key: cord-0962412-x9veckdi
authors: Loades, Maria Elizabeth; Chatburn, Eleanor; Higson-Sweeney, Nina; Reynolds, Shirley; Shafran, Roz; Brigden, Amberly; Linney, Catherine; McManus, Megan Niamh; Borwick, Catherine; Crawley, Esther
title: Rapid Systematic Review: The Impact of Social Isolation and Loneliness on the Mental Health of Children and Adolescents in the Context of COVID-19
date: 2020-06-03
journal: J Am Acad Child Adolesc Psychiatry
DOI: 10.1016/j.jaac.2020.05.009
sha: f24ea00a06dad35c4ff70edbb79d60b681439fa4
doc_id: 962412
cord_uid: x9veckdi

OBJECTIVE: Disease containment of COVID-19 has necessitated widespread social isolation. We aimed to establish what is known about how loneliness and disease containment measures impact on the mental health in children and adolescents. METHOD: For this rapid review, we searched MEDLINE, PSYCHINFO, and Web of Science for articles published between 01/01/1946 and 03/29/2020. 20% of articles were double screened using pre-defined criteria and 20% of data was double extracted for quality assurance. RESULTS: 83 articles (80 studies) met inclusion criteria. Of these, 63 studies reported on the impact of social isolation and loneliness on the mental health of previously healthy children and adolescents (n=51,576; mean age 15.3) 61 studies were observational; 18 were longitudinal and 43 cross sectional studies assessing self-reported loneliness in healthy children and adolescents. One of these studies was a retrospective investigation after a pandemic. Two studies evaluated interventions. Studies had a high risk of bias although longitudinal studies were of better methodological quality. Social isolation and loneliness increased the risk of depression, and possibly anxiety at the time loneliness was measured and between 0.25 to 9 years later. Duration of loneliness was more strongly correlated with mental health symptoms than intensity of loneliness. CONCLUSION: Children and adolescents are probably more likely to experience high rates of depression and probably anxiety during and after enforced isolation ends. This may increase as enforced isolation continues. Clinical services should offer preventative support and early intervention where possible and be prepared for an increase in mental health problems.

The COVID-19 pandemic has resulted in governments implementing disease containment measures such as school closures, social distancing and home quarantine. Children and adolescents are experiencing a prolonged state of physical isolation from their peers, teachers, extended family and community networks. Quarantine in adults generally has negative psychological effects including confusion, anger, and post-traumatic distress. 1,2 Duration of quarantine, infection fears, boredom, frustration, lack of necessary supplies, lack of information, financial loss, and stigma appear to increase the risk of negative psychological outcomes. 1 Social distancing and school closures may therefore increase mental health problems in children and adolescents, already at higher risk of developing mental health problems compared to adults 3 at a time when they are also experiencing anxiety over a health threat and threats to family employment/income. Social distancing and school closures are likely to result in increased loneliness in children and adolescents whose usual social contacts are curtailed by the disease containment measures.

Loneliness is the painful emotional experience of a discrepancy between actual and desired social contact 4 Although social isolation is not necessarily synonymous with loneliness, early indications in the COVID-19 context indicate that more than one third of adolescents report high levels of loneliness 5, 6 and almost half of 18-24-year olds are lonely during lockdown. 7 There are well established links between loneliness and mental health. 8 The purpose of this review was to establish what is known about the relationship between loneliness and mental health problems in healthy children and adolescents and to establish whether disease containment measures including quarantine and social isolation are predictive of future mental health problems. We included cross sectional, observational, retrospective and case control studies if studies included mainly children and adolescents, who had experienced loneliness or had used validated measures of social isolation and mental health problems. To capture the possible effects of social isolation and the expected mediator (loneliness) on mental health problems, we included search terms to capture these two areas.

We conducted a rapid review to provide timely evidence synthesis to inform urgent healthcare policy decision-making. 9 A rapid review adheres to the essential principles of systematic reviews, including scientific rigour, transparency, and reproducibility. 9,10 It uses "abbreviated" systemic review methodology including: limiting search criteria, faster data extraction, and using narrative synthesis methods. 11, 12 Search strategy and selection criteria [see Table S1 , Table S2, Table S3 for full search strategy]

We searched MEDLINE, PSYCHINFO, Web of Science and the Cochrane Library. Our search terms were informed by recent rapid reviews in the COVID-19 context 1 and included definitions of loneliness and social isolation to capture the impact of social distancing and school closures. Terms captured 'children' or 'adolescents' AND 'quarantine' or 'social isolation' or 'loneliness' AND 'mental health' with a focus on the most common mental health problems in this age group: depression and anxiety.

Peer reviewed studies were selected if they were published (1946 to 03/29/2020); reporting primary research; included predominantly children/adolescents (mean age < 21) 13 ; published in English (web of science only); participants had experienced either social isolation or loneliness; valid assessment of depression, anxiety, trauma, OCD, mental health, or mental wellbeing.

We checked 20% of all study eligibility results (both included and excluded) to ensure adherence to the eligibility criteria. Data were extracted into a purpose-designed database: A random 20% of the data was double entered to ensure accuracy.

A truncated quality assessment was conducted by one author (SR) using criteria adapted from the NIH 14 (see table 1 ).

[insert table 1 here]

We conducted a narrative synthesis within the following categories: (1) the impact of loneliness on mental health in healthy populations (further divided into cross-sectional and longitudinal evidence),

(2) pandemic-specific findings, and (3) intervention studies.

We located 4531 articles (see Figure 1 ) of which 83 articles (80 studies) met the inclusion criteria. Of these, 18 articles (17 studies) reported on the impact of loneliness in those with a variety of health conditions including mental health problems (12 studies), physical health problems (1 study) and neurodevelopmental conditions (4 studies). The remaining 65 articles reported on 63 studies which examined the impact of loneliness or disease containment measures on healthy children and adolescents. For the purposes of this rapid review, we will focus our analyses on these 63 studies. The 63 studies were mainly from the USA, China, Europe and Australia. Included studies were also conducted in India, Malaysia, Korea, Thailand, Israel, Iran, and Russia. 61 studies were observational and 2 studies reported on interventions. Of the 61 observational studies, 43 studies were crosssectional only, 6 longitudinal only and 12 reported both cross sectional and longitudinal findings. 1 study was a retrospective study after a pandemic. In cross sectional studies, likely confounders (e.g. adversity, SES) were rarely controlled, meaning that the association between loneliness and mental health outcomes in these studies is very likely to be inflated 16 . Four longitudinal studies used multiinformant approaches including self-report and parent and/or teacher report to assess mental health outcomes. Importantly, they typically assessed and controlled for confounds and could assess the most plausible direction of causality between loneliness/social isolation and mental health. Tables 2 and 3 describe the 60 studies which examined the impact of loneliness on mental health. 53 studies stated that they measured the impact of loneliness on mental health. 7 studies stated that they measured the impact of social isolation 17-23 on mental health, but the social isolation measures used were either subscales or questions from loneliness scales, or strongly overlapped with the construct of loneliness. Therefore, we have considered them together with studies that measured loneliness. Participants were mainly school or university students or taking part in longitudinal cohort studies.

Forty-five studies examined the cross-sectional relationship between depressive symptoms and loneliness and/or social isolation. 17, 20, 21, The majority were conducted in adolescent (N = 23) and young adult (N = 16) samples, although six studies included children under the age of 10. Most reported moderate to large correlations (0.12 ≤ r ≤ 0.81) and most included a measure of depressive symptoms. Two studies reported odds ratios, with those who were lonely 5.8 45 to 40 times 49 more likely to score above clinical cut-offs for depression. The associations were stronger in older participants 37 and in female participants. 46 However, the strength and direction of the associations did not differ by age of the sample. Fewer studies (N = 23) examined symptoms of anxiety. Those that did found small to moderate associations between anxiety and loneliness/social isolation (0.18 ≤ r ≤ 0.54). The duration of loneliness was more strongly associated with anxiety than intensity of loneliness. 42,67 Social anxiety was moderately to strongly associated with loneliness/social isolation (0.33 ≤ r ≤ 0.72) and there were moderate associations between generalized anxiety and loneliness/social isolation (r = 0.37, 0.40). 21, 34 One study found a small association between panic and loneliness (r = 0•13). 61, 62 In the single study which reported odds ratios, being lonely was associated with increased odds of being anxious by 1•63 to 5•49 times. 49 Positive associations were also reported between social isolation/loneliness and suicidal ideation, 24,27,28 self-harm, 24 and eating disorder risk behaviour. 24 Negative associations were reported between social isolation/loneliness and wellbeing 68,69 and mental health. 22 Eighteen studies followed participants over time (see table 3 ). [17] [18] [19] [55] [56] [57] [58] [60] [61] [62] [63] 65, [70] [71] [72] [73] [74] [75] Several of these were conducted in childhood (N = 6), or adolescence (N = 8), although three were in university students. Most (N = 12) had only one follow up time point usually between 1 and 3 years.

12 of the 15 studies found that loneliness is associated with depression and explained a significant amount of the variance in severity of depression symptoms several months to several years later. 55, [57] [58] [59] [60] [61] [62] [63] 71, 73, 74 Two studies found that loneliness in childhood at age 5 was not associated with depression several years later 59,60 although other studies which assessed loneliness during childhood found evidence that it is associated with subsequent depression 55,72 One large study of adolescents (n=3088) found that loneliness was not associated with depression one year later. 56 There were mixed findings in another large study of adolescents (n = 541) which found a significant association between loneliness and subsequent depression, although this did not hold in a cross-lagged model 17 suggesting a possible bidirectional relationships between the variables. A study of university students found evidence of a gender difference, with loneliness being associated with later depression in female participants but not in male participants. 18 In a large longitudinal cohort of   vulnerable young people, aged 11 to 17, after controlling for caregiver neglect and other relevant covariates, a substantial increase in self-reported peer isolation (1 S.D.) was associated with an increase in depression symptoms (0.49 S.D.). 71 Duration of peer loneliness rather than the intensity of peer loneliness is associated with depression 8 years later (i.e. from age 5 to age 13); in contrast family related loneliness was not independently associated with subsequent depression. 59

Three of the four studies which examined the longitudinal effect of loneliness on anxiety found that loneliness was associated with later anxiety. 56,64,75 Two of these studies assessed social anxiety, and one measured anxiety as a broad construct. One study did not find that loneliness/social isolation at age 5 was associated with anxiety at age 12. 19 One study of young adolescents found differences by gender, with loneliness being associated with later social anxiety in male participants but not female participants. 75 None of these studies measured loneliness during childhood.

Other mental health outcomes reported over time included internalizing symptoms which were associated with prior loneliness in primary school age children, 72 and suicidal ideation during adolescence, which was not associated with prior loneliness during childhood. 60 

One study 76 reported on mental health and social isolation in the context of different infections including H1N1, SARS, and avian flu (see table 2 ). This retrospective study included 398 parents of exposed children from the USA, Canada and Mexico, of whom 20•9% experienced social isolation and a further 3.8% had been quarantined. Parents of children reported on their child's experience of trauma and on their current mental health. One third of parents whose children had been subject to disease containment measures said their child had needed mental health service input because of their pandemic related experiences. The most reported diagnoses were acute stress disorder (16.7%), adjustment disorder (16.7%), grief (16.7%), and PTSD (6.2%). Two different parent-reported measures of PTSD symptoms found that those children exposed to disease containment measures scored significantly higher for PTSD symptoms post-pandemic. On the PTSD Checklist Civilian Version, 28% of children who had experienced isolation/quarantine scored about the cut-off for PTSD, compared to 5.8% of those who had not experienced isolation/quarantine. Similarly, on the UCLA PTSD Reaction Index, 30% of children who experienced isolation/quarantine scored about the cut-off for PTSD, compared to 1.1% of those who had not experienced isolation/quarantine (effect size: Cramer's V = 0.449). Mean scores were 4 times higher in the isolated/quarantined group than in those who had not been isolated/quarantined. The most common trauma symptoms in the quarantined/isolated group were avoidance/numbing (57.8%), re-experiencing (57.8%), and arousal (62.5%).

Two randomised control trials measured loneliness and mental health outcomes following an intervention aimed at the general population (peer mentoring 77 and classroom based, 78 see table 4).

In both instances the comparator was no intervention/with follow-up and education as usual. A relatively intensive peer mentor program, with an adult mentor, 4-6 hours per month for 4 months on average, reduced loneliness and mental health problems (small to medium effects) for victims of bullying and victimization. However, a brief (two session) universal classroom-based program delivered in schools including psychosocial support through peer mentors and a staff mental health support team did not reduce loneliness. Neither intervention specifically addressed mental health problems which had developed in the context of loneliness; therefore we are unable to answer our second review question which was what interventions are effective for those who have developed mental health problems as a result of social isolation or loneliness.

This rapid systematic review of 63 studies of 51, 576 participants found a clear association between loneliness and mental health problems in children and adolescents. Loneliness was associated with future mental health problems up to 9 years later. The strongest association was with depression.

These findings were consistent across studies of children, adolescents, and young adults. There may also be gender differences with some research indicating that loneliness was more strongly associated with elevated depression symptoms in girls and with elevated social anxiety in boys. 18, 75 The length of loneliness appears to be a predictor of future mental health problems 59 . This is of particular relevance in the COVID-19 context as politicians in different countries consider the length of time that schools should remain closed, and the implementation of social distancing within schools.

Furthermore, in the one study that examined mental health problems after enforced isolation and quarantine in previous pandemics, children who had experienced enforced isolation or quarantine were five times more likely to require mental health service input and experienced higher levels of post-traumatic stress. This suggests that the current social distancing measures enforced on children because of COVID-19 could lead to an increase in mental health problems, as well as possible posttraumatic stress. These results are consistent with preliminary, unpublished data emerging from China during the COVID-19 pandemic where children aged 3 to 18 are commonly displaying behavioural manifestations of anxiety including: clinginess, distraction, fear of asking questions about the pandemic, and irritability 79 Furthermore, a large survey of young adult students in China has reported that around one in four are experiencing at least mild anxiety symptoms 80 In the UK, early results from the Co-SPACE (COVID-19 Supporting Parents, Adolescents and Children in Epidemics) online survey of over 1500 parents suggest high levels of COVID-19-related worries and fears, with younger children (age four to 10) significantly more worried than older children (age 11 to 16) . 81, 82 In addition to the more direct effects of enforced isolation and quarantine, loneliness as an unintended consequence of disease containment measures seems to be particularly problematic for young people 5, 7 . This may be because of the particular importance of the peer group for identity and support during this developmental stage. 83, 84 This propensity to experience loneliness may make young people particularly vulnerable to loneliness in the COVID-19 context, which, based on our findings, may further exacerbate the mental health impacts of the disease containment measures.

More studies have examined the relationship between loneliness and depression than loneliness and anxiety. Losing links to other people and feeling excluded can result in an affective response of depression. 85 Social anxiety was more strongly associated with loneliness than other anxiety subtypes. This may be because social anxiety is triggered by a perceived threat to social relationships or status. 86 It is difficult to predict the effect COVID-19 will have on the mental health of children and young people. The subjective social isolation experienced by participants did not mirror the current features of social isolation experienced by many children and adolescents worldwide. Social isolation was not enforced upon the participants, nor was social isolation almost ubiquitous across their peer groups and across the communities in which they live. As loneliness involves social comparison, 87 it is possible that the shared experience of social isolation imposed by disease containment measures may mitigate the negative effects. The studies were also not in the context of an uncertain but dangerous threat to health. These features limit the extent to which we can extrapolate from existing evidence to the current context. In order to make evidence based decisions on how to mitigate the impact of a second wave, we need further research on the mental health impacts of social isolation in the disease containment context of a global pandemic. In this context, to more specifically understand the impacts of loneliness, measures such as the Loneliness and Aloneness Scale for Children and Adolescents (LACA) that assess the duration and the intensity of loneliness, and that separate peer-related loneliness from parent-related loneliness could be elucidating.

This rapid systematic review was conducted rapidly, in 3 weeks, to inform our response to COVID-19.

We double screened 20% of all articles and data extracted. In line with Cochrane rapid review guidance, 10 grey literature, and trial registry databases were not searched, hand-search strategies were not employed, and only English language publications were included, meaning that some relevant studies may have been missed. During the rapid data extraction phase, there was no scope to contact authors to request any missing information. The main limitation from this review is the lack of high-quality studies investigating mental health problems after enforced isolation. All but one study investigated social isolation that was not enforced on young people and was not common across a peer group. The effect of widespread social distancing could mitigate against the social isolation described with increased use of internet mediated relationships which can be beneficial to adolescents. 88 . Most studies were cross-sectional, and therefore the direction of the association cannot be inferred. Few studies used independent (i.e. not self-report) measures of mental health or social isolation/loneliness, increasing the risk of bias. Furthermore, the studies were mainly observational and did not consistently control for potential confounders. The majority of studies focused on depression and anxiety, and other mental health problems are important to measure in future research.

However, we used all available evidence on social isolation and loneliness to inform the likely outcome for healthy children and adolescents subjected to social isolation. The results were consistent across all study methodology for depression, (but less so for anxiety) suggesting these results are reliable. The results are also consistent with one study investigating mental health problems in children 76 after pandemics improving our confidence in the results. However, the post pandemic study has several limitations in that the sample was self-selecting, and the demographics of the children and the time elapsed since the experience were not reported. There is little evidence pertaining to interventions. We have focused on healthy populations in this review and will report on those with pre-existing conditions including mental health problems elsewhere.

The review indicates that felt loneliness is associated with adverse mental health in children and adolescents. There is limited evidence that indicates specific interventions to prevent loneliness or to reduce its effects on mental health and well-being. However, there are well-established practical and psychological strategies that may help promote child and adolescent mental health in the context of involuntary social isolation e.g. during the COVID-19 pandemic. Reducing the impact of enforced physical distancing by maintaining the structure, quality, and quantity of social networks, and helping children and adolescents to experience social rewards, feel part of a group, and know that there are others they can look to for support is likely to be important. 8 Finding ways to give children and adolescents a sense of belonging within the family and to feel that they are part of a wider community should be a priority. Therefore, providing accurate information about the relative risks and benefits of social media and networking to parents who overestimate the dangers of allowing their children too much screen time may help young people access the benefits of virtual social contact.

However, simply increasing the frequency of contact may not address young people's subjective experience of loneliness. 20 Helping young people to identify valued alternative activities and build structure and purpose into periods of involuntary social isolation may help to provide a wider range of rewards. 89 Addressing negative thoughts about social encounters (e.g. self-blame, selfdevaluation) may also be effective. 34,90 During periods of prolonged social isolation digital technology that provides evidence-based interventions to help young people to reappraise their thoughts and change their behaviour within the confines of the home setting may be particularly welcome.

Whilst this review did not provide evidence on interventions to improve social isolation or loneliness in healthy children and adolescents, given social distancing, digital interventions may be appropriate. The rapid review suggests that loneliness that may result from disease containment measures in the COVID-19 context could be associated with subsequent mental health problems in young people.

Strategies to prevent the development of such problems should be an international priority. Note: ES = effect size; NS = not specified.

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

Mental health outcomes of Quarantine and isolation for infection prevention: A systematic umbrella review of the global evidence. psyarxiv.com/dz5v2

Prevalence of mental health problems in schools: poverty and other risk factors among 28 000 adolescents in England. The British journal of psychiatry : the journal of mental science

Toward a social psychology of loneliness

Achieving Resilience During COVID-19 Weekly Report 2

Abstracts from the Poster and Oral Presentations from the 18th Neonatal and Paediatric Pharmacists Group, NPPG Annual Conference. Archives of disease in childhood

Coping with Loneliness

Social isolation in mental health: a conceptual and methodological review

World Health Organisation AfHPaSR. Rapid reviews to strengthen ealth policy and systems: a practical guide

Cochrane Rapid Reviews. Interim Guidance from teh Cochrane Rapid Review Methods Group

Expediting systematic reviews: methods and implications of rapid reviews

A scoping review of rapid review methods

Mother Nature versus human nature: public compliance with evacuation and quarantine

Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement

Statistically Controlling for Confounding Constructs Is Harder than You Think

Loneliness, depressive symptomatology, and suicide ideation in adolescence: cross-sectional and longitudinal analyses

Gender differences in the influence of social isolation and loneliness on depressive symptoms in college students: a longitudinal study

Social isolation and mental health at primary and secondary school entry: a longitudinal cohort study

Social isolation, loneliness and depression in young adulthood: a behavioural genetic analysis

The Moderating Role of Sleep in the Relationship Between Social Isolation and Internalising Problems in Early Adolescence

Psychosocial predictors and outcomes of loneliness trajectories from childhood to early adolescence

The development of loneliness from mid-to late adolescence: trajectory classes, personality traits, and psychosocial functioning

The interplay of loneliness and depressive symptoms across adolescence: exploring the role of personality traits

Disentangling the Longitudinal Relation Between Loneliness and Depressive Symptoms: Prospective Effects and the Intervening Role of Coping

Family dysfunction and Adolescents' anxiety and depression: A multiple mediation model

Longitudinal association between low self-esteem and depression in early adolescents: The role of rejection sensitivity and loneliness

Does self-esteem mediate the relationship between loneliness and depression among Malaysian teenagers?

Loneliness, perceived social support, and anxiety among Israeli adolescents

The Influence of Loneliness and Interpersonal Relations on Latina/o Middle School Students' Wellbeing

It Hurts To Be Lonely! Loneliness and Positive Mental Wellbeing in Australian Rural and Urban Adolescents

Mitigate the effects of home confinement on children during the COVID-19 outbreak

The joint impact of parental psychological neglect and peer isolation on adolescents' depression

Loneliness as a partial mediator of the relation between low social preference in childhood and anxious/depressed symptoms in adolescence

Changes in loneliness during middle childhood predict risk for adolescent suicidality indirectly through mental health problems

Loneliness and Depression or Depression-Related Factors Among Child Welfare-Involved Adolescent Females

The Role of Family for Youth Friendships: Examining a Social Anxiety Mechanism

Posttraumatic stress disorder in parents and youth after health-related disasters

LET's CONNECT community mentorship program for youths with peer social problems: Preliminary findings from a randomized effectiveness trial

Promoting Mental Health and Preventing Loneliness in Upper Secondary School in Norway: Effects of a Randomized Controlled Trial

Australian Journal of Psychology

Interparental Conflict and Family Cohesion: Predictors of Loneliness, Social Anxiety, and Social Avoidance in Late Adolescence

Psychosocial concomitants of loneliness among students of Cape Verde and Portugal

Life satisfaction and social anxiety among left-behind children in rural China: The mediating role of loneliness

Rapid Systematic Review: The Impact of Social Isolation and Loneliness on the Mental Health of Children and Adolescents in the Context of COVID-19

PhD The authors have reported no funding for this work. All research at Great Ormond Street Hospital NHS Foundation Trust and UCL Great Ormond Street Institute of Child Health is made possible by the NIHR Great Ormond Street Hospital Biomedical Research Centre. This report is independent research. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health and Social Care

on refining the review questions and focus. C. Borwick developed the search strategy

All co-authors reviewed and approved the final draft of the manuscript

The authors very much appreciate the valuable comments made by the anonymous peer reviewers. 

Appendix. Database searches -03/29/2020 Table S1 . Ovid MEDLINE (R) 1 exp Adolescent/ or exp Child/ or exp Child, Preschool/ or exp Infant/ or exp Minors/ or exp Pediatrics/ 3533050 2 (adolesc* or preadolesc* or pre-adolesc* or boy* or girl* or child* or infan* or preschool* or pre-school* or juvenil* or minor* or pe?diatri* or pubescen* or prepubescen* or prepubescen* or puberty or teen* or young* or youth* or school* or high-school* or highschool* or schoolchild* or school child*).tw,kf. (adolesc* or preadolesc* or pre-adolesc* or boy* or girl* or child* or infan* or preschool* or pre-school* or juvenil* or minor* or pe?diatri* or pubescen* or pre-pubescen* or prepubescen* or puberty or teen* or youth* or school* or high-school* or highschool* or schoolchild* or school child*).ti,ab,id. Full references saved as PsycINFO 290320 v1 TS=quarantine # 3 3,591,598 #2 OR #1 # 2 3,581,837 TS=(adolesc* or preadolesc* or pre-adolesc* or boy* or girl* or child* or infan* or preschool* or pre-school* or juvenil* or minor* or pe?diatri* or pubescen* or prepubescen* or prepubescen* or puberty or teen* or youth* or school* or high-school* or highschool* or schoolchild* or school child*) # 1 2,450,709 TS=(adolescent OR child OR child, preschool OR infant OR minor OR pediatrics) Applied 'English language' limit = 3012