pandemic has impacted substantially on the risk of major depressive disorder and anxiety disorders, estimating an additional 53.2 million cases of major depressive disorder and an additional 76.2 million cases of anxiety disorders globally due to the COVID-19 pandemic ( COVID-19 Mental Disorders Collaborators, 2021 ;Santomauro et al., 2021). Further support for the link between the COVID-19 pandemic and the onset of affective disturbances comes from the large cohort study of Lob et al. (March -April 2020, N = 51,417), showing that severe depressive symptoms were developed by 11% of their sample equaling 5656 participants, while moderate symptoms were experienced by 29% of the subjects during the COVID-19 pandemic ( Iob et al., 2020a ). In Italy ( N = 130), quarantine resulted in increased internalizing symptoms, particularly in those individuals with pre-existing psychopathology or experiencing negative economic consequences ( Castellini et al., 2021 ). In another study conducted in the US (March -June 2020, N = 7138), both increases and decreases in distress during the pandemic, assessed with the Patient Health Questionnaire-4 (PHQ-4), could be explained by perceived infection risk and risk of death, perceived financial risks, lifestyle changes resulting from the virus, perceived discrimination, and changes in substance use and employment status (explained variation: 70% for the increase in distress between March and April 2020, and 46% for the decline in April and June 2020) . Among Chinese students ( N = 68,685), levels of stress decreased after remission of the first outbreak (end of March -beginning of April), even though anxiety symptoms (22 to 26%) and depressive symptoms (11 to 15%) still increased, particularly in those with limited physical exercise and perceived social support ( Li et al., 2021b ). This indicates that the pandemic may have longer-lasting negative effects on mental health outcomes that might take time to fully emerge. Finally, a survey performed in 1,310 Spanish adults during the first lock-down period (March 2020) showed that regression models containing a series of variables (i.e. being female, having a younger age, having negative self-perceptions about aging, more time being exposed to news about COVID-19, having more contact with relatives different to those that participants co-reside with) explained 48% and 33% of the variance of distress and loneliness respectively ( Losada-Baltar et al., 2021 ). Although the methodological quality of studies was quite heterogeneous, with sometimes limited sample size, no longitudinal measures, and only sparse information about mental health status, adaptive responses after the first wave of the pandemic were frequently reported. In contrast to studies showing increases in mental health problems in the general population, there is also convincing evidence that most individuals are sufficiently able to cope with the pandemic and its associated measures and even that increased resilience building in the general population may have occurred. In the Netherlands, for example, a longitudinal study among adults ( N = 3,983) showed no apparent increase was found in anxiety and depression symptoms between March 2019 and March 2020 when the pandemic broke out ( van der Velden et al., 2020 ). Moreover, in a Chinese general population study ( N = 1738), which was conducted during the initial outbreak (end of January -beginning of February 2020) and the epidemic's peak four weeks later, there were no significant longi-tudinal changes in stress, anxiety and depression levels . In Switzerland, a survey study ( N = 10,472) documented increased stress in 50% of participants, but 24% showed no change and 26% even felt less stressed during the lockdown in April 2020 compared to the pre-pandemic period ( https://osf.io/jqw6a/ ). More encouraging news stems from the UCL COVID-19 Social Study showing that, between March and August 2020, in over 36,500 adults, the highest levels of depression and anxiety occurred in the early stages of lockdown but declined fairly rapidly as individuals adapted to the changing circumstances, even though the sample is not representative of the national population ( Fancourt et al., 2021 ). In Spain ( N = 3480), after the confinement was lifted, depressive symptoms rapidly decreased after an initial increase during the confinement, but no clear effects on anxiety were found ( Gonzalez-Sanguino et al., 2020 ). In Germany, worrying and depressive symptoms among the general population ( N = 2376) decreased on average between March and June 2020 ( Bendau et al., 2020b ). Another US sample ( N = 7319) found increased psychological distress (PHQ-4) between March and April 2020 as the COVID-19 crisis emerged and lockdown restrictions began, which subsequently declined to mid-March (baseline) levels by June 2020 . In 1166 UK adults, a low-stable profile characterized by little-to-no psychological distress was the most common trajectory for both anxiety-depression and COVID-19-related PTSD ( Shevlin et al., 2021 ). Assessment of loneliness in 1545 American adults in January, March, and April 2020 showed no significant changes in loneliness but rather increased perceived support from others ( Luchetti et al., 2020 ).
Conclusion: In the general population, the early stages of the COVID-19 pandemic were often associated with increased levels of distress and depressive and anxiety symptoms. However, the effects of the pandemic on mental health in the general population have been quite heterogeneous from the beginning, and a substantial group has been either largely unaffected or is even doing better during the pandemic period. More importantly, longitudinal follow up has shown signs of resilience in the general population, with surprising ability to bounce back and adapt. For an overview of findings from cross-sectional and longitudinal studies on the impact of COVID-19 on stress resilience and mental health in the general population, see Table 1 .

Health care personnel
Health care personnel have been particularly affected by the COVID-19 pandemic and exposed to more stressful circumstances than many other professional groups. This is due to several factors, including the increased infection risk, fear of infecting other people and being isolated from their families, working overtime, demanding work conditions with lengthy shifts, directly witnessing the suffering and death of patients, and witnessing the crowded and challenging situations in hospitals. Hence, it is not surprising that many studies on stress resilience and mental health outcomes during the COVID-19 pandemic have specifically focused on health care personnel. A meta-analysis of COVID-19-related stress and psychiatric symptoms in nurses al- Prevalence of depressive symptoms was more than 3-fold higher during COVID-19 (8.5% before COVID-19 and 27.8% during COVID-19). Being exposed to more stressors and low income was associated with greater odds of depressive symptoms.
( continued on next page )   Maqbali et al., 2021 ). In this metaanalysis, over one third of over 90,000 nurses reported stress, sleep disturbances and increased mood and anxiety symptoms. This seems considerably higher than findings from studies in nurses working during smaller-scale pandemics like SARS or when compared to the general population at the same time period ( Chen et al., 2005 ). A similar picture arises from other studies. A survey of dental academics across 28 countries (March -May 2020, N = 1862) indicated considerable psychological impact of the COVID-19 pandemic with significantly increased worries and altered individual behavior ( Ammar et al., 2020 ). Moreover, in a Spanish cohort of health care workers (April 2020, N = 1422) over half of the participants reported symptoms of posttraumatic stress disorder (PTSD) and anxiety disorders, and nearly 50% reported symptoms of depression, with women and younger people showing an even higher risk ( Luceno-Moreno et al., 2020 ). A study of medical staff in China (February -March 2020, N = 899) indicated a significantly increased prevalence of psychiatric symptoms such as depression, anxiety and insomnia compared to the general population ( Liang et al., 2020 ). A longitudinal study among Japanese adults (March 2020 and May 2020, N = 1015) showed that indices of fatigue, anxiety and depression increased among health care compared to non-health care workers during the COVID-19 outbreak ( Sasaki et al., 2020 ). In Portugal, a cross-sectional study (May 2020) showed that physicians working at the frontline of COVID-19 ( N = 420) presented worse mental health outcomes (anxiety, depression, stress and obsessive-compulsive symptoms) than other physicians. Moreover, this study found that being female and working at the frontline are risk factors for increased stress, while having a garden at home was a protective factor for anxiety and stress symptoms ( Ferreira et al., 2021 ). In Turkey, when 939 health care workers were assessed cross-sectionally in April-May 2020, more than 60% of the participants reported anxiety and depression symptoms ( Sahin et al., 2020 ). These studies stress the need for successful intervention or prevention strategies for health care personnel. To avoid long-term effects of stress, strategies to counteract the negative impact of the COVID-19 pandemic on mental health, particularly in highly affected populations, may be helpful. For instance, in a group of Italian general practitioners ( N = 102), Di Monte and colleagues found that the implementation of task-orientated project management seemed protective against symptoms of burnout during the pandemic ( Di Monte et al., 2020 ). In another study, specific pandemic-related stress factors were identified in health care personnel, including workload burden and fear of infection ( Mosheva et al., 2020 ). Heath and colleagues reviewed several strategies to increase resilience among health care workers during and after the COVID-19 pandemic, also guided by the experiences of previous pandemics ( Heath et al., 2020 ). They identified several interventions and approaches, ranging from individual strategies without professional help (e.g. increased selfcare or mindfulness practice) to strategies implemented at the group or organizational level (e.g. competency / resilience training, availability of psychological first aid or implementation of effective leadership and organizational justice). Importantly, some of these intervention strategies, including computer-assisted resilience training, have already been developed and successfully tested ( Aiello et al., 2011 ;Maunder et al., 2010 ;Weerkamp-Bartholomeus et al., 2020 ). Conclusion: Overall, health care workers appear to be at an increased risk of stress-related psychological symptoms during a pandemic, compared to the general population. However, longitudinal studies are still largely lacking, and it remains to be seen whether the increase in symptoms is transient and can be considered a normal response to an abnormal temporally limited situation. For more solid conclusions, large-scale prospective longitudinal studies on the specific risk of health care personnel during and after a pandemic are needed. Such studies are already being planned ( Roberts et al., 2020 ). This is particularly relevant as the COVID-19 pandemic has been present over a prolonged period and already spans several waves of infection. Importantly, specific prevention and intervention strategies at the individual as well as at the organizational level may be crucial, with studies already showing beneficial effects of these strategies. See Table 2 for an overview of findings from cross-sectional and longitudinal studies on the impact of COVID-19 on stress resilience and mental health in health care personnel.

Children, adolescents, and college students
It is plausible that the impact of the COVID-19 pandemic on mental health might vary as a function of age and levels of educational attainment. In this section, we present findings focusing first on the effects of the pandemic on mental health in children and their parents, and subsequently we discuss findings in studies performed in adolescents and college students.

Children and their parents
Due to measures to decrease the spread of the COVID-19 virus, schools have been closed and the opportunity to interact with peers, play outdoors and exercise decreased ( de Lannoy et al., 2020 ;Moore et al., 2020 ), whereas sleep and screen time have increased . Parental stress significantly increased after the school closures ( Hiraoka and Tomoda, 2020 ). Parents experienced financial and health problems and needed to comply with home schooling in addition to their own responsibilities, which affected parental routines. Parents' financial concerns increased verbal aggression, increased loneliness was associated with child neglect, whereas worries increased physical abuse of the children. In general, wellbeing of the parents significantly affected children's mental health ( Carroll et al., 2020 ). A national survey from the US in June 2020 ( N = 1011) reported worsening of mental health for 26.9% of the parents and 14.3% of the children, whereas the reported effect on physical health was smaller. Lost regular health care and delay in health care visits added to the worsening of children's mental health ( Patrick et al., 2020 ). In Italy (April -May 2020, N = 463) ( Cusinato et al., 2020 ) and in Spain (April 2020, N = 1049) Overall, 30.43%, 20.29%, and 14.49% of frontline medical workers in Hubei Province and 23.13%, 13.14%, and 10.64% of frontline medical workers in other regions reported symptoms of depression, anxiety, and insomnia, respectively. In addition, 23.33%, 16.67%, and 6.67% of the general population in Hubei Province and 18.25%, 9.22%, and 7.17% of the general population in other regions reported symptoms of depression, anxiety, and insomnia, respectively..  The COVID-19 emergency had a significant impact on GPs' work Implementing task-oriented problem management, rather than emotional strategies, appears to protect against burnout in these circumstances.

Mosheva et al., 2020
Israeli physicians Computer-assisted resilience training in healthcare workers appears to be of significant benefit and merits further study under pandemic conditions. Comparing three "doses" of the course suggested that the medium course was optimal. The proportion of participants who felt better able to cope after the session (76%) was significantly higher than the proportion who felt prepared to deal confidently with the pandemic before the session (35%).
Ten key themes emerged from a qualitative analysis of written comments, including family-work balance, antiviral prophylaxis, and mistrust or fear towards health care workers  , parental stress was associated with more conduct problems of their children. In Hong Kong (March 2020, N = 29,202 families), children with special education needs, chronic illnesses, mothers with mental illness, and children coming from single-parent and low income families were reported as more vulnerable to the increased stress levels as a result of the pandemic ( Tso et al., 2020 ). In Brazil (April -May 2020, N = 289), parents reported anxiety in 19.4% of their children. In particular, children living with a person other than their parents reported higher anxiety levels, especially when a guardian had a lower age and lower educational level ( Garcia de Avila et al., 2020 ). Having a house without an outdoor exit to a garden or terrace significantly predicted psychiatric symptomatology ( Francisco et al., 2020 ). Overall, there are concerns that the risk of child maltreatment might be heightened during the COVID-19 pandemic due to a range of stressors such as increased loneliness, reduced physical activity, economic stress, social distancing, homeschooling, marital conflicts and violence, and intensified childparent relationships ( Katz et al., 2020 ). In a recent study of 169 preschoolers, increased depressive and externalizing symptoms were observed compared to pre-COVID levels. A structured, predicable home environment adherent to family routines appears to mitigate these adverse effects ( Glynn et al., 2021 ).

Adolescents
In adolescents, determinants of resilience are less related to the parental situation and stressors associated with parental stress. A large-scale cross-sectional study from China (during the COVID-19 outbreak, N = 3613) revealed that among 3254 adolescents, anxiety and depressive symptoms were common during the COVID-19 pandemic ( Duan et al., 2020 ), as assessed by the Chinese version of the Spence Child Anxiety Scale (SCAS). In another crosssectional study conducted in China (February -March 2020, N = 1784), following almost 30 days of confinement, depressive and anxiety symptoms were reported in around 23% of adolescents, with an association with less optimism about the pandemic and with being more worried about being infected with COVID-19 ( Xie et al., 2020 ). A Swiss survey study (November 2020, during the second pandemic wave, N = 11,612) found that the youngest age group (14-24 years old, including students) were at higher risk for moderate-severe depressive symptoms (PHQ-9 ≥ 15) compared to individuals over 24 years old, and that this agedependent effect became more prominent over the course of the pandemic ( https://osf.io/6cseh/ ). A cross-sectional study from China (March -April 2020, N = 7890) reported a prevalence of 21.7% for anxiety and 24.6% for depression symptoms (HADS subscale score > 7) ( Li et al., 2021a ). Another study from China (March 2020, N = 8079) revealed that the prevalence of mild to severe depressive symptoms was 43.7% (assessed by the PHQ-9) and anxiety symptoms were 37.4% (assessed by the GAD-7) . Being in senior high school , female gender   , lack of physical exercise , and less social support in-creased the risk for depressive and anxiety symptomatology ( Qi et al., 2020 ). Furthermore, adverse childhood experiences, being exposed to COVID-19 and the presence of fear of exposure to COVID-19 were predictive of elevated levels of PTSD and anxiety in a Chinese cross-sectional cohort (February 2020, N = 6196) . Moreover, a longitudinal study among Australian adolescents ( N = 248) showed that adolescents experienced significant increases in depressive and anxiety symptoms and a decrease in life satisfaction during the epidemic compared to 12 months leading up to the COVID-19 outbreak. COVID-19 related worries, online learning difficulties, and increased conflict with parents negatively affected the mental health outcomes, whereas adherence to lockdown measures and feeling socially connected during lockdown were protective factors ( Magson et al., 2020 ). In Indonesia, decreased parental support was associated with total mental health difficulties, whereas anxiety due to the COVID-19 pandemic was associated with higher pro-social problems (April -May 2020, N = 113) ( Wiguna et al., 2020 ).
The previous findings seem to imply that the effects of the COVID-19 pandemic were unanimously negative regarding stress, coping and mental outcomes. However, the literature is more nuanced. A longitudinal study from the US ( N = 322) reported significantly lower internalizing, externalizing, and attention problems in adolescents during the lockdown (April 2020) compared to January 2020 (prior to the spread of COVID-19 in the US), which was associated with better family functioning in youth that reported elevated mental health problems before the pandemic ( Penner et al., 2020 ). Even though a decrease in physical activity and an increase in sleep and screen time were reported for adolescents in a Chinese cross-sectional study (May 2020, N = 10,082) ( Yang et al., 2020a ), an Italian study (April 2020, N = 306) showed that the great majority of the adolescents did not notice or only noticed very little changes in psychological well-being in the early phase of the pandemic ( Pigaiani et al., 2020 ). In Belgium and Italy (April -May 2020, N = 825), 5% of the participants reported having increased mental health care needs during the pandemic and 44% reported stability in needs for mental health care, with another 52% of the assessed youth reported no need for mental health care either before or after the pandemic ( Marchini et al., 2020 ). A study from Japan focusing on monthly suicide rates of people younger than 20 years old (January -May 2020, N = 138), found that suicide rates from March 2020 to May 2020 slightly decreased during the school closure time and were not significantly different from the previous two years ( Isumi et al., 2020 ). In Canada (March 2020, N = 683), almost half of a study sample of adolescents reported that the pandemic also exerted positive effects, with more time to spend with family and more time for exercise and hobbies. In addition, suicidal thoughts were reported to be lower than 6%. The type of motivation for social distancing was found to be associated with psychiatric symptomatology: social distancing due to fear of personally getting sick or to avoid judgement was related to higher anxiety, whereas social distancing due to the preference of staying home was associated with less anxiety and depressive symptoms ( Oosterhoff et al., 2020 ).

College students
Among university students in Spain (March 2020, N = 2530), moderate to extremely severe levels of anxiety, depression, and stress were reported by 21%, 34%, and 28% of the participants, respectively ( Odriozola-Gonzalez et al., 2020 ). In a Chinese cross-sectional study (February -March 2020, N = 407), the incidence of concerns about somatic symptoms, assessed by a somatic self-rating scale among college students, was 35%, mainly related to concerns regarding COVID-19 . In Bangladesh (May 2020, N = 476), over 80% of students exhibited any form of (mild to severe) depressive and anxiety symptoms, also related to worrying about academic activities ( Islam et al., 2020 ). In France (April -May 2020, N = 69,054), the prevalence of suicidal thoughts, severe depression, and high levels of anxiety were 11%, 16%, and 28%, respectively . In separate studies, female gender, social isolation, low quality of social relations were found as risk factors for lower mental health . In a US study (April 2020, N = 195), 71% of college students reported increased stress and anxiety, worry about their own and loved ones' health, sleep disruptions, difficulty with concentration, and concerns about academic performance as pandemic-related stressors ( Son et al., 2020 ). For Chinese college students, 25% of the students reported mild to moderate anxiety when cross-sectionally assessed during the pandemic ( N = 7143), with living in urban areas, income stability, social support, and living with parents as protective factors against anxiety . Worry about the economic influences of the pandemic, the academic delays and the influence of the pandemic on daily life also contributed to students' anxiety . In a Dutch study which ecologically followed students before and during the lockdown (March 2020, N = 78), mood homeostasis decreased significantly during lockdown, a finding that was in turn associated with lower mood and decreased engagement in activities that improved mood. Also, participants with previous mental illness showed a significantly higher decrease in mood homeostasis ( Taquet et al., 2020 ). In Italy (March -April 2020, N = 934), students' concerns about the COVID-19 pandemic increased PTSD symptoms, whereas positive thoughts about managing the epidemic showed the opposite effect ( Nania et al., 2020 ). Another longitudinal study conducted in the US ( N = 675) showed that a group of college students reported a slight improvement in internalizing symptoms, but worsening of externalizing symptoms and attention when assessed before (beginning of Spring semester 2020) and during the pandemic (end of Spring semester 2020) ( Copeland et al., 2021 ). Around 70% of a sample of 950 US students (March 2020) reported that using a coping strategy such as staying connected, trying to relax, keeping busy, having a day-to-day routine, hobbies, doing school work and exercising, were protective factors during the pandemic ( Waselewski et al., 2020 ). A cohort from the Zurich Project on the Social Development from Childhood to Adulthood ( N = 768) was assessed before the pandemic at the age of 20 and during the pandemic at the age of 22, and it was found that even though internalizing symptoms decreased, stress levels and anger increased during the pandemic. Moreover, only 30.5% reported feeling notably worse, whereas others were either feeling the same or better ( Shanahan et al., 2020 ).
Conclusion: In general, the pandemic has had negative effects on the mental health of children, adolescents, and students. Nevertheless, the effects vary widely within and across groups, and there are many methodological shortcomings in the current literature which is often crosssectional and relies on self-report. Moreover, it is difficult to directly compare results across many of the studies. In children and younger adolescents, parental stress and decreased physical activity seem to be important risk factors for worse mental health, but for adolescents and students, COVID-19-and academic related worries, familial conflicts, loneliness, and not engaging in health behaviors seem to be important risk factors. Although children, adolescent, and students seem particularly vulnerable for the negative effects of the pandemic in general, there are also some studies finding signs of resilience, such as the use of effective coping strategies. Findings from cross-sectional and longitudinal studies on the impact of COVID-19 on stress resilience and mental health in children, adolescents and college students are presented in Table 3 .

Elderly people
Elderly people have been more directly affected by the pandemic than younger age groups as the physical impact of COVID-19 is substantially higher in this population. In addition, even though elderly people have higher levels of comorbidities that need to be taken care of, the availability of physician appointments and medical care has also been negatively impacted by the lockdown periods ( Spalletta et al., 2020 ). In general, elderly people have lower access to technology and social media, which can make it more difficult to compensate for lockdown-related changes and have adequate access to food, news, and social interaction ( Martins Van Jaarsveld, 2020 ). They are more isolated from their families and social connections, leading to an increased risk of developing psychiatric symptoms and therefore impairing their cognitive performance and daily functioning . Elderly people also have been subject to change in health behaviors. A Cross-sectional study on the psychological, social and health-related challenges in Spanish older adults during the first COVID-19 wave (March 2020, N = 528) showed that while a quarter of the elderly people could increase their intellectual activity during the first lockdown, more than 60% of them reported decreased physical activity ( Rodriguez-Gonzalez et al., 2020 ). In Greece, 80% of the elderly reported moderate to severe anxiety and depression levels three weeks after a national lockdown (March 2020, N = 103) ( Parlapani et al., 2020 ), and these effects were also apparent in elderly with dementia in Argentina after the first 8 weeks of quarantine ( N = 119) . A German study (March -May 2020, N = 15,308) showed that while generalized anxiety significantly decreased with age, COVID-19 related fear significantly increased in elderly participants ( Schweda et al., 2021 ), even though this was not found in other studies.
In contrast, despite all the challenges that the elderly population faces, lower rates of psychiatric symptoms in this population compared to younger age groups were re-    During the COVID-19 quarantine period, more than 20% of adolescents had anxiety and depression. The prevalence was 21.7% (n = 1708) for anxiety and 24.6% (n = 1941) for depression. Exposure to COVID-19 predicted higher levels of PTSS and anxiety with effect sizes ranging from 0.06 to 0.15 (standardized betas). The largest variance in PTSS and anxiety problems was explained by adverse childhood events (ACEs), with more pre-pandemic maltreatment experiences predicting more PTSS and more anxiety.     Students reported how disruptive COVID had been to them personally. The mean level of disruptiveness (total score ranging from 0 to 10) was 7.8 (SD = 2.1),with 87.3% reporting a score of 6 or greater. After the onset of the COVID-19 pandemic, externalizing problems and attention problems increased after the onset of COVID, while internalizing symptoms did not increase.  ( Emerson, 2020 ). Another US survey study with elderly (March 2020, N = 825) revealed that less than 15% of the participants reported pandemic-related stress from confinement/restrictions, isolation, loneliness and concern for others and the unknown future. In particular, stress from concern for others and the unknown future was associated with poorer psychological well-being ( Whitehead and Torossian, 2021 ). In Spain (March -April 2020, N = 1639) ( Garcia-Fernandez et al., 2020 ) and the UK (April -May 2020, N = 15,530) , elderly people reported less psychopathology compared to younger people. Also, when compared to younger age groups, lower COVID-19 Peritraumatic Distress Index (CPDI) scores (Brazil, March 2020, N = 638) , lower rates of suicidal ideation (Greece, April -May 2020, N = 5116) ( Papadopoulou et al., 2021 ), and lower rates of depression, anxiety and stress (Northern Spain, March -April 2020, N = 1933), were reported in the elderly ( Ozamiz-Etxebarria et al., 2020 ). These findings might indicate that at least a sub-population of the elderly is remarkable resilient, potentially due to their complex experiences during their previous lives. The discrepancy of the findings in elderly people might heavily depend on the geographical location, timing of the measurements, and the nature of the assessed sample.

Elderly with cognitive symptoms and dementia
Elderly with cognitive symptoms may be more prone to the mental effects of the pandemic. In Italy (April -May 2020, N = 126), daily physical activity and adherence to a healthy diet were found to be decreased in this population. However, less than 20% of the participants reported depression and anxiety, which was related to living alone, having less social interaction and reduction in leisure activities ( Di Santo et al., 2020 ). Based on a review on patients with a dementia diagnosis (summarizing 20 studies from March 2020 and June 2020), anxiety, apathy, and agitation were the most reported neuropsychiatric symptoms during the COVID-19 pandemic, probably due to decreased social interaction ( Simonetti et al., 2020 ). Conclusion: Recent evidence points out that even though elderly people are more vulnerable to the physical effects of COVID-19, they also report lower psychopathology during the pandemic period compared to younger age groups. In the elderly population, patients with cognitive decline may be more vulnerable to the mental health effects of the pandemic. However, there is quite some heterogeneity within groups, and a significant proportion of the elderly people may still be at risk for worse mental health outcomes. Table 4 presents findings from cross-sectional and longitudinal studies on the impact of COVID-19 on stress resilience and mental health in elderly people.

Pregnant women
During the first days of the pandemic, it was not clear whether a COVID-19 infection would affect pregnant women and whether the virus would be transmitted to the fetus. Many pregnant women therefore avoided visiting hospitals, and obstetricians in India reported that a great majority of their pregnant patients experienced anxiety about their hospital visits during the first wave of the COVID-19 pandemic (April -May 5, 2020, N = 118) ( Nanjundaswamy et al., 2020 ). These worries in pregnant women were related to COVID-19-related concerns about being present in public places, perceived infection risk, visiting hospitals, health of the fetus, delivery concerns, a family member being infected, or transmission of COVID-19 to the baby during delivery ( Akgor et al., 2021 ;Taubman-Ben-Ari et al., 2020 ;Zhang et al., 2020 ). A nationwide cross-sectional study from Mexico (May -June 2020, N = 503) found that 33.2% of pregnant women reported being stressed based on the Perceived Stress Scale, with perceived stress being significantly correlated with later gestational age ( Medina-Jimenez et al., 2020 ). In China (February -March 2020, N = 560), over half of pregnant women reported feeling horrified, apprehensive, or helpless during the pandemic . A Turkish study reported that half of the 172 enrolled pregnant women in the third trimester reported feeling vulnerable to the effects of the pandemic (April 2020) ( Yassa et al., 2020a ). In a Pakistani (August 2020, N = 552) and Chinese study (February 2020(February , N = 1947, more than 80% of the pregnant women reported that they themselves and their fetuses were more vulnerable to the effects of the pandemic compared to the general population Shahid et al., 2020 ), but moderate to severe anxiety was reported in only 3% . A systematic review, that included 15 studies examining depression and anxiety symptoms in pregnant or delivered women during the COVID-19 pandemic, reported a pooled overall prevalence of 30% for depression and of 34% for anxiety ( Sun et al., 2020 ). In addition, the prevalence of depression and anxiety symptoms was around two times higher compared to non-pregnant women ( Sun et al., 2020 ). Another systematic review and meta-analysis that included 19 studies about the mental health status of pregnant women during the COVID-19 pandemic reported a prevalence between 5 and 38% for depression and anxiety symptoms (overall prevalence of 25%) and a pooled overall prevalence of anxiety of 42% ( Fan et al., 2020 ). A rare longitudinal study in Argentina confirmed significant increases in depressive, anxiety and negative affect in 102 pregnant women at 2, 14, and 47 days after the start of the lockdown compared to 102 non-pregnant women ( Lopez-Morales et al., 2021 ). During the pandemic, 1754 pregnant women in Canada reported significantly higher levels of depressive, anxiety, dissociative and post-traumatic stress symptoms compared to a pre-COVID-19 cohort of pregnant women ( Berthelot et al., 2020 ). However, again, conflicting results are reported. A Turkish study (April 2020) found lower state anxiety in 203 pregnant women compared to 101 non-pregnant women  Older people rated quality of life, life satisfaction, and well-being higher than young people and scored less than young people on anxiety (mean difference = −9.19, SE = 1.90, p < 0.01) and greater than young people on risk tolerance (mean difference = 1.38, SE = 0.33, p < 0.01 difference = 0.91, SE = 0.31, p < 0.05).
( continued on next page ) The ≥60 age group showed lower depression scores and lower acute distress scores than the < 60 age group. There were no gender differences in any of the clinical measures.  Adherence to the Mediterranean diet decreased in almost 1/3 of respondents and over 35% reported weight gain. Social activities were abolished and 1/6 of participants also decreased productive and mental-stimulating activities. 19.8% were depressed, 9.5% anxious, and 9.5% apathetic. ( Yassa et al., 2020b ). In China (February -March 2020, N = 859;), pregnancy was associated with a lower risk for depression, anxiety, insomnia, and PTSD ( Zhou et al., 2020b ). Another Israelian study (March -May 2020, N = 369) also reported lower rates of depression in pregnant women hospitalized during the pandemic, compared to pregnant women that were hospitalized before the pandemic ( Sade et al., 2020 ). Pre-existing psychiatric disorders appear to increase the experience of depressive, anxiety and dissociative symptoms in pregnant women during the pandemic ( Berthelot et al., 2020 ;Liu et al., 2021 ;Ravaldi et al., 2020 ), and previous anxiety disorders increases health-related and society-related anxiety levels ( Berthelot et al., 2020 ;Ravaldi et al., 2020 ).
There are several factors that may impact on the level of stress resilience specifically in pregnant women. Maternal social support in China and Ethiopia significantly decreased anxiety levels in pregnant women ( Yue et al., 2020 ), and also increased health-related quality of life ( Dule et al., 2021 ). Social support in Canada was also negatively correlated with depression and insomnia, whereas negative cognitive appraisal positively correlated with these symptoms ( Khoury et al., 2021 ). Risk of COVID-19 infection ( Bo et al., 2020 ), social isolation ( Durankus and Aksu, 2020 ), financial and relationship difficulties ( Bo et al., 2020 ;Khoury et al., 2021 ;Lebel et al., 2020 ;Matsushima and Horiguchi, 2020 ;Mortazavi et al., 2021 ), marital life satisfaction ( Effati-Daryani et al., 2020 ), intimate partner violence ( Almeida et al., 2020 ), sleep difficulties ( Lin et al., 2021 ), were reported factors that relate to psychopathology in pregnant women during the COVID-19 pandemic. In addition, having a relative with COVID-19 infection, a history of abortion, and an age below 30 increased pregnant women's worries ( Mortazavi et al., 2021 ). In the US (April -May 2020, N = 787), compared to white women, black women reported significantly higher rates of depression, more pregnancyrelated worries, more worries about the financial burden of the pandemic and having a job that is negatively affected by the pandemic ( Gur et al., 2020 ). Finally, a report from Iran (March -April 2020, N = 580) showed that fear of COVID-19 was associated with suicidal ideation, quality of life and depression in 290 pregnant women ( Ahorsu et al., 2020 ).
Conclusion: Although several cross-sectional studies, mostly surveys, report that pregnant women experienced higher levels of stress, depression and anxiety during the pandemic period compared to non-pregnant women, several other studies found comparable or even better mental health outcomes. It remains quite challenging to assess the prevalence of anxiety and depression in pregnant women during the COVID-19 pandemic compared to the pre-pandemic period. Regarding risk factors, lack of social support and fear about pandemic-related issues reduced the stress resilience of pregnant women. An important aspect for future studies will be to assess the impact of the pandemic-related stress exposure during pregnancy on the children of these mothers. For an overview of findings from cross-sectional and longitudinal studies on the impact of COVID-19 on stress resilience and mental health in pregnant women, see Table 5 .

Adults with a psychiatric disorder
Whilst facing the COVID-19 pandemic has important negative consequences in terms of mental health and increases the vulnerability for psychological problems, severe mental illness in turn has been shown to represent a vulnerability factor for COVID-19 infection. Several studies have shown that patients affected by severe mental illness have an increased risk to become infected compared to the general population (odds ratios ranging from 5.7 to 7.6) Wang et al., 2020b ), which is linked to poorer environmental conditions, such as socioeconomic deprivation. Furthermore, they might have more difficulties with being compliant to the rules and obligations established to fight the pandemic and tend to be generally exempt by wearing personal protective equipment such as masks ( Ayuso-Mateos et al., 2020 ). Indeed, in their analysis of anonymized electronic health records of 62,354 US patients affected by COVID-19 (January -August 2020), Taquet et al. clearly showed that the presence of a preexisting psychiatric illness was significantly associated with a higher risk of a COVID-19 diagnosis (RR = 1.65, 95% CI: 1.59-1.71, p < 0.0001), independent of known physical health or economic and housing risk factors ( Taquet et al., 2021 ). This finding was confirmed by a recent analysis of deidentified population-level electronic health records data ( N = 61,783,950) from US hospitals, which showed that individuals with a recent diagnosis of a mental disorder had a significantly increased risk for COVID-19 infection, with an effect strongest for depression (OR = 7.6) and schizophrenia (OR = 7.34) . Of interest, this increased risk was further exacerbated among African Americans and women. It is plausible that these patterns of associations might be also related to the effect that COVID-19 has exerted on the levels of clinical care in psychiatry which showed a substantial decrease since the start of the pandemic ( Carpiniello et al., 2020 ;Yao et al., 2020 ).
In this context, stress resilience appears among one of the many plausible moderators of the identified increased risk of infection in patients affected by severe mental illness ( Ameis et al., 2020 ;Jacob et al., 2020 ). This is of relevance given that building resilience in the general population and at-risk patient populations will be a key instrument to decrease the impact of the COVID-19 related socioeconomic shock . Indeed, there is evidence that specific factors such as having a higher academic level, being autonomous, having self-efficacy, and the presence of optimism have been shown to be significant predictors of resilience in the general population during the COVID-19 pandemic ( Robles-Bello et al., 2020 ), protecting individuals from the development of mental disorders. This points to the importance of building resilience in at risk population such as those affected by severe mental disorders. Indeed, the work from Burrai et al. has shown that Italian psychiatric patients in residential community (April -May 2020, N = 82) scored lower than healthy individuals in levels of stress although, as expected, they showed higher levels of anxiety, perceived risk of getting infected with COVID-19 and worry about the emergency situation ( Burrai et al., 2020 ). This finding is probably justified by the perceived and experi- Participating pregnant women reported high COVID-19-related anxiety, with leaving the home (taking public transportation or being in public places) being the greatest cause for concern. Specifically, the use of public transportation (87.5%) was reported as the cause of the highest anxiety, followed by the potential infection of other family members (71.7%), being in public places (70%), concern for the fetus (70%), going for pregnancy check-ups (68.7%), being infected themselves (59.2%), and the delivery (55.4%). Arab women were more distressed and anxious compared to Jewish women. Zhang et al., 2020 Chinese pregnant women February 13-16 2020 Cross-sectional, N = 1901 (mean age 28.9 years ( ±4.7) Presence of prenatal depression (PND) and post-traumatic stress disorder (PTSD) during the COVID-19 pandemic Being in the second or third trimester of pregnancy During the early stage of the COVID-19 outbreak, high anxiety levels, a high prevalence of probable PND (34%) and a high prevalence of suspected PTSD (40%) was observed among pregnant women.  In a time range of 50 days of quarantine, pregnant women showed a higher increase in depression, anxiety and negative affect and a higher decrease in positive affect compared to non-pregnant women. The prevalence of depression, anxiety, physical discomfort, insomnia and post-traumatic stress disorder (PTSD) during the COVID-19 pandemic Having a childbearing age During the COVID/19 pandemic, pregnant women had lower scores of symptoms of depression, anxiety, and PTSD (all p < 0.05) compared to non-pregnant women. The relationship between social support, risk perception and anxiety among third-trimester pregnant women during the COVID-19 pandemic Having a current pregnancy (third trimester) The third trimester pregnant women had a high level of social support, a medium level of risk perception to COVID-19 and were susceptible to anxiety. Risk perception played a mediating role between social support and anxiety. The prevalence of depression (9-item Patient Health Questionnaire -PHQ-9) during the COVID-19 pandemic and its associated factors in women in the perinatal stages Being a woman in the third semester of pregnancy, or post-partum period (from the beginning of pregnancy to one week after childbirth), being 18 years or older, not having a pre-existing psychiatric disorder A high prevalence of depression in women across the perinatal stages was observed. Worries about infection and interrupted routine medical check-ups were associated with an increased risk of depression.

Durankus & Aksu, 2020
Turkish pregnant women Not reported Cross-sectional, N = 260 (mean age 29.56 years ( ±3.83))) The effects of the COVID-19 pandemic on depression and anxiety in pregnant women, using the Edinburgh Postnatal Depression Scale (EPDS) Not having a history of a psychiatric disorder 35.4% of the participating pregnant women scored higher than 13 on the EPDS and were thus being considered as being at risk of developing depression. COVID-19 pandemic effects regarding psychology and social isolation, anxiety symptoms and depressive symptoms contributed to increased depression in pregnant women. The percentage of women experiencing a low well-being state was relatively high. Predictors of experiencing low well-being were worry about own health, health of others and the fetus, having at least one infected person with COVID-19 among relatives. enced support that residential patients receive from mental health workers and peers.
At the same time, there is evidence that COVID-19 has exerted a role as a precipitating factor for an exacerbation of existing psychiatric disorders, particularly in those disorders where stress is a key trigger, such as PTSD, mood disorders, and schizophrenia ( Ettman et al., 2020 ;Horn et al., 2020 ;Jolly et al., 2020 ;Ma et al., 2020 ;Pinkham et al., 2020 ;Rutherford et al., 2021 ). For instance, in China (January -April 2020), 30 patients with schizophrenia who were socially isolated after having close contact with a COVID-19 case showed a higher severity of symptoms, including higher levels of perceived stress and anxiety and lower quality of sleep, as well as a higher inflammatory load compared to 30 patients with schizophrenia not subjected to quarantine measures . Of interest, there is anecdotal evidence of two cases in the US where symptoms of PTSD might have been exacerbated by public masking ( Jolly et al., 2020 ). However, Pinkham et al. showed in 148 US individuals with severe mental illness (92 with schizophrenia spectrum illnesses and 56 with affective disorders) that affective symptoms and sleep were stable after five months from the start of the pandemic ( Pinkham et al., 2020 ). Convincing evidence stems from three longitudinal Dutch case-control cohorts (NESDA: N = 2329 cases and N = 652 controls; NESDO: N = 378 cases and N = 132 controls; NOCDA: N = 419 cases), showing that patients with depressive, anxiety, or obsessive-compulsive disorders are not experiencing a large detrimental impact on their mental health during the COVID-19 pandemic compared to before, even though symptom severity remained substantially high -and many more times higher than healthy controls, indicating the burden and severity of psychiatric disorders compared to general population symptom levels ( Pan et al., 2021 ). With regard to suicidality, a series of studies have shown that the suicide risk increased significantly during the pandemic ( Iob et al., 2020b ;John et al., 2020 ;Nomura et al., 2021 ;Sáiz et al., 2020 ;Singh, 2020 ;Tanaka and Okamoto, 2021 ). While the grim forecast from predictive models showed increased rates of suicide during the pandemic , data emerging from epidemiological observations in countries as the Netherlands and Japan found that monthly suicide rates declined during the first months of the pandemic, even though levels in Japan increased by 16% during the second wave ( Tanaka and Okamoto, 2021 ). Other studies showed increased rates of various suicidal behavior components in the general population including passive suicidal ideation ( Sáiz et al., 2020 ) and self-harm ( Iob et al., 2020b ).

Children and adolescents with a psychiatric disorder
Children and adolescents with an eating disorder diagnosis represent a vulnerable group. Reactivation of eating disorder symptoms occurred in 42% of 365 young patients followed up in an eating disorder clinic in Spain during the first 8 weeks of COVID-19, particularly in adolescents (March -May 2020, N = 365). In half of the cases, the clinical worsening was associated with eating restriction and excessive exercising due to reactivation of weight phobia. Social iso-lation and increase in family conflicts were reported as relevant factors ( Graell et al., 2020 ). Obsessive compulsive disorder symptoms were also reported to increase in the first months of the pandemic in most children and adolescents diagnosed with OCD, in addition to increased anxiety and depressive symptoms (Denmark, April-May 2020, N = 67) ( Nissen et al., 2020 ). In a Turkish study, 61 participants aged 6-18 years who had been diagnosed with OCD were assessed in April 2020. 54% reported worsening of their symptoms and 36% reported more than a 30% increase in CY-BOCS scores. On the other hand, 11.4% of the patients reported decreased symptoms. More than half of the patients who were in remission also reported increased symptoms that reached a clinically significant threshold. Among the factors that predicted worsening of OCD symptoms during the pandemic were daily preoccupation with COVID-19, searching on social media about COVID-19, duration of OCD diagnosis and a diagnosis of COVID-19 in someone familiar ( Tanir et al., 2020 ).
Children with neurodevelopmental disorders such as attention deficit hyperactivity disorder (ADHD) and autism have shown to be at risk during the COVID-19 outbreak. A survey completed by 241 Chinese parents of children aged 6-15 years old who were diagnosed with ADHD, showed that ADHD symptoms significantly worsened during the lockdown . In a survey study in Italy (April 2020, N = 527), parents reported that respectively 36% and 42% of children diagnosed with an autism spectrum disorder (ASD) experienced more intense and more frequent behavioral problems compared to the period before the pandemic . Further, Turkish children diagnosed with ASD reported more sleep problems during the home confinement period which mediated autism symptom severity (May 2020, N = 46)) ( Turkoglu et al., 2020 ). In another Turkish study that included 87 ASD patients (aged 3 to 29 years old), it was found that parents' anxiety levels were significantly correlated with the child's total score on the ABC (Aberrant Behavior Checklist). Half of the parents reported that their children became more aggressive, around one third of the parents reported sleep and appetite changes in their children and a quarter of the parents reported that their child's tics increased, or new tics emerged ( Mutluer et al., 2020 ).
Conclusion: Overall, people with existing psychiatric disorders are experiencing a detrimental impact on their mental health from the COVID-19 pandemic, for example in OCD and PTSD, which requires close monitoring in clinical practice. The COVID-19 pandemic, however, does not seem to have further increased symptom severity in adult patients with depressive and anxiety disorders compared to their pre-pandemic levels ( Pinkham et al., 2020 ); ( Pan et al., 2021 ). Longitudinal observations with adequate time of follow-up suggest an increased risk for suicidality associated with the pandemic, even though there are conflicting reports. Regarding children and adolescents diagnosed with a psychiatric disorder, studies have generally reported a worsening of symptoms in young patients with eating disorders, obsessive compulsive disorders, and neurodevelopmental disorders such as ADHD and ASD. Severe mental illness in turn has been shown to represent an important vulnerability factor for COVID-19 infection. Basing on the increased vulnerability to COVID-19 in psychiatric patients, European Neuropsychopharmacology 55 (2022) 22-83 several European countries have prioritized them for vaccination. Many other countries are currently evaluating this option ( De Picker et al., 2021 ). See Table 6 . for an overview of findings from cross-sectional and longitudinal studies on the impact of COVID-19 on stress resilience and mental health in patients with a psychiatric disorder.

COVID-19 patients
Previous SARS and MERS pandemics have shown that infection with the virus itself can be associated with increased symptoms as well as new diagnoses/symptoms of anxiety, depression, impaired memory, fatigue and insomnia in the acute as well as post-illness phase, and there is burgeoning evidence for substantial psychiatric symptoms related to COVID-infection ( Rogers et al., 2020 ;Taquet et al., 2021 ). An electronic health record network cohort study using data from 69 million individuals showed that in the three months following testing positive for COVID-19, 1 in 5 survivors was recorded as having a first time diagnosis of anxiety, depression or insomnia. This was about twice as likely as for other groups of patients in the same period ( Taquet et al., 2021 ). Furthermore, Horn et al. found that the prevalence of PTSD in patients with COVID-19 was around 6.5%, and a similar rate was also reported in COVID-19 patients discharged from hospitals in Wuhan . In an Italian crosssectional study (April -October 2020) that included 381 patients who had recovered from COVID-19 within 30 to 120 days, a PTSD prevalence of 30.2% was reported after acute COVID-19 infection ( Janiri et al., 2021 ). It is crucial to determine protective factors increasing resilience against mental health impairment following infection with COVID-19. This is illustrated by findings in a sample of 296 Chinese patients with mild symptoms of COVID-19, where higher resilience measured with the Connor-Davidson Resilience Scale was correlated with lower anxiety ( r = −0.391, p < 0.001) and depression ( r = −0.472, p < 0.001) scores. Patients with high resilience (upper 27%) were much less likely to display symptoms of anxiety (OR = 0.362, p < 0.001) or depression (OR = 0.301, p < 0.001) . Education about disease-related facts, emotional support and confidence of rehabilitation may enhance resilience in COVID-19 affected patients . Following up on these first data, longitudinal studies correcting for potential confounders such as pre-existing mental disorders, concomitant somatic disorders, severity of infection, degree of physiological compromise, immunological response, extent of medical interventions and socioeconomic situation, are warranted to further elucidate the role of resilience and its determining factors in promoting mental health in COVID-19 patients in order to develop targeted preventive interventions strengthening coping skills, self-efficacy, will power, daily routines and opportunities to share the emotional burden ( Richards and Scowcroft 2020 ).
Conclusion: In sum, given the high risk of psychiatric sequelae of COVID-19 infection, preventive measures promoting mental health as well as intensified screening for symptoms of mental disorders should be routinely implemented in the standard care of COVID-19 patients to increase resilience towards mental disorders particularly in this patient group. Table 7 presents findings from cross-sectional and longitudinal studies on the impact of COVID-19 on stress resilience and mental health in COVID-19 patients.

Interindividual differences in stress resilience: implications for the pandemic
From the previous sections it is apparent that the effects of the pandemic, either related to COVID-19 itself or the associated measures, are surprisingly heterogeneous across populations. Trajectories of mental distress varied markedly by resilience level during the early months of the COVID-19 pandemic ( Riehm et al., 2021 ). It is thus of paramount importance to understand which individuals are resilient or vulnerable to apply a personalized medicine approach ( Willis and Lord, 2015 ). This approach harnesses the individual's genetic, genomic, proteomic, clinical, socioeconomic and lifestyle information to identify the factors causing the differential resilience/vulnerability to the virus. Such information allows to define resilient/vulnerable subpopulations, to refine targeted therapeutic strategies and to develop an effective public health approach. A meta-analysis examining 68 studies comprising 288,830 participants from 19 countries on factors associated with psychological distress during the COVID-19 pandemic (December 2019 -July 2020), showed that being female, being younger than 35 years old, living in rural areas, lower socioeconomic status, higher COVID-19 infection risk, longer social media exposure and having pre-existing physical or mental conditions were associated with higher anxiety and depression odds ( Wang et al., 2020c ). Higher social/family support, physical activity and positive coping strategies were associated with lower odds of anxiety and depression and thus a reduced risk of psychological distress ( Wang et al., 2020c ).
With regard to mental health, psychological attitudes towards how to manage the risk of infection, as well as towards specific medical, socioeconomic, personality and lifestyle factors, have been suggested to be key for the individual's resilience to distress and psychiatric disorders during a pandemic ( Chen and Bonanno, 2020 ;WHO, 2020 ;Zager Kocjan et al., 2021 ). Individual trait resilience and well-being scores, measured respectively with the 10-item Connor-Davidson Resilience Scale and the Recovery from War Scale, have been found to predict effective coping with the COVID-19 threat ( Kimhi et al., 2020 ;Ran et al., 2020 ). In addition, strategies aimed at reducing psychological distress such as paying attention to a healthy lifestyle, social support, good quality of sleep, acceptance of negative emotions, and avoidance of suppression and substance abuse, have been suggested to increase psychological resilience and may be key in coping with the COVID-19 related distress ( Bozdag and Ergun, 2020 ;Petzold et al., 2020 ). By contrast, loneliness, or negative psychological reactions, including panic response, hysteria, hopelessness and desperation, have been associated with negative outcomes, including suicidal ideation ( Killgore et al., 2020a ;Lee, 2020 ;Serafini et al., 2020 ;Thakur and Jain, 2020 ). Although the impact of the pandemic is still under investigation and initial large scale data analysis show that suicide numbers have remained largely unchanged or declined in the early months of the pandemic compared with the expected levels based on the pre-pandemic period in high-income and Having an electronic health record Survivors of COVID-19 appear to be at increased risk of psychiatric sequelae, and a psychiatric diagnosis might be an independent risk factor for COVID-19.
( continued on next page ) The level of resilience of the general Spanish population exposed to a traumatic situation by the COVID-19 pandemic Being 18 years or older, having a Spanish nationality, being a resident in Spain, having read the information sheet and accepted the informed consent, having completed the questionnaire.
The Spanish population exposed to confinement presented high levels of resilience, but no relevant post-traumatic growth took place. Having a higher academic level, being autonomous), along with self-efficacy) and to a lesser extent optimism predicted a resilient outcome. Burrai et al., 2020 Italian Results showed that 6.5% of the patients presented with probable PTSD. Psychotropic medication, hospitalization, and distress during the acute phase of COVID-19 were significantly associated with the severity of the PTSD symptoms. Although people with depressive, anxiety, or obsessive-compulsive disorders scored higher on all four symptom scales than did individuals without these mental health disorders, both before and during the COVID-19 pandemic, they did not report a greater increase in symptoms during the pandemic Patterns of abuse, self-harm and thoughts of suicide/self-harm in the UK during the first month of the COVID-19 pandemic Presence of data on abuse, self-harm and thoughts of suicide or self-harm on at least one occasion The reported frequency of abuse, self-harm and thoughts of suicide/self-harm was higher among women, Black, Asian and minority ethnic (BAME) groups and people experiencing socioeconomic disadvantage, unemployment, disability, chronic physical illnesses, mental disorders and COVID-19 diagnosis.   upper-middle-income countries it is plausible that more accurate figures will be detected in the long-term even after the pandemic will decrease its burden globally. Mental distress can be exacerbated by being quarantined ( Xin et al., 2020 ). It is worth noting that young people have reported to experience greater psychological distress than adults ( McGinty et al., 2020 ;Pierce et al., 2020 ;Varma et al., 2020 ). Unfortunately, to the best of our knowledge, no data on potential genetic, epigenetic, or brain function markers of resilience to mental illness in the face of the pandemic have been published yet. However, it can be hy-pothesized that neurobiological factors usually involved in the stress response, emotion-regulation, and the ability to adapt to new life conditions, such as immune system activation, hypothalamic-pituitary-adrenal axis activity and neural plasticity processes, might be involved ( Branchi and Giuliani, 2020 ;Zorn et al., 2017 ). The three-dimensional vulnerability-stress-coping model may help understanding risk and resilience for mental disorders in relation to stress during the COVID-19 pandemic. This gene x environment x coping (G x E x C) model has been proposed to include the three factors a) genetic vulnerabil- Significant negative correlations were found between individual/community resilience and sense of danger ( −0.220 and −0.255 respectively; p < .001) and distress symptoms (-0.398 and −0.544 respectively; p < .001). Individual resilience and well-being showed to be the first and foremost predictors of COVID-19 anxiety.   ity, b) risk-increasing external stressors and c) resilienceincreasing coping mechanisms, such as self-efficacy buffering the impact of a high genetic vulnerability and/or adverse environment. Exemplary studies have demonstrated this complex, three-fold interaction regarding risk or resilience towards anxiety phenotypes ( Schiele et al., 2020( Schiele et al., , 2016. Finally, epigenetic mechanisms at the interface between biology and biography and impacting on stressresponse and emotion-regulating capacities might add another level to confer individual risk as well as resilience towards mental disorders under adverse and advantageous environmental conditions ( Vinkers et al., 2015 ). Therefore, an extended (Epi) G x E x C model might prove useful in informing future personalized preventive interventions increasing resilience towards mental disorders. At the same time, existing hypotheses as the stress inoculation or match/mismatch hypothesis for psychiatric disorders may help to understand why selected population groups show increased resilience levels, for instance due to previous life experiences that now enable them to deal appropriately with the current challenges ( Nederhof and Schmidt, 2012 ;Schmidt, 2011 ). Conclusion: There is an urgent need for a personalized approach when it comes to identifying individuals at risk or resilient for the stressful effects of the COVID-19 pandemic. The effects of stress and the resilience capacity are dependent on (neuro)biological, psychological, and environmental factors and are heavily dependent on an individual's unique context. Intensified research into (epi)genetic, proteomic, immunological, clinical, neuropsychological, socioeconomic and lifestyle factors conveying mental disorder risk or resilience in the context of the present pandemic is urgently warranted to provide individually tailored and thus most efficient resilience-increasing preventive measures. Findings from cross-sectional and longitudinal studies on interindividual differences in stress resilience are presented in Table 8 .

Conclusions and future perspective
We believe resilience research is in a unique position to make a significant contribution to understand the psychological and psychiatric impact of this pandemic and in-form future clinical and research directions. This is not limited to the current COVID-19 pandemic but also applies to other local or global challenges. From the current literature, a surprising level of resilience is apparent across populations, even though there are individuals and groups that are at increased risk for the stressful effects of the COVID-19 pandemic ( Fig. 1 ). Nevertheless, firm conclusions cannot be drawn as most of the current literature has major methodological limitations. Most studies on stress resilience and mental health outcomes during the pandemic are observational, cross-sectional, using convenience samples with often rather small sample sizes and rather limited assessment of contextual and personal characteristics that are essential to understand stress vulnerability and resilience. Interpretation of observational studies from nonrepresentative samples is likely to suffer from bias, particularly regarding collider bias ( Griffith et al., 2020 ). There is a stark contrast between the very limited number of extensive longitudinal studies with pre-pandemic assessments and a broad array of outcomes, and the vast number of cross-sectional studies with one or two outcomes. Where newer cohorts lack baseline data from before the pandemic, large established cohorts move relatively slowly and mostly sample infrequently. This makes more fine-grained assessments of resilience and mental health more challenging. Finally, there is lack of observational data on how the general public and patients with psychiatric disorders actually deal with self-care, nutrition, physical activity or restorative sleep during confinement ( Balanza-Martinez et al., 2020 ). Thus, public policies will need to be informed by data gathered in observational studies of lifestyle behaviors during the compulsory isolation ( Balanza-Martinez et al., 2020 ). This research gap has been partly filled by very recent data confirming that psychiatric patients, particularly those affected by depression and anxiety tended to have higher levels of psychopathological distress , and that the presence of depressive symptoms was a predictor of poorer resilience .
Nevertheless, studies during the pandemic consistently show that children and young people are the most vulnerable group with increased psychological distress, probably because their needs for social interactions are stronger. Moreover, young women appear to be more vulnerable than young men, and parents with young children appear to be at particularly high risk for mental health problems. However, these are rather broad conclusions which cannot be used at the individual level. There is still an urgent need to identify individuals and populations with higher risk of psychological distress during the COVID-19 pandemic to offer targeted mental health care and to improve social support, physical activity, and coping strategies in these individuals. These approaches are needed to boost resilience factors protecting the individual against psychological distress. For example, social support from and connectedness with family, friends, and a special caring loved one were each independently associated with greater resilience ( Killgore et al., 2020b ;Nitschke et al., 2020 ) ( South et al., 2020 ) ( Prime et al., 2020 ). Moreover, there seems to be a role for media with regard to resilience and mental health during the pandemic. Nuanced and balanced news coverage around the COVID-19 pandemic is essential in order that nocebo effects as a result of negative and alarming news coverage occur ( Bendau et al., 2020a ). Next, employers are taken up on their promise to assist their employees by building up resilience strategies at the organizational level. It remains to be seen how stress resilience will be shaped because of longer-term effects of the pandemic, or when society will need to find a new balance after the lockdown and pandemic. This is particularly pressing considering possible economic sequelae that will emerge while at the same time individuals will be forced to change from a lockdown situation to a more proactive attitude when all societal processes will start to function again.
About the methodological quality of the resilience literature during the COVID-19 pandemic, several lessons can be learned for future research to provide a more detailed and fine-grained picture, with possibilities for targeted prevention and intervention. First, the use of longitudinal data and large samples is important to identify potentially causal relationships, and changes over time ( Kalisch et al., 2017 ). How we respond to stress is a surprisingly dynamic process. Stress initiates a cascade of behavioral, (neuro)biological, and physiological changes. To this end, we need prospective data of adequate duration and of sufficient temporal resolution to observe (dis)continuous changes in resilience. Second, concerning outcomes and context, biological, psychological, and environmental data should be combined and integrated to understand the impact of (pandemic-related) stress at different levels with understanding of the individual's unique context. Third, we would encourage interdisciplinary collaborations, for example between physicians treating COVID-19 patients, psychiatrists, sociologists, advanced data experts, and neurobiologists. Fourth, we need not only to identify how stress resilience is shaped during a (prolonged) pandemic, but also develop efficient mental health interventions at a governmental, institutional, and individual level to minimize its long-term consequences. In conclusion, it is apparent that we have shown a remarkable level of resilience during the prolonged COVID-19 pandemic, but that large interindividual differences exist. Above all, it provides an opportunity but also an imperative for scientists and clinicians to work together to help understanding and addressing the pandemic.

Contributions
CV wrote the first draft. All authors provided critical input and revisions.

Conflicts of Interest
There are no conflicts of interest