Introduction

The COVID-19 pandemic

In December 2019, the outbreak of a respiratory disease caused by the novel virus “Severe Acute Respiratory Syndrome Coronavirus type 2” (SARS-CoV-2) was reported in the People’s Republic of China [1]. The outbreak rapidly developed into an epidemic in that country by January 2020 and quickly spread to other regions of the world. On  11 March 2020, The World Health Organization (WHO) declared the spread of the virus-induced disease, COVID-19, a global pandemic [1]. By the end of February 2023, more than 757 million confirmed cases of COVID-19, over 6.85 million of which were fatal [2], were reported to the WHO. Further, it was estimated that in the years 2020 and 2021, 14.83 to 14.9 million excess deaths were associated with the pandemic [3, 4]. As a result, to contain the spread of the virus, drastic measures were taken worldwide, which led to massive restrictions on everyday life and the global economic crisis of 2020 and 2021.

Effects of the COVID-19 pandemic on the daily life of children and adolescents

For many children and adolescents, especially in high-income countries, the still ongoing pandemic represents the first confrontation with a threat of severe disease, potential death, and grief in their lives [5,6,7]. In many, the pandemic has caused anxiety and worries about infection and the health of themselves and of their family and friends [6, 8]. Feelings of safety, trust, and security have been replaced by the perception that the world is a dangerous and unsafe scary place [5, 9, 10]. Young people in a British survey conducted by Youngminds [10] reported loneliness and isolation, concerns about school work, and breakdown of routines to be the most stressful experiences during the pandemic.

The global economic crisis caused by pandemic-related restrictions has led to a sharp rise in unemployment and poverty around the world [11, 12]. Especially in low-income countries, these economic hardships have forced many children into exploitative and dangerous work to support their families [11]. Further, since the start of the pandemic, essential physical and mental health services have been disrupted in over 90% of countries worldwide [13, 14]. Child and adolescent access to routine immunizations and examinations has, therefore, been impaired, leading to increased morbidity and mortality [12, 13]. In the first months of the pandemic, critical mental health services were halted worldwide, which has led to massively reduced access to psychological, psychiatric, and psychosomatic in- and out-patient services [14,15,16]. The work of child protection services has been impaired in at least 104 countries, making it hard to prevent, report, detect, and respond appropriately to cases of child maltreatment [17, 18].

One of the biggest and most far-reaching changes in the lives of children and adolescents during the pandemic has been the closure schools and childcare facilities. These closures have been globally regarded as necessary due to the impracticability of distancing practices in school because of limited space, frequent and varied interactions among large groups of pupils, and difficulties for children (especially younger ones) to follow hygiene and distancing guidelines [19]. According to the WHO [13], school closures have led to the largest disruption of education systems in history, affecting nearly 1.6 billion students in more than 190 countries.

It is likely that closer social relationships of children and adolescents have suffered during the pandemic due to limited and discouraged socialising and imposed isolation from friends and extended family members [17, 20]. Entire family systems have experienced disrupted daily activities and pandemic-related stress. As a result, inner-familiar tension has increased and family dynamics in many households have changed [9, 12].

Taken together, the COVID-19 pandemic and the related containment measures have massively changed the daily lives of children and adolescents. For many, normal development has been impaired and stress and strain have increased while the availability of many coping, support, and protection resources has been limited.

Effects of disasters on mental health

The unpredictability, community impact, fatalities, and persistent effects of the COVID-19 pandemic make it a disastrous event according to literature on the mental health effects of disasters [15, 21,22,23]. Exposure to a disastrous event requires adaption from every individual involved [24]. This adjustment process is influenced by various factors from before (e.g., mental health history), during (e.g., injury, life threat, cumulative risk exposure), and after the event (e.g., coping behaviour, family functioning) that depend on individual experiences and resources [21, 22, 25,26,27,28,29,30]. Therefore, disastrous events can produce multiple patterns of outcome (e.g., stress-resistance, transient distress with healthy adaption, breakdown without recovery, post-traumatic growth) [21, 22, 26]. This is in line with the transactional stress model by Lazarus and Folkman [31] and classic vulnerability-stress theories, which claim that the effects of negative life events on mental health depend on individual vulnerability factors, such as how an individual perceives, attends to, appraises, interprets, copes with, and remembers such events [32, 33]. Thus, the effects of exposure to the COVID-19 pandemic, a disastrous event, seem to depend on a complex interaction of individual and environmental factors.

In general, disasters have been associated with an increase in a variety of mental health problems (e.g., depression, anxiety, post-traumatic stress, acute stress reactions, adjustment disorders, substance abuse, somatic complaints, and prolonged grief) [21, 22, 25, 28]. These mental health effects can be long-lasting, sometimes found even 12–20 years after a disastrous event [25, 28, 30].

However, catastrophes typically studied in disaster research are restricted to relatively short one-time events with ultimate and intense consequences in a particular area (e.g., earthquakes, hurricanes, oil spills, and terrorist attacks). In contrast, the COVID-19 pandemic is a global and long-lasting emergency state with stress factors of changing intensities and differential effects unfolding over time. There is still a lack of theoretical models on both the effects of this kind of disaster on mental health and on factors that make individuals more vulnerable to suffer negative mental health consequences.

Research on the effect of the COVID-19 pandemic on child and adolescent mental health

Since the COVID-19 pandemic began in spring 2020, a lot of research into its mental health consequences has been conducted. Some reviews and meta-analyses have already tried to summarise this enormous body of research. For instance, Santomauro et al. [34] estimated in their systematic review that there has been an increase of 27.6% in depression and of 25.6% in anxiety disorders in the general population due to the pandemic. In addition, a number of reviews and meta-analyses which assessed the psychological consequences of the COVID-19 pandemic on different age groups have found a strong increase of mental health problems in children and adolescents [9, 35,36,37,38,39,40,41].

These previous reviews and meta-analyses have included mostly cross-sectional studies. This makes it difficult to distinguish between pre-existing levels of psychopathology and the pandemic-specific consequences and can lead to an overestimation of effects.

Longitudinal studies are more appropriate for investigating these effects because they allow a direct detection of developmental pathways across time. Unfortunately, the few meta-analyses that use longitudinal assessments of mental health symptoms focus on the general population and not on children and adolescents in particular [42,43,44].

The current review

The current study fills in some gaps by reviewing only longitudinal and repeated cross-sectional studies on the mental health consequences of the COVID-19 pandemic on child and adolescent mental health that have been published over a course of two years. The objectives are to assess the pandemic’s effects on a broad spectrum of mental health outcomes, to learn about factors influencing these effects, to investigate long-term psychological consequences over the course of the pandemic, and to create a large sample of studies conducted in many different countries to obtain a higher generalisability of the findings.

This study might also be beneficial to disaster theory and research because the COVID-19 pandemic significantly differs from disasters frequently studied in terms of duration, intensity, and differential effects. Previously studied disasters include, for example, natural catastrophes (e.g., hurricanes, oil spills, floods), human-made technological disasters (e.g., the nuclear accident at Chernobyl), acts of mass violence and terrorism (e.g., the attack on the World Trade Center in 2001), and previous epidemics and pandemics (e.g., the H1N1 influenza (swine flu) pandemic) [21, 25,26,27, 30, 45].

The following research questions were investigated in this paper:

  1. (1)

    How did the prevalence of mental health problems in the general population of children and adolescents change from before the COVID-19 pandemic to during the pandemic?

  2. (2)

    How did the prevalence of mental health problems in the general population of children and adolescents develop over the course of the pandemic?

  3. (3)

    What are factors that influence the potential effects of the COVID-19 pandemic on child and adolescent mental health?

Mental health can be measured by both broader outcomes (i.e., quality of life, well-being, life satisfaction, psychological stress, and affect) and by specific symptoms of common mental health disorders in childhood and adolescence. These involve internalising symptoms (including depression and anxiety), externalising symptoms (including hyperactivity, inattention, and disruptive behaviour problems), psychosomatic complaints, addictive behaviour (including substance abuse and excessive use of electronic media), and post-traumatic stress symptoms. These symptom categories match commonly described outcomes in research discussed previously and cover a broad range of mental health problems that can occur in childhood and adolescence.

Methods

Literature research

This review is in accordance with the guidelines proposed by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) [46]. There is no registered protocol. The literature search was conducted in the PubMed, Web of Science, and APA PsycInfo databases, each consulted last on 12 January 2022. The following was our search strategy to identify studies that examined the effect of the COVID-19 pandemic on child and adolescent mental health: “(COVID-19 OR corona* OR SARS-CoV-2 OR novel coronavirus) AND (psychological OR psychopatholog* OR mental health OR psychiatric disorders OR interna* OR externa* OR depress* OR anxi* OR sleep* OR hyperact* OR ADHD OR attention OR autis* OR neurodevelopment* OR disruptiv* OR conduct OR defiant OR oppositional OR emotional OR behavio* OR addict*) AND (adolescent* OR child* OR youth OR minor* OR teenager* OR juvenile*)”. To ensure sufficient data, we initially included a broad range of mental health symptoms in the search strategy. The terms used were based on the strategy used in a published review on child and adolescent psychopathology in the context of social competence [47].

If available at the respective databases, we used automatic filters as follows: The search terms named above had to appear in the title and/or abstract of the studies, the studies had to be in peer-reviewed journal articles written in the English language and published between December 2019 and December 2021, the participants had to be aged between 0 and 18 years, and the methodology was restricted to quantitative designs.

We identified additional studies via the ancestry approach by examining reference lists of studies included in this review and reviews and meta-analyses on the same topic. The first author manually screened the identified studies multiple times based on titles, abstracts, and methodology (i.e., longitudinal, repeated, or one-time cross-sectional). Eligible studies were then textually reviewed, and the results were summarised in a table. The second author confirmed the literature search and inclusion process. This research procedure is equivalent to other systematic reviews (e.g., [47]).

We filtered the results further using the following inclusion criteria: The research question should focus on the effect of the COVID-19 pandemic (i.e., by comparing measurements prior to and during the pandemic) or a pandemic-related stressor (e.g., confinement, quarantine) on a mental health outcome. Mental health could be assessed via global measures of psychological distress, quality of life, life satisfaction, affect, and well-being or via measures of specific symptoms of common mental disorders in childhood and adolescence.

Due to the large body of research found, we placed a closer focus on the following symptom groups: internalising symptoms, including depression, anxiety, and post-traumatic stress disorder; externalising symptoms, including hyperactivity, inattention, and disruptive behaviour problems; psychosomatic complaints; and addictive disorders such as substance abuse or excessive use of electronic media.

Because changes over time are more directly detectable in longitudinal research and the general population of children and adolescents is best represented in community samples, we decided to include only studies with a longitudinal or repeated cross-sectional research design with at least one measurement during the pandemic and only studies that used community samples when it became clear that enough of such studies were available. Because most identified studies assessed mental health outcomes via self- or others-report questionnaires, we also decided to include only studies that used this methodological approach to gain higher homogeneity and thus better comparability among studies.

Exclusion criteria were cross-sectional research designs with a one-point assessment of mental health, qualitative assessments of mental health outcomes, clinical sample use (i.e., children and adolescents with pre-existing mental or physical health disorders), or specific sub-sample use (e.g., young athletes or LGBTQ+ youth). Similarly, studies with a predictor only loosely connected to the COVID-19 pandemic (e.g., use of electronic media) and/or an outcome not focussed on mental health (e.g., family functioning, academic learning, or lifestyle changes) met the exclusion criteria. Research assessing intervention or prevention programmes, or diagnostic instruments was also excluded.

The process of study selection is depicted in Fig. 1.

Fig. 1
figure 1

Adapted from Page et al. 2021 [48]

Process and results of literature research.

Full size image

Synthesis of results

To synthesise the body of research that supports changes in distinct mental health outcomes more explicitly, we sorted the findings of studies comparing measures before and during the pandemic by examined outcome (e.g., depressive symptoms) and reported change in outcome (i.e., increase, no change, or decrease from before to during the pandemic). We calculated and tabulated the number of studies that examined each mental health indicator and the number of findings for each possible direction of change.

Wherever possible, we grouped together the outcomes that measured overlapping constructs and symptom categories. The larger pools of studies for each indicator allow for more profound conclusions. We summarised mental health indicators from outcomes as follows:

  • “Psychological well-being” and “quality of life” were grouped together (c.f., [49]).

  • Total scores of scales assessing multiple mental health outcomes were summarised as “global measures of mental health problems”.

  • “Internalising symptoms/problems” and “emotional symptoms/problems” were summarised as “internalising symptoms” (c.f., [50, 51]).

  • “Externalising symptoms/problems” and “behavioural symptoms/problems” were summarised as “externalising symptoms” (c.f., [50, 51]).

  • “Conduct”, “oppositional”, “challenging”, and “impulsive behaviour symptoms” were summarised as “disruptive behaviour symptoms” (c.f., [52]).

  • “Inattention”, “attentional problems”, and “sluggish cognitive tempo” were summarised as “cognitive symptoms”.

In addition to those above, we included the following psychopathological indicators and symptoms because they appeared in studies chosen for this review: resilience, suicidal symptoms, neurotic symptoms, dissociative symptoms, eating disorder symptoms, and psychotic symptoms.

To examine the overall directions of changes in mental health, we added the numbers of findings that supported each of increase, no change, and decrease in a mental health indicator. We grouped the outcomes as “indicators of good mental health” if they were positively associated with mental health (e.g., well-being, quality of life, life satisfaction) and “indicators of mental health problems” if they were negatively associated with mental health (e.g., stress, negative affect, psychopathological symptoms). We synthesised the results of studies of changes over the course of the pandemic separately. The emphasis was thereby on findings on the effects of certain health protection measures (i.e., lockdown, confinement) and the general severity of restrictions on mental health outcomes.

To examine the factors that influence the effect of the pandemic on mental health (i.e., moderators, mediators, correlates), we collected those that were reported in the studies as “COVID-19 related stressors” or “other variables” then tabularly sorted them into the following factor categories:

  • Sociodemographic

  • Intra-individual

  • Parental/family

  • Social

  • COVID-19-related

  • Behavioural

We used two additional approaches post-hoc to review the studies’ findings and identify sources of variance among outcomes. We sorted the studies that examined changes from before to during the pandemic by both age of the participants (<  six years, six–12 years, > 12 years) and country where the study was conducted to investigate systematic effects. To evaluate age effects, we counted findings of an increase, no change, or decrease in each mental health indicator separately. To evaluate the influence of region, we sorted findings by direction of change in mental health and not by specific indicators due to the small number of studies for each country.

Results

Sample characteristics

In total, we included 69 studies, published between September 2020 and December 2021 that assessed around 130,000 participants in this review. Fifty-eight of the studies used a longitudinal, within-subjects design. Thirteen studies applied a repeated cross-sectional, between-subjects design (e.g., cohort designs, matched convenience samples). Two studies analysed both within-subjects (i.e., intra-individual development) and between-subjects comparisons (i.e., age group comparisons) [53, 54].

All included studies conducted their in-pandemic assessment between February 2020 and June 2021, which covers a period of 17 months. Most were carried out in the months of April (n = 33), May (n = 37), and June 2020 (n = 29), which correlates with the first wave of the pandemic and the first lockdown in many countries. Far fewer studies were conducted after the summer of 2020 and in the first half of 2021. In 45 of the 69 included studies, data were collected during lockdown, confinement, or stay-at-home orders. Fourteen studies assessed children’s and adolescents’ mental health indicators and/or symptoms multiple times during the pandemic [53,54,55,56,57,58,59,60,61,62,63,64,65,66].

In total, our review included data from 21 countries, comprising 14 European, four Asian, two North American countries, and one South American country. Most studies were conducted in the United States of America (n = 14), China (n = 8), the United Kingdom (n = 7), Germany (n = 6), and Canada (n = 6).

All age groups between three and 18 years were included in this review. Early teenage years (i.e., 12–14 years) appeared most frequently in the study population, while research investigating effects on early childhood was rather rare. More concretely, nine studies included participants aged six years or younger, 34 studies included participants aged six–12 years, and 44 studies included participants aged above 12 years. Note that some studies fell into multiple groups.

Overview of included studies

Table 1 displays the 58 longitudinal studies and the two studies using both longitudinal and repeated cross-sectional approaches. Table 2 depicts the remaining 11 repeated cross-sectional studies.

Table 1 Overview of longitudinal studies (Same Sample)
Full size table
Table 2 Overview of repeated cross-sectional studies (different samples)
Full size table

Changes in mental health due to the COVID-19 pandemic

Table 3 displays the mental health indicators and symptoms assessed in the studies with a pre-pandemic measurement and how frequently the respective indicators and symptoms were assessed. Table 3 also summarises the studies’ findings concerning the change of the prevalence or intensity of the respective mental health indicators and symptoms from before the pandemic to during the pandemic. The seven studies without a pre-pandemic measure are not included in the table [56, 59,60,61, 63,64,65].

Table 3 Summary of main results of studies comparing pre-pandemic levels of mental health with measures during the pandemic
Full size table

The results of the 14 studies that investigated changes over the course of the pandemic are mixed. The overall tendency, however, indicates a decrease in mental health symptoms (internalising and externalising symptoms, stress, depression, anxiety, conduct problems, attention problems, problematic smartphone and Internet use) and an increase in well-being when we compared assessments taken during periods of lockdown/stronger restrictions and higher infection rates with those taken during or after the lift of restrictions and easing of the pandemic situation [55,56,57, 59,60,61, 63, 64, 66]. However, eight studies did not find a significant change in certain symptoms (i.e., internalising and externalising symptoms, depression, anxiety, attention problems, substance use) and thus do not completely support this trend [53, 55, 57,58,59,60,61, 65]. Two of those studies even reported an increase in certain mental health symptoms (i.e., eating disorder symptoms, hyperactivity, conduct problems) [60, 61]. One study that compared measurements between a period of lifted regulations and a following period of reinforced restrictions reported mainly an increase in symptoms of poor mental health (i.e., anxiety, depression, and psychosomatic symptoms, decrease in health-related quality of life) [54].

Although most studies reported a decrease in at least some psychopathological symptoms after an ease of the pandemic situation, it is less clear if children and adolescents return to a pre-pandemic level of functioning. Of the seven studies with both a pre-pandemic measurement and multiple during-pandemic measurements, five reported an increase in some symptoms from the pre-pandemic measure to assessment during loosened restrictions within the pandemic [53,54,55, 57, 62]. Three studies reported a decrease in symptoms from pre-pandemic levels to during less-restricted periods of the pandemic [53, 57, 58]. Three studies did not find a difference between the two time points [55, 57, 66]. Note that some of the seven studies assessed multiple outcomes and reported multiple directions of change (e.g., decrease in depression, anxiety, sluggish cognitive tempo and oppositional/defiant symptoms and no change in hyperactivity and impulsivity symptoms from spring 2020 to summer 2020 [55]).

Factors that influence COVID-19-related changes in mental health

Table 4 summarises factors reported to be associated with mental health outcomes during the COVID-19 pandemic. For simplicity, we classified the factors as “risk factors” if they were associated with more negative mental health effects and as “protective factors” if they were associated with less severe negative mental health outcomes.

Table 4 Risk and protective factors of the COVID-19 pandemic’s mental health effects
Full size table

Table 5 portrays the change of mental health indicators and symptoms from before to during the pandemic and how frequently these symptoms were assessed in three separate age groups (< 6 years, six–12 years, > 12 years).

Table 5 Summary of the main results of studies that compare pre-pandemic levels of mental health with levels measured during the pandemic sorted by age group
Full size table

Table 6 summarises regional differences in mental health outcomes by listing the countries with more than three studies in this review.

Table 6 Change of mental health from pre-pandemic to during the pandemic sorted by country
Full size table

Discussion

The COVID-19 pandemic and the protection measures to contain its spread have massively changed the daily lives of billions of children and adolescents worldwide. Here we present a scoping literature review on the longitudinal effects of the COVID-19 pandemic on child and adolescent mental health. There is a lack of reviews and meta-analyses that (a) investigate longitudinal changes in child and adolescent mental health, and (b) include studies conducted after the initial months of the pandemic. We executed the current review according to PRISMA-ScR [46].

Sixty-nine longitudinal and repeated cross-sectional studies published between September 2020 and December 2021 assessing child and adolescent mental health over the first 17 months of the pandemic are included. We have assessed changes in a range of both broad mental health indicators and specific psychopathological symptoms from before to during the pandemic and over its course. Furthermore, we summarised factors that influenced these effects.

Changes in mental health from before to during the pandemic

Among all the indicators of mental health problems, such as psychological distress, negative affect, and specific psychopathological symptoms, 84 findings of an increase in a single mental health indicator were reported. However, almost as many findings did not support an increase in mental health problems, with 49 showing no change and 27 showing a decrease in such an indicator. The findings for the change of broad measures of poor mental health (i.e., outcomes not specific to a certain indicator or symptom group, e.g., total scores of mental health questionnaires) are also mixed but more in line with an increase in these constructs. Five findings support this trend compared to four that do not. On the other hand, there is clearer support for an increase in the indicators “stress” and “negative affect”. Each had only one finding not supportive of an increase, compared to four and three findings, respectively, that support an increase.

The findings for indicators of good mental health, such as quality of life, life satisfaction, positive affect, resilience, and well-being, are more homogeneous and support a decrease in mental health from before to during the pandemic (14 supportive findings vs.  six non-supportive findings). The trend for global measures of internalising symptoms is less clear and would rather suggest no change (nine supportive findings) or an increase (eight supportive findings) than a decrease in symptoms (five supportive findings).

Negative effects of the pandemic are clearer if symptoms of individual internalising syndromes are inspected separately. The support for an increase in depressive symptoms is strong (19 supportive findings vs. 10 non-supportive findings). This increase fits the presumed effect based on previous literature and appears logical in the face of enforced isolation, lack of social contacts, pandemic-related worries, and loss of many positive leisure time activities and daily structure. However, there is more variability in data regarding anxiety symptoms with 13 studies finding an increase in symptoms and the same number of studies not reporting an increase in symptoms.

A rise in the general scores of externalising symptoms from before to during the COVID-19 pandemic is supported by most of the reviewed studies assessing these symptoms (eight supporting findings vs. six non-supporting findings). When specific externalising syndromes are regarded separately, different results can be observed. While symptoms of hyperactivity and inattention seem to have increased (hyperactivity: five supportive findings vs. four non-supportive; inattention: five supportive findings vs. two non-supportive), there is greater evidence for no change in the prevalence of disruptive behaviour symptoms (six supportive findings) than for an increase (four supportive findings) or a decrease (two supportive findings).

The increase in hyperactive symptoms can be explained by the restricted opportunities for physical activity during the pandemic due to stay-at-home orders, closures of schools, sport clubs, playgrounds, gyms, public swimming pools, etc., and the cancellation of leisure time activities associated with physical exercise. The lack of evidence indicating an increase in disruptive behaviour problems can be explained by the stability of underlying disorders and dispositions. Furthermore, the display of such symptoms is often situation-specific, meaning symptoms only become evident in certain social settings, for example, in school. Exposure to these situations was reduced during the pandemic due to stay-at-home orders, and school closures potentially led to limited chances to exercise and observe these problematic behaviours.

In line with previous research, the findings show an increase in mental health problems and symptoms of depression and anxiety in children, adolescents, and adults in the context of disasters in general [21, 22, 25, 28] and the COVID-19 pandemic in particular [9, 34, 35, 39,40,41, 122, 123]. However, due to highly heterogeneous data, the few existing meta-analytic studies regarding mental health effects of the COVID-19 pandemic based on longitudinal data [42,43,44] indicate no or only slight changes in mental health in the general population. This high variability in findings across studies is also observed in this review, in particular regarding the more global measures of mental health indicators (e.g., internalising symptoms). Negative effects of the pandemic on children’s mental health became more detectable when specific symptom groups were examined separately and when studies were sorted by age of participants. This fits with the assumption that effects might differ across different regions, social groups, and contexts, which we discuss below.

Changes in mental health during changes in pandemic intensity and restrictions

According to the literature, it can be expected that pandemic-related health protection measures, in particular confinement and quarantine, have a negative effect on child and adolescent mental health [37, 124,125,126]. Consequently, a reduction in restrictions over the course of the pandemic should be associated with decreasing symptoms of poor mental health. However, not all reviewed studies that assessed the changes in mental health indicators and symptoms over the course of the pandemic support this notion. Nonetheless, more findings indicate a decrease in symptoms after a lift of restrictions (n = 9) than no change in symptoms (n = 8) or an increase in symptoms (n = 2).

This conclusion is in line with data from the newest wave of the study by Ravens-Sieberer et al. [127]. This German study shows that the levels of health-related low quality of life and internalising symptoms in autumn 2021 were higher than pre-pandemic levels. However, health-related quality of life and mental health improved from spring 2020 and winter 2020/21 to autumn 2021. The study authors explain this effect by lower infection rates, higher vaccination rates, and loosening of restrictions.

Reasons for the negative effect of quarantine, confinement, and lockdown on child and adolescent mental health are diverse. For example, Mohler-Kuo et al. [128] found in their survey on stress factors during lockdown in Switzerland that the most common sources of stress were the disruption in social life, the breakdown of normal routines, the cancellation of important plans and events, and the uncertainty and unpredictability of the duration of the pandemic. These were in addition to distressing news coverage, rapidly changing recommendations, and fear of infection and the pandemic itself. Overall, it must be considered that protection measures usually correlate highly with pandemic intensity and infection rates. Contrarily, restrictions are usually lifted when infection rates ease. Thus, a unique contribution of the lifting of containment measures independent from infection rates cannot be assumed [129].

Changes in mental health related to individual risk and protective factors

Overall, the reviewed studies suggest that low socio-economic status, financial worries, material hardship, lack of space, negative home-schooling experience, bad physical health, and diagnosis of a neurodevelopmental disorder are the key risk factors for experiencing stronger negative mental health effects due to the pandemic (see Table 5). There is also evidence that children and adolescents reported to have experienced constant high levels of mental health problems before and during the pandemic (due to early childhood stress, maltreatment experiences, certain chronic mental health problems, special needs, and socio-economic disadvantage) were less likely than their healthy peers to display an increase in mental health problems in response to the onset of the pandemic [63, 77, 86, 88]. Their mental health was also mostly unrelated to the changing infection rates and health protection measures. This means that they did not benefit from the lifting of restrictions or the easing of the pandemic situation as their same-aged peers did. This is a rather unexpected finding in light of previous research that has proposed that factors such as prior traumatic experiences or dependency on special psychological support enhance the risk of experiencing negative mental health effects in the face of a disaster [21, 27, 41].

We further investigated age effects by grouping the reviewed studies by participant age (see Table 4). Most evidence for an increase in mental health problems (59 supportive vs. 50 non-supportive findings) and a decrease in indicators of good mental health (14 supportive findings vs. four non-supportive findings) was reported for the adolescent age group (i.e., individuals approximately 12–18 years of age). Within this age group we found strong and convincing support for an increase in depressive (17 supportive findings vs.  nine non-supportive findings) and anxiety symptoms (13 supportive vs. seven non-supportive findings), and a decrease in quality of life and well-being (eight supportive findings vs. two non-supportive findings).

A general trend of increasing mental health problems also appeared in school-aged children (i.e., individuals approximately six–12 years of age) with 37 studies reporting an increase in indicators of mental health problems, 22 not finding a change in symptoms, and 17 reporting a decrease in mental health problems. In contrast to adolescents, we found no clear evidence for an increase in depressive and anxiety symptoms in this younger age group.

Hence, adolescents might have suffered more strongly from reduced contact with peers and from heightened demands for personal responsibility (e.g., self-directed learning). Additionally, literature on the psychological development in childhood and adolescence states that adolescents are more vulnerable to social stressors, such as isolation and loneliness [130] and that symptoms of affective disorders rise significantly in adolescence [131, 132]. This age-related difference might also be partially due to the increasing introspective ability of adolescents that allows them to report more reliably than do children on internalising symptoms. Another reason might be that parent-reports, typically associated with an underreporting of internalising symptoms, were used for children but not for adolescents.

For younger children (i.e., individuals approximately under six years of age), most studies do not indicate a change in mental health problems (11 supportive vs. eight non-supportive findings). However, this age group is severely under-investigated. Moreover, the reliability and validity of results might be limited due to difficulties in assessing symptoms in such young children and the reliance on parent- or educator-reports in this review. Therefore, it is not possible to draw clear conclusions for this age group, though it seems that it is less affected by negative mental health effects of the pandemic than are older age groups.

Gender effects were not consistently investigated and reported in the reviewed studies. Of the 14 studies finding and reporting a significant gender effect, 11 studies identified female gender as a risk factor for higher levels and/or a stronger increase in certain indicators of poor mental health and five studies identified male gender as such a risk factor (see Table 5). Females were reported to be at a higher risk for higher levels and/or stronger increases in internalising, anxiety and depressive symptoms, stress, and lower levels of well-being than males. At the same time, males seemed more prone to attention problems, addictive gameplay, and sharper decreases in quality of life and life satisfaction than females. These gender differences were nearly exclusively found in adolescent samples.

Regarding socio-economic variables, children and adolescents who grow up with a single parent and those who do not have siblings were particularly at risk to show decreased levels of mental health during the pandemic. Further, housing situations that provided only limited living space and no access to green spaces were associated with greater increases in and/or levels of mental distress. A low socio-economic status and family financial worries were also identified as important risk factors, especially if the children did not already experience heightened psychological strain prior to the pandemic.

Other individual factors that acted as risk factors for mental health include poor physical health prior to the pandemic, diagnosis of a neurodevelopmental disorder, and dysfunctional or lacking coping and emotion regulation strategies. Parental strain and psychological strain, particularly in the form of parental symptoms of anxiety, depression, and substance abuse were found to be important influential factors for child and adolescent mental health during the pandemic. Further risk factors for poor mental health in the family environment include negative parental coping strategies, dysfunctional parenting styles, and overall family stress and instability.

Pandemic-specific factors that led to more negative mental health effects were the perceived lifestyle impact of the pandemic-related policies on the one hand, and stress due to the pandemic situation on the other. The former includes negative changes in the parental job situation, the experience of isolation and loneliness, frustration and boredom, and negative experiences during home-schooling. The latter comprises fears of oneself or family and friends becoming infected with the virus, general health-related concerns, and worries due to the uncertainty of the pandemic situation.

Behavioural factors reported to contribute to negative mental health outcomes were prolonged screen time, less physical exercise, disrupted sleep patterns, and staying mainly inside during times of confinement. Factors reported to promote mental health or mitigate negative mental health effects include access to green spaces and time spent in nature during times of confinement. Good physical health and health-related behaviours such as regular physical exercise, a healthy diet, structured routines, and regular sleep patterns reportedly helped protect child and adolescent mental health. Children’s and adolescents’ positive and adaptive coping mechanisms and emotion regulation abilities and social support through families and friends have also been identified as protective factors. Family functioning and a positive family climate have been found to help reduce negative mental health effects.

While many of the risk factors named are hardly modifiable (e.g., socio-economic variables, chronic mental and physical health conditions, parental mental health and parenting abilities, pandemic-related stressors), it is important to point out that most of the described protective factors are behaviour-based and can be adjusted (e.g., regular physical exercise, regular sleep, contact with friends, spending time in nature). Consequently, even though there might be unpreventable factors that heighten the risk for negative mental health consequences, there are also multiple viable preventive measures that can foster positive coping and protect mental health.

Reasons for inconsistencies in findings

The variability in the study samples, for example, times and locations (i.e., country/region) of assessments, might be a strong factor in the inconsistency of findings among the reviewed studies. Differing phases of the pandemic and locations are related to variations in infection rates and health protection measures and to the duration and intensity of exposure to the pandemic at the time of assessment.

The timing of the assessments is particularly important in studies that investigated changes in symptoms over the course of the pandemic. An explanation for the fact that some studies did not find a decrease in symptoms after a lift of restrictions might be that recovery takes time and might not be immediately visible in the assessments. Some stressors named above, such as unpredictability of the situation and fear of infection, have most likely continued to be a burden on mental health even with lifted restrictions.

Additional influencing factors leading to higher heterogeneity in findings might be differences in sample characteristics (e.g., age, socio-economic status), methodological approaches (over 60 different assessment tools are represented), conceptualisation of mental health problems, and definition and operationalisation of outcome variables. An important reason for the heterogeneous findings on the effects of health protection measures might be that mental health is not only influenced by the severity of protection measures but also associated with other characteristics of the pandemic (e.g., infection rates, death tolls). It is difficult to distinguish the effects of the intensity of protection measures from those of the severity of infection rates and the perceived pandemic threat. For example, one of the reviewed studies compared mental health effects among three European countries and showed a greater increase in mental health problems in the countries with greater restrictions, which also had higher infection and death rates at the time [60].

In fact, it has been proven that both policy stringency and pandemic intensity affect mental health to a similar degree [129]. This means that minimising transmission of the virus and death rates by potentially stricter health protection measures might indeed be protective against negative mental health effects due to the pandemic’s intensity. It also seems more appropriate to use separate measures for distinct symptom categories (e.g., depressive symptoms, psychosomatic symptoms) than global measures (e.g., internalising symptoms) because the latter might be less suitable for accurately capturing a change in single symptoms. The same explanation can be applied to the heterogeneous findings for complex symptom groups (e.g., anxiety) that comprise many different forms of a syndrome (e.g., generalised anxiety, health-related anxiety, school-related anxiety) that might have developed differently during the pandemic (see e.g., 119). Finally, there is evidence for age-related differences in the change of mental health symptoms due to the pandemic. For instance, it seems that adolescents have experienced an increase in anxiety symptoms while children have not. Looking at these different effects altogether in the total study sample of the scoping review might have contributed to the heterogeneity of our findings.

Strengths and limitations of the current review

To our knowledge our review is the first to include only studies with multiple assessment waves, which allows for the detection of changes in mental health that can be clearly attributed to the COVID-19-pandemic. This is a huge advantage over previous papers that mainly relied on cross-sectional estimations of the prevalence of certain mental health indicators. Furthermore, this review goes beyond studies conducted in the early months to include research that assessed mental health over the first one and a half years of the pandemic. The large sample size of our study and inclusion of data from 21 countries on 4 continents allow for a higher generalisability of results. Conclusions drawn from this work are, thus, not restricted to the initial outbreak and beginning of the pandemic but are suited to estimate broader and more long-term effects.

There is potential for errors in the study selection because the identified studies were primarily assessed for eligibility, selected, and synthesised by only one researcher. Another reason for a possible distortion in the data might be the effects of publication bias due to, for example, the under-publication of non-significant results (the file drawer problem; see, e.g., [133, 134]). This review includes published and peer-reviewed articles only. We did not attempt to search for unpublished articles to ensure the scientific quality of the studies. There is a chance that the results reviewed here, therefore, represent a biased sample of relevant data. Because we did not statistically assess the size of the publication and reporting bias, we cannot estimate effects these biases may have on our results.

The conclusions drawn in this paper are most representative of western, educated, industrialised, rich, and democratic societies (WEIRD-bias, c.f., [135]) because most studies were conducted in North American and Western European countries. In the face of a global pandemic that has most likely affected socially and economically disadvantaged groups the most, this is a serious limitation to this review.

Most of the studies in the sample concentrated on the first months of the pandemic, March to June 2020. Furthermore, two-thirds of all included studies were conducted during stricter confinement rules. The lack of studies conducted both after summer 2020 and outside heightened restrictions means that the long-term effects of the pandemic situation, which has lasted much longer the initial period, are not completely evaluable.

The sample is most representative of early adolescent years and does not allow for clear statements about effects on younger age groups, especially early childhood.

A major limitation of this scoping literature review is that all included studies rely only on assessments of mental health via child or adolescent self-, caretaker-, or teacher-reports on questionnaires, which were presented online in most cases. None of the included studies used clinical evaluations by mental health professionals (e.g., diagnostic interviews), which would be necessary to form solid clinical diagnoses [136]. Moreover, the lack of studies using multiple informants to report on a child’s or adolescent’s mental health limits the validity of results. Consequently, the trends observed in this study do not allow for conclusions about the change of the prevalence of diagnoses but are rather suited to estimate changes in the prevalence of mostly self- or parent-reported symptoms.

We neither assessed nor accounted for the methodological quality of the reviewed studies. We also did not analyse the effect sizes of findings. Additionally, it is nearly impossible to distinguish the extent to which the reported effects were caused by the pandemic itself or by certain pandemic-related stressors, health protection measures, or methodological issues. This reduces the options for summarising the findings to the qualitative method of vote-counting, which is substantially inferior to quantitative statistical methods of synthesising research findings. Hence, the purpose of this paper is an accumulation and summary of the research currently available and an indication of general directions of changes in child and adolescent mental health due to the COVID-19 pandemic.

Implications for intervention strategies to meet the mental health needs of children and adolescents

This paper shows that children’s and adolescents’ mental health problems and symptoms of mental health disorders increased from before to during the pandemic. Missed chances and milestones of development and the delay in diagnosis and treatment will have long-term mental health consequences. It is obvious, therefore, that the need for psychological care for children and adolescents has risen and is still rising worldwide. The situation of mental health care was already precarious before the pandemic. It has now become even more painfully evident that there is a great dissonance between demand and availability of mental health care in many places worldwide. Therefore, it is important to invest in research, creation, and implementation of intervention strategies to react to this growing need for psychological care. Furthermore, this review has shown that it is important to balance health protection and infection control measures with child and adolescent protection and guarantee of societal participation.

In general, continuity and stability of access to school, mental and physical health care, and social services are essential for children’s and adolescents’ mental health. Further disruptions should be avoided at all costs. Decisions about the implementation of certain interventions should always be based on expertise and empirical research to allow for efficiency in both the prevention of infections and protection of children’s and adolescents’ mental well-being. Some concrete measures that could help to reach this goal are proposed in the following.

It is important to facilitate access to mental and physical health care and to social and community services to support children, adolescents, parents, and families in need. The pandemic has shown that there is oftentimes room for improvement in the online resources of such services, which can be especially helpful in cases where more physical distancing is required, for instance, in times of increasing infection rates or for individuals with physical health risks. Easily and quickly accessible psychosocial support services, such as cost-free helplines, can also promptly provide care to individuals in need even when in-person services should be less accessible.

It is crucial to prepare safe day-care and school environments for children and adolescents, for example, by providing air filters in classrooms and creating concepts for health protection in these institutions. These concepts could include recommendations for group and class sizes, personal hygiene measures, or regulations for mask wearing, vaccination, and testing. These measures are not only necessary to protect child and adolescent physical health but also to protect the health of teachers, thereby preventing cancellation of lessons and further strain on the education system.

Improved digital schooling is necessary to prepare for the worsening of the pandemic situation and for future emergency situations. This could also help children who have to stay home due to an infection to not lose touch with school. This measure includes, for example, extension of online structures, provision of online materials, and training in media competence for both teachers and students. Additionally, education authorities must ensure that all students possess the necessary tools and services at home to participate in online schooling (e.g., stable Internet connection, electronic devices).

Even if structures for online schooling are improved, it is important to provide concepts to support vulnerable families by, for example, guaranteeing access to in-person schooling or day-care for children and adolescents in these families. Special attention should be directed at vulnerable families with limited economic and social resources because these factors put not only children but also their parents at higher risk for experiencing psychological distress. Therefore, social services and structures are needed to reliably identify families in need and support them in receiving the help they need (e.g., parenting classes, financial aids, access to day-care).

It is known that the worldwide school closures have led to a loss of education so teachers should consider the deficits associated with home-schooling and school closures when it comes to evaluations, grading, and exams. Further, teacher sensitivity is needed to identify children and adolescents that might have particularly suffered due to the pandemic situation and connect them to support services. Another important task for both educators and parents is to provide children and adolescents with accurate information about the current pandemic situation and necessary health protection guidelines in age-appropriate ways. This can help children and adolescents understand the current situation, give them room to voice their worries and fears, and prevent the spread of misinformation and panic.

In the family context, it is not only important to consider child and adolescent mental health but also the mental health of parents who have also suffered from the pandemic situation. It is known that parent psychopathology is closely linked to their parenting abilities and children’s psychological distress [137]. Healthy, empathic, and friendly family interaction, communication, and parenting competence should be promoted and fostered by, for example, providing access to parenting programmes or family counselling [138]. Positive parent–child relationships, the presence of secure attachment figures, good caregiving, parenting that encourages emotional expression, acceptance, and positive reframing have been reported to protect child and adolescent mental health after exposure to a disaster [23].

Families should also be encouraged to establish structured routines that include time spent outside, physical activity, and regular sleep times, especially in times of home confinement or quarantine. It can be helpful to prepare programmes to animate children and adolescents to engage in physical exercise at home [137]. These approaches can help to counteract the negative effects of quarantine or home confinement on mental and physical health. Last, effects of loneliness and social isolation can be prevented by encouraging children and adolescents to stay in contact with same-aged peers by providing access to safe and age-appropriate ways of online communication [137]. Parents should respectfully monitor their children’s online activities to protect them from the dangers of the online world.

Implications for further research

The current study makes evident that there is need for more longitudinal or repeated cross-sectional studies that investigate the changes in certain symptom categories, because we did not identify enough studies to draw cohesive conclusions about their development. These categories include psychosomatic, post-traumatic stress, and compulsive symptoms, symptoms of substance abuse and addiction and symptoms of behavioural addiction to electronic media.

Our focus on the COVID-19-pandemic’s mental health effects on community samples of children and adolescents helps the generalisability of findings, however, it does not allow for conclusions about the pandemic’s effects on special vulnerable groups. These groups include children belonging to marginalised ethnic, religious, or social groups, children of frontline workers, children and adolescents growing up in low-income families, and children and adolescents suffering from chronic physical or mental health conditions, such as neurodevelopmental disorders. There is evidence that these groups of children and adolescents have been more strongly affected by the pandemic’s effects than the general population [138,139,140,141].

Lastly, further research into the COVID-19 pandemic’s mental health effects and into strategies for the mitigation of negative outcomes is important far beyond the current health crisis. There is a causal link between climate change and an increase in the frequency of pandemic outbreaks (e.g., [142, 143]). This means that if we fail to appropriately react and fight against the current climate crisis, we will have to prepare ourselves for future pandemics. Therefore, it is important and urgent to analyse the effects of the current pandemic in depth and to plan and prepare future actions accordingly based on facts and empirical evidence.

Conclusion

This scoping review of 69 longitudinal and repeated cross-sectional studies has demonstrated that mental health problems in children and adolescents have increased globally in the first 17 months of the COVID-19 pandemic compared to pre-pandemic data. We have shown that the mental health problems of children and adolescents are positively associated with both the intensity of the pandemic situation (e.g., infection rates, death tolls) and the severity of protection measures, such as confinement and quarantine. Among many other factors we have identified female gender, adolescent age, socio-economic disadvantage, parental psychopathology, dysfunctional family environment, social isolation and loneliness, loss of routines and structure, and the experience of distressing emotions due to health-related worries and uncertainty of the pandemic situation as increasing the chances of suffering negative mental health consequences.

Future research with assessments of the long-term effects of pandemic exposure and of the changes in clinical diagnoses is needed to extend the current findings. This is necessary to more concretely estimate the increased need for child and adolescent mental health care. The protection of children’s and adolescents’ mental health must be prioritised at all costs in a crisis like the COVID-19 pandemic to prevent life-long harm to future generations.