Table 1.

Summary of included studies

Author, year, country	Sample, age, informants and response rates	Before and during pandemic time points	Outcome(s) and measure(s)	Key findings	Before/during data (mean ± SD otherwise stated), Changes analysis (eg. statistical significance, effect sizes)	Direction of mental health changes	Mental health changes based on age, gender or existing poor mental health
Repeated cross-sectional
Chen et al. (2021), China [33]	Snowball sampling using smart phone-based WeChat-Wenjuanxing. Self-reports from students aged 11–20 in all 34 Chinese provinces before pandemic: 9554. During: 3886 (no response rates stated)	February 2020 vs April 2020	(1) Depression (Center for Epidemiological Studies Depression scale) (2) Anxiety (Generalized Anxiety Disorder 7 scale)	Increased depression and anxiety symptoms during the pandemic	(1) Depression (2) Anxiety; Before: (1) 14.06 ± 10.80, (2) 2.28 ± 3.54; During: (1) 19.29 ± 11.82, (2) 3.98 ± 4.46; Changes: (1) p < 0.001, (2) p < 0.001	(1) Depression deteriorated (2) Anxiety deteriorated	Gender: increased risk of depression and anxiety among female; Age: increased risk of depression and anxiety among senior secondary school (age 16–20)
Gray et al. (2020), Wales (UK) [40]	Online snowball sampling, advertised via social and mass media. 2019 data from on those aged 16–24 from large-scale, self-reported National Survey for Wales. Before pandemic: not stated. During: 703 (no subpopulation response rates stated)	April 2018 to March 2019 vs June and July 2020	Well-being (Warwick-Edinburgh Mental Well-being scale)	Large decrease in psychological well-being during pandemic	Mental well-being; Before: 50.3; During: 41.2; Changes: p < 0.001, Hedge’s g 0.95, 95% CI 0.83–1.06	Mental well-being deteriorated	Age: greater increase in psychological distress among young people (18–24) compared to older general population
Liébana-Presa et al. (2020), Spain [38]	Two self-reported cross-sectional samples obtained by convenience, non-probability sampling of students aged 13–17 from two schools with similar socio-demographic characteristics in a Spanish town. Before pandemic: 202. During: 98 (response rates during 36%)	February 2020 vs May 2020	(1) Emotional stress (2) Physiological stress (3) Behavioral stress (all subscales of Student Stress Inventory-Stress Manifestations)	Statistically significant and higher manifestation of emotional stress during pandemic	Stress: (1) Emotional (2) Physiological (3) Behavior; Before: (1) 21.02 ± 8.88, (2) 24.63 ± 9.83, (3) 9.38 ± 4.16; During: (1) 24.63 ± 9.83, (2) 10.15 ± 4.54, (3) 9.95 ± 4.21; Changes: (1) p < 0.05, (2) p = 0.340, (3) p = 0.266	(1) Emotional stress deteriorated (2) No significant change (physiological stress) deteriorated (3) No significant change (behavioral stress)	N/A
Ravens‐Sieberer et al. (2021), Germany [7]	Pandemic population-based representative online survey Corona and Psyche (COPSY) compared with pre-pandemic data from nationwide, longitudinal, representative BELLA cohort study Before: 1556. During: 1586 families with adolescents aged 7–17 years. Self-reported and/or parent-reported COPSY response rate 45.8%. No response rate for BELLA	BELLA 2017 vs COPSY May and June 2020	(1) Quality of life (KIDSCREEN-10 index) (2) Depression (German version of the Center for Epidemiological Studies Depression scale) (3) Anxiety (Screen for Child Anxiety Related Disorders (4) Mental health (5) Emotional symptoms (6) Conduct problems (7) Hyperactivity (8) Peer problems (4–8 all Strengths and Difficulties Questionnaire)	Adolescents experienced significantly lower quality of life, more mental health problems (anxiety), conduct problems, hyperactivity, and peer problems during the pandemic	(1)Quality of life (2) Depression (3) Anxiety (4) Mental health (5) Emotional (6) Conduct (7) Hyperactivity (8) Peer problems; Before (%): low, normal/high: (1) 15.3, 84.7 high: (2) not available, (3) 14.9% normal, borderline, abnormal: (4) 82.4, 7.8, 9.9, (5) 83.6, 6.2, 10.2, (6) 86.9, 6.5, 6.6, (7) 87.2, 5.1, 7.7, (8) 88.6, 3.9, 7.5; During (%): low, normal/high (1) 40.2, 59.8 high: (2) Not available, (3) 24.1% normal, borderline, abnormal: (4) 69.6, 12.5, 17.8, (5) 79.0, 7.7, 13.3, (6) 80.8, 9.1, 10.0, (7) 76.4, 8.9, 14.6, (8) 78.2, 10.2, 11.5; Changes: (1) p < 0.001, (2) p > 0.05, (3) p < 0.001, (4) p < 0.001, (5) p = 0.007, (6) p < 0.001, (7) p < 0.001, (8) p < 0.001	(1) Quality of life deteriorated (2) No significant change in depression, (3) mental health (anxiety) deteriorated (4) mental health (total mental health) deteriorated (5) no significant change in emotional symptoms (6) mental health (conduct problems) deteriorated (7) Mental health (hyperactivity) deteriorated (8) mental health (peer problems) deteriorated	Age: higher increases in mental health problems among boys (parental report 7–17)
Longitudinal
Alivernini et al. (2020), Italy [36]	Adolescents aged 14–19 from several different geographical areas of Italy who participated in a self-reported online education project 347 adolescents. Mean age 16.4 ± 1.12 years (no response rates stated)	April 2019 vs May and June 2020	(1) Positive affect (2) Negative affect (both Positive and Negative Affect Schedule—Children)	Increase in negative affect, decrease in positive affect during the pandemic	(1) Positive affect (2) Negative affect; Before: (1) 3.74 ± 0.61, (2) 2.35 ± 0.70; During: (1) 3.59 ± 0.75, (2) 2.66 ± 0.68; Changes: (1) p < 0.001, (2) p < 0.001	(1) Positive affect deteriorated (2) Negative affect deteriorated	N/A
Banks and Xu (2020), UK [27]	Self-reported panel data from the nationwide representative UK understanding society household longitudinal study sub-population of 1851 subjects aged 16–24 years (No subpopulation response rates stated)	2009 to 2019 vs April 2020	Psychological distress (General Health Questionnaire-12)	Substantial effects on young adults who already had lower levels of mental health before COVID-19	Psychological distress; Before: 11.83 ± 5.92; During: not available; Changes: original article authors interpretation of significant deterioration among young adults	Psychological distress deteriorated	Gender: significantly worsened psychological distress among young women (16–24) compared to young men and older general population; Existing poor mental health: young women aged 16–24 recorded increasing severe problem
Bignardi et al. (2020), UK [35]	Sample from the Resilience in Education and Development study, from a cohort of children aged 7.6–11.6 years living in the East of England 168 children, mixture of caregiver, teacher and child reports (29% response rate)	June 2018 to September 2019 vs April to June 2020	(1) Depression (2) Anxiety (both Revised Child Anxiety and Depression scale) (3) Emotional problems (Strengths and Difficulties Questionnaire)	Significant increase in depression during pandemic. No significant changes in anxiety and emotional problems subscale	(1)Depression (2) Anxiety (3) Emotional; Before and During: not available; Changes: (1) p < 0.001, (2) p = 0.035, (3) p = 0.173	(1) Depression deteriorated (2) No significant change (anxiety) (3) No significant change (emotional problems)	Age: did not significantly alter the changes; Gender: did not significantly alter the changes
Giménez-Dasí et al. (2020), Italy [26]	Families of preschool and primary school children aged 3.2–11.1 years in two educational centers in northwestern Madrid 113 families, parents’ reports (67% response rate). Mean age 7 years and 2 months ± 2.64 (no response rates stated)	February 2020 vs April 2020	(1) Attentional Problems, (2) Depression, (3) Challenging Behaviors, (4) Emotional Regulation, (5) Hyperactivity, (6) Willingness to study* (System of Evaluation of Children and Adolescents) *(only for the primary education version)	Unchanged for 3-year-old children during pandemic. Lower scores for children aged 6–10 years for attention, emotional regulation, hyperactivity and willingness to study	(1) Attentional problems, (2) Depression, (3) Challenging Behaviors, (4) Emotional regulation, (5) Hyperactivity, (6) Willingness to study; Before: Preschool (1) 2.11 ± 0.68, (2) 1.19 ± 0.24, (3) 2.64 ± 0.69, (4) 2.26 ± 0.67, (5) 2.41 ± 0.77 Primary school (1) 2.21 ± 0.81, (2) 1.40 ± 0.45, (3) 2.30 ± 0.77, (4) 2.15 ± 0.74, (5) 0.71 ± 0.75, (6) 3.02 ± 0.65; During: preschool (1) 2.31 ± 0.76, (2) 1.28 ± 0.45 (3) 2.67 ± 0.83, (4) 2.43 ± 0.92, (5) 2.44 ± 0.87 primary school (1) 2.42 ± 0.83, (2) 1.60 ± 0.64, (3) 2.66 ± 0.89, (4) 2.42 ± 0.93, (5) 0.75 ± 0.75, (6) 2.38 ± 0.74; Changes: preschool: no significant changes for all primary school: (1) p = 0.02, (4) p = 0.01, (5) p < 0.001, (6) p < 0.001	Preschool (no significant change for all 5) (1) Attentional Problems, (2) Depression, (3) Challenging Behaviors, (4) Emotional Regulation, (5) Hyperactivity primary school (significant change for 1, 4, 5 and 6, and no significant change for 2 and 3) (1) Attentional problems, (2) Depression, (3) Challenging behaviors, (4) Emotional regulation, (5) Hyperactivity, (6) Willingness to study	Age: no significant difference; Gender: no significant difference
Huckins et al. (2020), USA [25]	Students aged 18–22 from Dartmouth college who participated in the self-report StudentLife study 178 students (no response rates stated)	Academic terms September 2017 until September 2018 vs Winter 2020 academic term	(1) Depression (Patient Health Questionnaire-2) (2) Anxiety (Generalized Anxiety Disorder-2)	Significantly increased symptoms of anxiety and depression	(1) Depression, (2) Anxiety; Before and During: weekly EMA; Changes: (1) p < 0.001, (2) p < 0.001	(1) Depression deteriorated (2) Anxiety deteriorated	N/A
Li H.Y. et al. (2020), China [31]	Undergraduate students attending Hebei Agricultural University in Baoding, China, who had completed the first wave of the self-reported survey in December 2019 were contacted again 555 students took part at a mean age of 19.6 ± 3.4 years (response rate 89%)	December 2019 vs February 2020	(1) Positive affect (2)Negative affect (both Positive and Negative Affect Schedule) (3) Anxiety and depression (Patient Health Questionnaire-4)	Increases in negative affect and symptoms of anxiety and depression	(1) Positive affect, (2)Negative affect, (3) Anxiety and depression; Before: (1) 3.21 ± 0.79, (2) 2.38 ± 0.79, (3) 0.95 ± 0.65; During: (1) 3.26 ± 0.79, (2) 2.24 ± 0.80, (3) 0.76 ± 0.61; Changes: (1) p = 0.107, (2) p < 0.001, (3) p < 0.001	(1) No significant change (positive affect) (2) Negative affect deteriorated (3) Anxiety and depression Deteriorated	N/A
Li W. et al. (2020), China [30]	Second and third-year students across five disciplines from a university in a southeast city of China took part in this self-reported study the 173 students had a mean age of 19.81 ± 0.98 years (33% response rate)	October 2019 vs January 2020 vs March and April 2020	(1) Depression (2) Anxiety (3) Stress (all Chinese Depression Anxiety Stress Scale)	Stress, anxiety and depression decreased during lockdown, before increasing as restrictions eased	(1) Depression (2) Anxiety (3) Stress; Before: (1) 6.25 ± 6.15, (2) 9.23 ± 6.16, (3) 10.95 ± 7.05; During time 1: (1) 4.99 ± 6.15, (2) 5.09 ± 5.90, (3) 7.11 ± 7.18; During time 2: (1) 6.01 ± 6.10, (2) 6.74 ± 5.97, (3) 8.77 ± 7.04; Changes: (1) p < 0.001, p < 0.001, (2) p < 0.001, p < 0.001, (3) p < 0.001, p < 0.001	(1) Depression deteriorated (2) Anxiety deteriorated (3) Stress deteriorated	N/A
Magson et al. (2020), Australia [39]	Sample comprised adolescent aged 13–16 who lived in an urban area of New South Wales, Australia, who were part of the larger self-reported longitudinal Risks to Adolescent Well-being Project 248 responded at a mean age of 14.4 ± 0.5 years (53% response rate)	Throughout 2019 vs May 2020	(1) Depression (the Short Mood and Feelings Questionnaire—Child version) (2) Anxiety (Spence Children’s Anxiety Scale)	Significant increase in symptoms of depression and anxiety during. pandemic	(1) Depression (2) Anxiety; Before: (1) 3.81 ± 4.31, (2) 4.60 ± 3.74; During: (1) 6.12 ± 6.04, (2) 5.10 ± 4.05; Changes: (1) p < 0.001, (2) p < 0.001	(1) Depression deteriorated (2) Anxiety deteriorated	Age: no significant difference; Gender: more significant changes among girls
Manjareeka and Pathak (2021), India [34]	First year medical school students aged 18–24 at KIIT University, India 101 took part in a self-reported study at a mean age of 19.7 ± 0.7 years (67% response rate)	February 2020 vs May 2020	Anxiety (state version of State Trait Anxiety Inventory)	Mean anxiety scores were significantly lower before than during the COVID-19 lockdown period	Anxiety; Before: 45.70 ± 11.42; During: 47.97 ± 10.80; Changes : p = 0.0394	Anxiety deteriorated	Gender: no significant difference
Munasinghe et al. (2020), Australia [29]	Instagram and Facebook used to recruit adolescents aged 13–19 from general population in Sydney 301 had before and during data (no response rates stated)	18 November 2019 to 22 March 2020 (before) vs after 23 March-19 April 2020	Psychological distress (Kessler Psychological Distress 6-item scale)	Social distancing measures during the pandemic were associated with slightly higher increases in psychological distress	Psychological distress; Before (median interquartile range): 15 (11–20); During (median interquartile range): 15.5 (11–20); Changes: OR = 1.48, 95% CI 95% 0.74–2.95	Psychological distress deteriorated	N/A
Niedzwiedz et al. (2021), UK [24]	Self-reported panel data from the nationwide representative UK understanding society household longitudinal study 5875 adults aged 18–24 took part (no subpopulation response rates stated)	2015–2019 vs COVID-19 wave 24th to 30th April 2020	(1) Psychological distress (General Health Questionnaire-12) (2) Loneliness (1) loneliness item: how often felt lonely in last 4 weeks)	Psychological distress and loneliness increased during the pandemic	(1) Psychological distress, (2) Loneliness; Before (%): (1) N/A (2) 13.3 (95% CI 11.6–15.3); During (%): (1) N/A (2) 20.2 (95% CI 16.0–25.2); Changes: (1)Psychological distress (23% to 40% based on original authors figure), (2) loneliness—both showed significant increase (non-overlapping confidence intervals)	(1) Psychological distress deteriorated (2) Loneliness deteriorated	Age: significantly worsened psychological distress among young people (18–24) compared to older general population
Pierce et al. (2020), UK [28]	Self-reported panel data from the nationwide representative UK Understanding Society Household Longitudinal Study adolescents aged 16–24 took part with 1,543 in COVID-19 2020 wave (no subpopulation response rates stated)	2014 to 2019 vs 23–30 April 2020	Psychological distress (General Health Questionnaire-12)	Significant increase in GHQ-12 scores of individuals aged 16–24 compared to pre-pandemic surveys	Psychological distress; Before (mean and 95% CI): 2014–15: 10.9 (10.6–11.1) 2015–16: 10.8 (10.6–11.0) 2016–17: 11.1 (10.8–11.3) 2017–18: 11.6 (11.3–11.8) 2018–19: 12.0 (11.6–12.5); During (mean and 95% CI): 14.7 (14.1–15.3); Changes: GHQ score pandemic 14.7 95% (14.1–15.3). The increases were significant (as shown by their non-overlapping confidence intervals)	Psychological distress deteriorated	Age: significantly worsened psychological distress among young people (18–24) compared to older general population; Existing poor mental health: steepest increase in the proportion of clinically significant psychological distress among female aged 16–24 with already the highest proportions before the pandemic
Rogers et al.(2020), USA [8]	Project AHEAD, a two-wave self-reported longitudinal study, focused on a stratified random sample of adolescents aged 14–17 drawn from a nationally representative database 407 adolescents. Mean age 15.42 ± 1.16 (no response rates stated)	October 2019 vs April 2020	(1) Depression (Children’s Depression Inventory-short version) (2) Anxiety (Generalized Anxiety Disorder scale) (3) Loneliness (Three-item Loneliness Scale)	Small but significant increases in depressive symptoms, anxiety symptoms and loneliness during the pandemic	(1) Depression, (2) Anxiety, (3) Loneliness; Before: (1) 1.75 ± 0.52, (2) 1.64 ± 0.77, (3) 1.3 ± 0.47; During: (1) 1.84 ± 0.56, (2) 1.85 ± 0.79, (3) 1.44 ± 0.53; Changes: (1) t(406) = 3.88, p < .001; Cohen’s d = 0.19, (2) t(406) = 5.92, p < .001; Cohen’s d = 0.28, (3) t(406) = 5.52, p < .001; Cohen’s d = 0.27	(1) Depression deteriorated (2) Anxiety deteriorated (3) Loneliness deteriorated	Existing mental health: depressive symptoms before strong indicator of depressive symptoms during the pandemic
Thorisdottir et al. (2021), Iceland [41]	Self-report Youth in Iceland school surveys carried out by the Icelandic Centre for Social Research and Analysis targeted all secondary schools students aged 13–18 59,701 adolescents took part (63%-86% response rates)	October/February 2016 and 2018 vs October 2020	(1) Depression (Symptom Checklist-90) (2) Mental well-being (the short Warwick Edinburgh Mental Well-being scale)	Increase in depressive symptoms and worsened mental well-being were observed across all age groups during the pandemic, compared with peers of the same age before COVID-19	(1) Depression, (2) Mental well-being; Before (2016): 13 years (1) 16.52 ± 6.89, (2) 25.21 ± 6.02 14 years (1) 17.36 ± 7.43, (2) 25.04 ± 5.79 15 years (1) 18.11 ± 7.83, (2) 24·73 ± 5.93 16 years (1) 18.23 ± 7·50, (2) 25.62 ± 5.75 17 years (1) 18.94 ± 7.66, (2) 25.68 ± 5.81 18 years (1) 19.09 ± 7.62, (2) 25.96 ± 5.80; During (2020): 13 years (1) 18.72 ± 7.21, (2) 23.13 ± 5.44 14 years (1) 19.40 ± 7.77, (2) 22.86 ± 5.64 15 years (1) 19.41 ± 7.69, (2) 23.01 ± 5.62 16 years (1) 20.65 ± 8.10, (2) 23.92 ± 5.26 17 years (1) 22.52 ± 8.29, (2) 23.37 ± 5.19 18 years (1) 22.41 ± 8.29, (2) 23.66 ± 5.19; Changes: (1) (β 0·57, 95% CI 0·53 to 0·60) (2) (β –0·46, 95% CI − 0·49 to −0·42)	(1) Depression deteriorated (2) Mental well-being deteriorated	Gender: increase in depressive symptoms and worsened mental well-being significantly worse in adolescent girls; Age: increase in depressive symptoms and worsened mental well-being significantly worse among 16–18-year-olds
Wang et al.(2020), China [32]	Non-graduating college students aged 18–22 in a top multidisciplinary and research-oriented university in China took part in a self-report study. From 34 provincial-level administrative units and 260 cities in China 1,172 students (30.84% response rate before)	February 2020 vs March 2020	Anxiety (the Self-Rating Anxiety Scale)	Significantly higher anxiety during the pandemic. Significant differences among all males, females, and students majoring in arts and sciences between the two studies	Anxiety; Before: 40.39 ± 9.98; During: 40.77 ± 10.51; Changes: p < 0.05	Anxiety deteriorated	Gender: significant increase in anxiety symptoms among female freshmen but not for male freshmen
Wendel et al. (2020), Canada [37]	Parents from 32 kindergarten classrooms in six schools from a large city in Eastern Canada 113 parents provided date on children aged 4–6 years with a mean age of 4.66 ± 0.54 (no response rates stated)	December 2019 to January 2020 vs May to June 2020	(1) Inattention (2) Hyperactivity/Impulsivity (both The ADHD Rating Scale–5 for children and adolescents, home version)	Children’s inattention and hyperactivity/Impulsivity symptoms increased during the pandemic	(1) Inattention, (2) Hyperactivity/Impulsivity; Before and During: not available changes: (1) p = 0.00, (2) p = 0.02	(1) Inattention deteriorated (2) Hyperactivity/Impulsivity deteriorated	N/A
Xiang et al. (2020), China [23]	Students aged 6–17 from five schools, selected using cluster sampling from 14 districts in Shanghai, took part in both surveys 2427 school students and/or parents reported (79.8% response rate)	3–21 January 2020 vs 13–23 March 2020	Depression (Children’s Depression Inventory-Short form)	The mean CDI-S score significantly decreased between the period before school closure and when schools were closed during the pandemic	Depression; Before: 4.19 ± 2.82; During:3.90 ± 2.56; Changes: p < 0.001	Depression improved	N/A

N/A not available