Table 1 Summary table for the characteristics and methodological quality of observational studies included in the review

Alves, et al., USA, Cross-sectional [56]

9–15 years; 64 (24/40)

CQ (SR)

Positive affect, Anxiety, Negative affect; STAIC, Positive and Negative Affect Schedule for Children

MVPA was associated with reduced anxiety levels in children with overweight/obesity (r = − 0.52, p < 0.05)

Medium

Alves, et al., USA, Cross-sectional [57]

9–15 years; 65 (25/40)

CQ (SR)

Anxiety; STAIC

Children who engaged in VPA had lower mean anxiety (β = − 2.8, p = 0.01); Children who reported more time spent in MVPA also had lower anxiety (β = − 0.2, p = 0.01)

Medium

Awais, et al., Pakistan, Cross-sectional [58]

17.9 ± 1.22 years; 225 (107/118)

GLTEQ (SR)

Distress; Kessler-10

A moderate negative correlation of PA (r = − 0.340, p < 0.001) was found with psychological distress levels

Medium

Bosselmann, et al., Germany, Cross-sectional [59]

15.83 ± 1.73 years; 122 (53/69)

GSLTPAQ (SR)

Boredom, COVID-19 Fear; FOC-19, Multidimensional State Boredom Scale

COVID-19 Fear significantly correlated with total PA (r = − 0.214, p = 0.017), quantity of strenuous PA (r = − 0.325, p < 0.01), and boredom (r = − 0.211, p < 0.05)

Medium

Breidokiene, et al., Lithuania, Cross-sectional [60]

9.65 ± 1.94 years; 306 (144/162)

CQ (PR)

General well-being; CQ

There was no significant correlation between PA and a Child’s emotional well-being/behavior

Low

Chi, et al., CHN, Cross-sectional [61]

15.26 ± 0.47 years; 1794 (1,077/787)

IPAQ-SF (SR)

Anxiety, Depression, Insomnia; YSIS; PHQ-9; GAD

Moderately active physically was significantly associated with a lower level of depressive symptoms (β = − 0.16, p = 0.002) and anxiety symptoms (β = − 0.16, p = 0.005), while highly active physically was associated with a lower level of insomnia symptoms (β = − 0.05, p = 0.020), depressive symptoms (β = − 0.17, p < 0.001) and anxiety symptoms (β = − 0.15, p = 0.004)

Medium

Constantini, et al., Israel, Cross-sectional [62]

17.4 ± 0.8 years; 473 (302/171)

CQ (SR)

Resilience; CD-RISC

PA was important factor associated with resilience (β = 0.008, p < 0.001)

Medium

Cosma, et al., Czekh, Cross-sectional [63]

13.45 ± 1.62 years; 3438 (1574/1866)

CQ (SR)

General well-being, Life satisfaction; WHO-5, Cantril ladder

Sports and PA components were positively associated with WHO-5 (β = 0.158, p < 0.001) and Life Satisfaction (β = 0.110, p < 0.001)

Medium

Deng, et al., CHN, Cross-sectional [64]

18–22 years; 1607 (1041/566)

CQ (SR)

Anxiety, Depression, Stress; DASS-21

Lower DASS-21 scores were significantly correlated with regular exercise, maintaining exercise habits during the outbreak of COVID-19, exercising more than 1 to 2 times a week, exercise duration > 1 h, and > 2000 pedometer steps (all p < 0.05)

Medium

Ellis, et al., CAN, Cross-sectional [65]

16.68 ± 0.78 years; 1054(231/805) (18 people who chose the “other”)

GLTEQ (SR)

Depression, Loneliness, COVID-19 Stress; CQ, Brief Symptom Inventory, UCLA Loneliness Scale

PA significantly correlated with Loneliness (r = − 0.12, p < 0.01) and depression (r = − 0.08, p < 0.05). More PA (β = − 0.09, p < 0.001) was significant predictor of less loneliness

Medium

Fang, et al., CHN, Cross-sectional [66]

20.17 ± 0.47 years; 1011 (328/683)

The five-item scale was developed by Boothby, Tungatt, and Townsend (SR)

General well-being; The eight-item scale developed by Campbell, The 28-item scale developed by Burckhardt and Anderson

Participation in sports was positively related to general well-being (r = 0.02, p < 0.01). The linear regression analysis of how participation in sports influences well-being (β = 0.25, p < 0.001) revealed a significant amount of explained variance

Medium

Ghorbani, et al., Iran, Cross-sectional [67]

16.28 ± 0.97 years; 136 (60/76)

The accelerometer ActiGraph wGT3X-BT

Anxiety, Depression, Stress; DASS-21

The analyses showed that MVPA per day was negatively associated with all of the mental health indicators; depression (β = − 0.290, p = 0.001), anxiety (β = − 0.404, p < 0.001), and stress (β = − 0.298, p < 0.001)

Medium

Gilbert, et al., USA, Cross-sectional [68]

8.01 ± 1.75 years; 144 (80/64)

CQ (PR)

General well-being; Child Mood and Feelings Questionnaire,

Children with a decrease in total PA, outside play and organized sports programs had lower general well-being (p < 0.05)

Medium

Huang, et al., CHN, Cross-sectional [69]

NR, Middle school students; 1493 (778/715)

CQ (SR)

Negative affect, Feelings of loss of control; CQ

There were significant differences among students with low, medium, and high PA in terms of their self-perceived feelings of loss of control (p < 0.01, effect size = 0.274) and negative affect (p < 0.01, effect size = 0.257)

Medium

Kang, et al., CHN, Cross-sectional [70]

16.3 ± 1.3 years; 4898 (2359/2539)

IPAQ (SR)

Vigor, Self-esteem, Tension, Depression, Anger, Fatigue, Confusion; Chinese Profile of Mood States

Higher levels of PA were significantly associated with lower levels of total mood disturbance in this population (High PA group according to IPAQ: B = − 3.22, SE = 0.40, p < 0.001; Moderate: B =  − 1.47, SE = 0.37, p < 0.001, compared to Low PA group)

Medium

Khan, et al., Bengladesh, Cross-sectional [71]

NR, college and university students; 505 (317/188)

CQ (SR)

Anxiety, Depression, Stress, Event-specific distress; DASS-21, Impact of Event Scale

Physical exercise was significantly associated in lowering scores of DASS depression subscale (B = − 2.10, p < 0.05)

Medium

Lai, et al., CHN, Cross-sectional [72]

NR, University Students; 124 (45/79)

CQ (SR)

Resilience, Anxiety, Depression, insomnia; Stress; PSS-10,

PHQ-4, The seven-item Insomnia Severity Scale, CD-RISC

Exercise was significantly related to anxiety and depression (r =  − 0.194, p < 0.05)

Medium

Lee, Korea, Cross-sectional [73]

14–19 years; 1,046 (521/525)

Model of sports participation developed by Snyder

COVID-19 Stress; COVID-19 Stress Scale

PA was significantly related to COVID-19 stress (r = − 0.162, p < 0.001)

Medium

Lee, et al., Korea, Cross-sectional [74]

14–16 years; 333 (153/180)

Ware’s “Health Perception” Scales (SR)

Mental health importance, Mental health performance; Ware’s “Health Perception” Scales

PA performance was significantly related to mental health importance (r = 0.533, p < 0.001), mental health performance (r = 0.520, p < 0.001)

Medium

Li, et al., CHN, Cross-sectional [75]

5.21 ± 1.40 years; 21,526 (11,281/10,245)

CQ (PR)

Mental health problems; SDQ

PA < 1 h/day (OR: 1.21, p < 0.001) was associated with increased risks for child mental health problems

Medium

Lindoso, et al., Brazil, Cross-sectional [76]

10–18 years; 355 (140, 215)

CQ (SR)

Mental health problems; SDQ

There was a negative correlation between the mental health problems score and PA per week (r = -0.222, p < 0.001)

Medium

Lu, et al., CHN, Cross-sectional [77]

15.26 ± 0.46 years; 965 (556, 409)

IPAQ-SF (SR)

Anxiety, Depression, COVID-19 Fear, Insomnia; YSIS, PHQ-9, GAD, FOC-19

People with high PA were less likely to experience insomnia (OR = 0.71, p < 0.05) and depression (OR = 0.71, p < 0.05) compared to those with low PA

Medium

Lukacs, Hungary, Cross-sectional [78]

24.52 ± 7.15 years; 2162 (599/1552) (11 people who chose the “other”)

CQ (SR)

General well-being, Perceived health; CES-D

The test indicated students with reduced MVPA had lower general well-being and perceived health status scores than students who were unchanged (p < 0.001) or increased their activities (p < 0.001)

Medium

Masse, et al., CAN, Cross-sectional [79]

13 ± 0.1 years; 254 (117/ 137)

CQ (SR)

General well-being, Anxiety; CQ

Children's general well-being had a significant impact on PA (β = 0.286, 95% CI 0.138–0.424, p < 0.05)

Medium

Maximova, et al., CAN, Cross-sectional [80]

9-12 years; 1095 (538/557)

CQ (SR)

General well-being, Positive outlook, Internalizing and functioning problems, Tiredness, Loneliness, Boredom; CQ

Girls who were more physically active during than before the lockdown were less likely to experience ‘internalizing and functioning problems, tiredness and loneliness’ and more likely to have a ‘positive outlook on future and time during lockdown’ relative to those who were less physically active. Boys who were physically active during the lockdown were more likely to have a ‘positive outlook on future and time during the lockdown’

Medium

McArthur, et al., CAN, Cross-sectional [81]

9–11 years; 846 (447/398)

CQ (SR)

Happiness, Anxiety, Depression; CQ

PA was significantly related to anxiety (r = − 0.11, p < 0.05) and happiness (r = 0.15, p < 0.05)

Medium

McGuine, et al., USA, Cross-sectional [82]

15.7 ± 1.2 years; 559 (313/244)

The Hospital for Special Surgery Pediatric Functional Activity Brief Scale (SR)

Health-related quality of life, Anxiety, Depression; GAD-7, PHQ-9, Pediatric Quality of Life Inventory 4.0

The did not play (DNP) group had a higher (ie., worse) GAD-7 score than the did play (PLY) group (p < 0.001) as well as a higher (ie., worse) PHQ-9 score than the PLY group (p < 0.001). the PLY group had a higher Pediatric Quality of Life Inventory total scores (p < 0.001)

Medium

Mitra, et al., CAN, Cross-sectional [83]

9–15 years; 800 (377/423)

CQ (SR)

General well-being ; CQ

Lower general well-being was related to less PA (OR = 1.54, p = 0.028)

Medium

Moriarty, et al., USA, Cross-sectional [84]

21.3 ± 3.8 years; 550 (135/408)

IPAQ-SF (SR)

Stress; PSS-4

Stress was negatively associated with exercise during COVID-19 (r = − 0.162, p < 0.001)

Medium

Morres, et al., Greece, Cross-sectional [85]

14.41 ± 1.63 years; 950 (518/432)

IPAQ-SF (SR)

Positive energy, Relaxation, General well-being, Negative arousal, Tiredness; The 4-Dimensional Mood Scale, WHO-5

Total PA was significantly related to general well-being (r = 0.35, p < 0.01), positive energy (r = 0.36, p < 0.01), relaxation (r = 0.12, p < 0.01), and negative arousal (r = − 0.10, p < 0.01)

Medium

Oliva, et al., Italy, Cross-sectional [86]

1-18 years; 9688 (5066/4622)

CQ (PR)

Anxiety, Depression, Emotional and behavioral problems; The Pediatric Symptom Checklist, CES-D, The Screen for Child Anxiety Related Disorders

Emotional and behavioral problems (Estimate = − 5.7980, p < 0.001), anxiety (Estimate = − 4.3827, p < 0.001), and depression (Estimate = − 3.0091, p < 0.001) were negatively correlated with PA

Medium

Pigaiani, et al., Italy, Cross-sectional [87]

18.1 ± 0.9 years; 306 (223/83)

CQ (SR)

General well-being; CQ

PA was significantly related to general well-being (OR = 2.609, p = 0.007)

Medium

Qi, et al., CHN, Cross-sectional [88]

11–20 years; 9554 (4557/4997)

CQ (SR)

Depression; CES-D

Duration of PA > 60 min/day (OR = 0.686, p < 0.001) and 30–60 min/day (OR = 0.636, p < 0.001) were significantly associated with lower risk of depression

Medium

Qin, et al., CHN, Cross-sectional [89]

13–16 years; 248 (114/134)

CQ (SR)

Anxiety; Mental Health Test

Exercising for ≥ 1 h per day (OR = 0.23, p < 0.01) was a protective factor for anxiety

Medium

Qin, et al., CHN, Cross-sectional [90]

12.04 ± 3.01 years; 1,199,320 (619,144/580,176)

CQ (SR)

Distress; GHQ-12

Students who spent less than 0.5 h exercising had increased odds of self-reported psychological distress compared with students who spent more than 1 h exercising (OR = 1.64, p < 0.001)

High

Ren, et al., CHN, Cross-sectional [91]

13.14 ± 1.55 years; 1487 (727/760)

CQ (SR)

Depression; CES-D

PA time (β = − 0.07, p < 0.001) was negatively associated with depressive symptoms

Medium

Sikorska, et al., Poland, Cross-sectional [92]

15.38 ± 2.10 years; 455 (121/243) (2 non-binary people and 4 people who chose the “other”)

CQ (SR)

Resilience, General well-being, Anxiety, Depression, Stress; CD-RISC, DASS

Physical exercise was significantly associated resilience (r = 0.183, p < 0.01), emotional well-being (r = 0.153, p < 0.01), psychological well-being (r = 0.191, p < 0.01), Social well-being (r = 0.126, p < 0.05)

Medium

Song, et al., Korea, Cross-sectional [93]

NR, middle school and high school students; 836 (412/424)

Social Aspect of Sport (SR)

Optimism, Pessimism, COVID-19 Stress; A scale with Verified Reliability and Validity as Reported by Chang, The scales Used by Gaumer Erickson, The COVID Stress Scale

Sports participation exerted a positive effect on optimism (β = 0.659, p < 0.001). Sports participation exerted a negative effect on pessimism (β = − 0.156, p = 0.037) and COVID Stress (β = − 0.656, p < 0.001)

Medium

Swansburg, et al., CAN, Cross-sectional [94]

10.14 ± 3.06 years; 587 (166/412) (9 people who chose the “other”)

A daily activities table with the same activity categories as the United Kingdom Co-SPACE study (SR)

Anxiety, Depression, Hyperactive impulsive, Inattention; PHQ-9, GAD, The Swanson, Nolan, and Pelham 26-question scale

Exercising < 1 h/day correlated positively with the PHQ-9 (r = 0.110, p < 0.01) but negatively with the hyperactive/impulsive (r = − 0.086, p < 0.05) score

Medium

Szwarcwald, et al., Brazil, Cross-sectional [95]

12-17 years; 9470 (4716/4754)

The National School Health Survey (SR)

Insomnia, Irritability, Sadness; World Health Survey

PA for 60 min or over at least twice a week was inversely correlated with the problem (At least two problems from frequent sadness, frequent irritability, and sleep problems) (OR = 0.82, p < 0.001)

Medium

Tandon, et al., USA, Cross-sectional [96]

10.8 ± 3.5 years; 1000 (517/467)

CQ (SR)

Mental health problems; SDQ

For younger children (6–10 years), engaging in the recommended 7 day/week of PA was associated with mental health problems (β = − 2.4, p = 0.04). For older children (11–17 years), engaging in 1–6 (β = − 3.5, p < 0.01), and 7 day/week (β = − 3.6, p < 0.01) of PA was significantly associated with mental health problems

Medium

Thomas, et al., United Kingdom, Longitudinal [97]

16–24 years; 64 (34/30)

CQ (SR)

Self-esteem, General well-being; The Short Warwick-Edinburgh Mental Wellbeing Scale

Changes in general well-being were positively associated with changes in moderate PA (r = 0.24, p < 0.05), and total PA (r = 0.28, p < 0.05)

Medium

Wang, et al., CHN, Cross-sectional [98]

NA, primary scholar; 6017 (3287/2730)

CQ (PR)

Emotional and behavioral problems; SDQ

Exercise duration of 30 min/day could reduce the risk of Emotional and behavioral problems

Medium

Wang, et al., CHN, Cross-sectional [99]

6–16 years; 12,186 (6357/5829)

CQ (SR)

Psychosocial and behavioral problems; The Child Behavior Checklist Score

PA time per day was a significant risk factor associated with mood and behavior problems: outside Wuhan (OR = 0.510, p < 0.001), within Wuhan (OR = 0.416, p < 0.01)

Medium

Wang, et al., USA, Longitudinal [100]

13–18 years; 349 (140/209)

GLTEQ (SR)

Positive Affect, Stress; The Multicultural Events Schedule for Adolescents, The Positive and Negative Affect Scale for Children

Daily exercise was significantly associated Stress (r = − 0.05, p < 0.001) and positive affect (r = 0.23, p < 0.001)

Low

Wright, et al., United Kingdom, Cross-sectional [101]

15.9 ± 1.48 years; 165 (65/100)

CQ (SR)

Vigor, Anxiety, Depression, Fatigue, Stress; PSS-10, The Hospital Anxiety and Depression Scale, The 20-item Multidimensional Fatigue Inventory, The Subjective Vitality Scale

Higher levels of PA were associated with lower levels of stress (r = -0.26, p < 0.01), depression (r = − 0.31, p < 0.01), fatigue (r = − 0.38, p < 0.01), as well as higher levels of vigor (r = 0.34, p < 0.01)

Medium

Wunsch, et al., Germany, Cross-sectional [102]

10.36 ± 4.04 years; 1711 (750/961)

The Momo Physical Activity Questionnaire (SR)

Health-related quality of life, The KIDSCREEN-10 Index

PA within-COVID-19 was positively predicted by pre-COVID-19 health-related quality of life (standardized estimate = 0.07; p = 0.003)

Medium

Xu, et al., CHN, Cross-sectional [103]

NA, college and university students; 11,254 (4054/7200)

CQ (SR)

Anxiety, Depression, Insomnia, Posttraumatic stress disorder (PTSD); PHQ-9, GAD-7, ISI, Posttraumatic Stress Disorder Checklist for DSM-5

Participants who engaged in regular exercise during the pandemic reported a lower risk of symptoms of depression (OR = 0.86, p = 0.01), and insomnia (OR = 0.82, p < 0.01)

Medium

Zhang, et al., CHN, Cross-sectional [104]

NA, college and university students; 2270 (877/1393)

CQ (SR)

Anxiety, Depression; The Self-Rating Anxiety Scale, The Self-Rating Depression Scale

Exercise during the epidemic outbreak (OR = 0.456, p < 0.001) was protective factors for anxiety and depression

Medium

Zhang, et al., CHN, Cross-sectional [105]

11.63 ± 1.23 years; 9979 (5131/4848)

IPAQ-SF (SR)

Self-esteem, Vigor, Anger, Confusion, Depression, Fatigue, Tension; Chinese Profile of Mood States

Moderate PA (β = − 3.031, p < 0.001) and High PA (β = − 1.309, p = 0.047) were significantly correlated with mood states in children and adolescents

Medium

Zhang, et al., CHN, Cross-sectional [106]

20.51 ± 1.88 years; 11,787 (5056/6731)

CQ (SR)

Depression; PHQ-9

Compared with PA ≥ 3 day/week, PA < 3 day/week was positively associated with depression symptoms (β = 0.01, 95% CI 0.008–0.012)

Medium

Zhang, et al., CHN, Longitudinal [107]

20.70 ± 2.11 years; 66 (25/41)

IPAQ (SR)

Aggressiveness, Anxiety, Depression, Stress; DASS-21, The Buss-Perry Aggressive Questionnaire

Each 100-unit increase in METs of total PA corresponded to a change of (Point Estimate = − 0.12, p < 0.05) in the global DASS score. PA also significantly alleviated depression (Point Estimate = − 0.04, p < 0.05)

Medium

Zhou, et al., CHN, Cross-sectional [108]

11–18 years; 4805 (0/4805)

CQ (SR)

Depression; CES-D

Compared with physical exercise duration/day of > 30 min, duration/day of < 30 min (OR = 1.641, p < 0.001) was a risk factor for depression

Medium

Zhu, et al., CHN, Cross-sectional [109]

12.6 ± 1.3 years; 2860 (1346/1502) (12 people missing)

The Mental Health Lifestyle Scale (SR)

Being more stressed, Feeling more horrified, Feeling more apprehensive, Feeling more helpless, Having larger study pressure, Perceived vulnerability; CQ, The adapted Perceived Risk of the HIV Scale

Those who perceived more vulnerability were more likely than others to spend more time to exercise (OR = 1.28, p < 0.001)

Medium