Child and Adolescent Mental Health Volume **, No. *, 2021, pp. **–** doi:10.1111/camh.

12485

Review: Physical activity interventions for the

mental health and well-being of adolescents – a
systematic review
Gabrielle E. Hale , Luke Colquhoun, Deborah Lancastle, Nicky Lewis &
Philip J. Tyson
School of Psychology and Therapeutic Studies, University of South Wales, Pontypridd, UK

Background: Rates of physical activity decline throughout adolescence, and evidence indicates that this has an
adverse impact on psychological health. This paper aims to synthesise available evidence for physical activity
interventions on the mental health and well-being of young people (11–19 years) from the general popula-
tion. Method: Nine databases were searched to identify studies published between January 2005 and June
2020: Web of Science, ProQuest Psychology Journals, PsycINFO, Pub Med, ASSIA, CINHAL PLUS, SPORTDiscus,
EMBASE and Wiley Online Library. Key search terms included ‘physical activity intervention’, ‘mental health’
and ‘adolescen*’. Eligible studies were independently screened by two authors based on inclusion/exclusion
criteria. Results: Twenty-eight interventions were narratively synthesised in four categories: Quality of Life
(QOL), self-esteem, psychological well-being and psychological ill-being (e.g. depression, stress). A large pro-
portion (67.9%) of interventions were multicomponent and combined physical activity with other features
such as health education (e.g. nutrition). However, only a limited number (N = 5) specifically addressed mental
health. Findings suggest that interventions are useful in improving psychological well-being and QOL, yet evi-
dence for self-esteem is mixed. Conclusions: Although effectiveness in improving well-being is evident, evi-
dence for a reduction in the frequency and severity of mental health problems is less clear. A summary of the
overall impact of physical activity interventions on the mental health of young people is presented.

Key Practitioner Message

• The impact of physical activity on youth mental health is evident, and further support has been provided
for the effect of physical activity interventions on QOL and psychological well-being.
• Findings illustrate a lack of interventions combining physical activity and mental health promotion in a mul-
ticomponent framework and identify a pathway for future development and research.

Keywords: Physical activity intervention; mental health; well-being; adolescent; young people

mental health due to the absence of the positive influ-

Introduction
ence of the biopsychosocial factors outlined above
Adolescents who engage in physical activity display (Lubans, Smith, et al., 2016). Moreover, lack of MVPA is
increased self-esteem and lower levels of anxiety and linked to various psychological and physical health
depression (Ahn & Fedewa, 2011; Beauchamp, risks, including depressive symptomatology, psychologi-
Puterman, & Lubans, 2018) due to various biopsychoso- cal distress, obesity and cardio-metabolic disease (Car-
cial influences, including changes to brain function, son et al., 2016; Hoare, Milton, Foster, & Allender,
opportunities for social interaction and development of 2016). To address concerns, there is a need to establish
competence (Lubans, Richards, et al., 2016). Despite effective interventions that target both mental and phys-
the marked benefits, many do not participate in the ical health. Immersing mental health interventions in a
recommended average 60 min of moderate to vigorous physical health context can increase physical activity,
intensity, mostly aerobic, physical activity per week, enhance mental health, and prevent the development of
with rates disproportionately lower for females com- psychological and physical problems in later life.
pared to males (Chaput et al., 2020; Farooq et al., , 2018; Existing physical activity interventions can be deliv-
Hallal et al., 2012; World Health Organisation [WHO], no ered alone or as part of a multicomponent framework,
date). Research exploring participation barriers has whereby physical activity is combined with additional
found that body image, self-esteem, experiences of phys- components such as health (e.g. portion control), beha-
ical education (PE) at school and peer reactions can con- vioural (e.g. goal-setting) or psychological education (e.g.
tribute to lack of participation (Martins, Marques, body image) (Spruit, Assink, van Vugt, van der Put, &
Sarmento, & Carreiro da Costa, 2015). Adolescents may Stams, 2016). Of reviews that have explored the impact
avoid physical activity due to low body image and self- of physical activity interventions on the mental health of
esteem, yet this lack of activity could exacerbate poor the general adolescent population (Biddle & Asare,

© 2021 Association for Child and Adolescent Mental Health

2 Gabrielle E. Hale et al. Child Adolesc Ment Health 2021; *(*): **–**

2011; Biddle, Ciaccioni, Thomas, & Vergeer, 2019), most an update on the literature on physical activity interven-
findings have been positive. Spruit et al. (2016) identified tions that have evaluated impact on adolescent mental
small-to-medium effects on internalising problems such health, (b) identify any gaps in knowledge and research
as anxiety (d = 0.316) and self-esteem (d = 0.297), in this area, and (c) summarise any key components
whereas Brown, Pearson, Braithwaite, Brown, and Bid- used in intervention design (i.e. activity type, education
dle (2013) found small but significant effects (Hedges’ components) and discuss their effectiveness in improv-
g = 0.26) on depression. Subgroup analyses also ing adolescent mental health.
revealed that interventions that included health educa-
tion and were less than three months in duration con-
Methods
tributed to the greatest declines in mental health
problems. A later meta-analysis identified positive Protocol and registration
effects on self-esteem following interventions that only The protocol was registered with the International Database for
used physical activity, yet significance did not emerge Prospective Register of Systematic Reviews PROSPERO (No.
during interventions that combined additional compo- CRD42018105356) and conducted in line with the Preferred
Reporting Items for Systematic Reviews and Meta-Analyses
nents (Liu, Wu, & Ming, 2015). Inconsistent evidence for
(PRISMA) guidelines (Moher et al., 2015; Popay et al., 2006;
the impact of multicomponent (Brown et al., 2013) and Shamseer et al., 2015). A deviation from the protocol includes a
purely physical activity-based interventions (Liu et al., change to population. Due to the number of studies identified
2015) suggests that further investigation is needed to during the original search, the review was split into interven-
determine the vital components necessary to enhancing tions for children and adolescents. Hence, the current review
overall mental health. focuses on studies that included participants aged 11 to 19.
Previous reviews have evaluated the impact of physi-
cal activity interventions on anxiety, depression and Search strategy
self-esteem outcomes, yet a wide range of factors that Nine databases were searched for studies published between
can contribute to mental health and well-being have January 2005 and June 2020: Web of Science, ProQuest Psy-
chology Journals, PsycINFO, Pub Med, ASSIA, CINHAL PLUS,
been overlooked. Peer relationships, body image, life sat-
SPORTDiscus, EMBASE and Wiley Online Library. The search
isfaction, resilience and stress can all impact mental comprised of three elements: intervention (physical activity),
health or ill-health during adolescence (Currie et al., population (adolescent) and outcome (mental health). Out-
2009; Jansen, van de Looij-Jansen, de Wilde, & Brug, comes of interest included mental health, self-esteem, emo-
2008; Tennant et al., 2007), and it remains unclear tional functioning, depression, anxiety, stress, body image,
whether physical activity interventions enhance these resilience, quality of life, relationships, life satisfaction and
social inclusion. For an outline of the search strategy, please see
outcomes too. Although a recent meta-analysis
Figure 1. Articles were independently screened by two authors
(Rodriguez-Ayllon et al., 2019) considered a broader (GH and LC), and a Kappa statistic of .802 (95% CI .592–1.00,
range of outcomes – including happiness, life satisfac- p = .000) indicated substantial inter-rater consistency (Viera &
tion and self-image – interventions were restricted to Garrett, 2005). Data extraction was carried out by two authors,
those purely using physical activity. Of the 12 interven- and additional studies were identified by manually checking ref-
tions included in the analysis, a small but significant erence lists of included papers.
effect was found on overall adolescent mental health yet
the impact of interventions that incorporate additional Inclusion/exclusion criteria
components (e.g. health education) remains unknown. Inclusion criteria were as follows: (a) quantitative peer-reviewed
It is important to review the impact of both multicompo- studies published in English or easily translated; (b) physical
activity was the main intervention component (≥50% of deliv-
nent and solely physical activity programmes, as the ery); and (c) participants’ mean age was 11 to 19. There were no
WHO (2009) states that interventions that incorporate comparator restrictions as although using a control is prefer-
physical activity and additional health education compo- able, it is not always appropriate to withhold physical activity
nents are most effective in modifying adolescent lifestyle from young people (Annesi, Tennant, Westcott, Faigenbaum, &
factors (e.g. physical inactivity, poor diet) and can be Smith, 2009; von Hippel, Powell, Downey, & Rowland, 2007)
expected to have similar impact on mental health. and can lead to schools or other sectors being unwilling to coop-
erate.
Increasing health literacy via multicomponent interven-
Exclusion criteria were as follows: (a) samples with physical
tion has the potential to build resilience and improve health problems (excluding obesity); (b) samples with develop-
well-being (Kickbush et al., 2013), and the European mental disorders or severe mental health problems; (c) interven-
Commission maintains that including a specific mental tions delivered in psychiatric or prison settings; and (d)
health component is fundamental to promoting good interventions delivered in combination with other therapies (e.g.
mental health and preventing the development of prob- Cognitive Behavioural Therapy [CBT], pharmacological treat-
lems in later life (Kutcher et al., 2016; Rampazzo et al., ment). Samples with common mental health problems, such as
anxiety or depression, were eligible for inclusion. Many adoles-
2015).
cents experience symptoms without receiving clinical diagnosis
Addressing adolescent mental health literacy has (Mental Health Foundation, 2018); hence, it was anticipated
gained reasonable attention in school-based pro- that studies not using anxiety/depression criterion would
grammes in recent years (Kutcher et al., 2016; Ram- include participants presenting with mild or moderate symp-
pazzo et al., 2015), yet it is unclear whether the topic has toms.
been combined with a multicomponent physical activity
framework to date. Thus, the current review aimed to Risk of bias analysis
identify both multicomponent interventions and solely
Risk of bias was assessed using the Cochrane Collaboration tool
physical activity programmes and review the impact on a (Higgins et al., 2011) by reporting high, low or unclear risk for
broad range of mental health outcomes that have been four domains: selection bias (i.e. sequence generation, alloca-
previously overlooked. Taking previous evidence into tion concealment), attrition bias (i.e. incomplete outcome data),
account, the aim of the present review was to (a) provide reporting bias (i.e. selective outcome reporting) and detection

© 2021 Association for Child and Adolescent Mental Health

doi:10.1111/camh.12485 Review of physical activity interventions and mental health 3

The following example search was adapted for each database: TITLE-ABS-KEY “exercise
intervention” OR “physical activity intervention” OR “physical exercise intervention” OR “sport
intervention” OR “yoga intervention” AND TITLE-ABS-KEY “mental health” OR
“psychological well-being” OR “mental well-being” OR “self-esteem” OR “emotional
functioning” OR “depression” OR “anxiety” OR “stress” OR “body image” OR “resilience” OR
“quality of life” OR “relationships” OR “life satisfaction” OR “social inclusion” AND TITLE-
ABS-KEY “children” OR “young people” OR “adolescen*” OR “teenage*” OR “youth”
PUBYEAR > 2005.

Studies identified through database searching Additional studies identified through other sources
(n = 14,650) (n = 8)

Duplicates removed
(n = 3,268)

Studies screened at title and

abstract Studies excluded
(n = 11,390) (n = 11,153)

Studies excluded after full-text

reading (n = 209)

Full-text articles assessed for Different intervention (n = 74)

eligibility Not relevant outcomes (n = 60)
(n = 237) Different study design (n = 29)
Different population (n = 40)
Unable to translate (n = 6)

Studies included in synthesis

(n = 28)

Figure 1. Study selection process PRISMA flow chart

bias (i.e. other potential threats). Performance bias (i.e. blinding groups and outcome measures (Higgins & Green, 2011). To
of participants, personnel and outcome assessors) was not ensure caution was taken with interpreting findings, a narrative
assessed as, due to the nature of PA interventions and use of synthesis of identified studies was employed instead. A narra-
self-report, blinding was not considered a high risk. Pre–post tive synthesis was conducted by organising outcomes into four
designs were not assessed as they have unquestionable bias by categories: (a) quality of life (QOL); (b) self-esteem; (c) psycholog-
not using a comparator (Higgins et al., 2011). ical well-being; and (d) psychological ill-being. Effect sizes
(Cohen’s d) were calculated to supplement findings where possi-
Quality assessment ble (Sullivan & Feinn, 2012; Wilson, 2013) using Cohen’s (1992)
original suggestion of small (d = 0.2), medium (d = 0.5) and
Methodological quality was assessed using the National Heart
large effect (d = 0.8). Effect size has been considered in conjunc-
Lung and Blood Institute (2014) Quality Assessment Tool for
tion with methodological quality (i.e. sample size, comparator)
Before-After (Pre-Post) Studies with No Control Group, as stud-
throughout the review.
ies that did not use a comparator were eligible for inclusion but
were at greatest risk of bias (Higgins & Green, 2011). The quality
assessment focused on specification of eligibility criteria, gener- Results
alisability, intervention description, outcome assessment and
incomplete data. There were 12 items assessed in the quality After removing duplicates, 11,390 articles were screened
assessment as shown in Appendix S1, and two authors inde- at abstract from which 11,155 were excluded for not
pendently each item for risk of bias and any discrepancies were
meeting criteria. The remaining papers were examined
resolved through discussion.
at full-text, and subsequently, 28 interventions were
included in the review. Due to the range of outcomes
Synthesis of findings
reported, studies were characterised into four overarch-
A decision was made not to include a meta-analysis in the cur-
rent review due to clear heterogeneity across included studies
ing categories. Among studies, seven used a measure of
with regard to study designs, intervention types, comparison QOL, 13 measured self-esteem, nine measured

© 2021 Association for Child and Adolescent Mental Health

4 Gabrielle E. Hale et al. Child Adolesc Ment Health 2021; *(*): **–**

psychological well-being (including emotional func-

tioning, resilience and positive mood) and 10 mea-
Intervention characteristics
Designs included randomised controlled trials (RCT)
sured psychological ill-being (including anxiety,
(N = 17) (Butzer et al., 2017; Costigan et al., 2016; Dud-
depression, stress and negative mood). No identified
ley et al., 2010; Dunker & Claudino, 2018; Eather et al.,
studies measured outcome ‘relationships’ or ‘social
2016; Ha et al., 2015; Haden et al., 2014; Ho et al., 2017;
inclusion’.
Kalak et al., 2012; Lindwall & Lindgren, 2005; Lubans
et al., 2012; Lubans, Smith, et al., 2016; Morgan et al.,
Participant characteristics
2012; Neumark-Sztainer et al., 2010; Roh et al., 2018;
Sample size varied from 9 (Hainsworth et al., 2018) to
Schranz et al., 2013; Smith et al., 2018), nonrandomised
1,386 (Ha, Burnett, Sum, Medic, & Ng, 2015) (mean
(N = 8) (Bahram et al., 2014; Bonhauser et al., 2005;
[M] = 231; standard deviation [SD] = 297.6) and mean
Elnaggar & Shendy, 2016; Felver et al., 2020; H€ oner &
age of participants ranged from 11.4 (Roh, Cho, & So,
Demetriou, 2014; Lopera et al., 2016; Schneider et al.,
2018) to 18.3 years (Kalak et al., 2012). Four studies
2008; Toulabi et al., 2012), single-group pre–post inter-
used male-only groups (Bahram, Akkasheh, & Akka-
ventions (N = 2) (Dopp et al., 2012; Hainsworth et al.,
sheh, 2014; Lubans, Smith, et al., 2016; Morgan, Saun-
2018) and a three-group cohort (Choukse et al., 2018).
ders, & Lubans, 2012; Schranz, Tomkinson, Parletta,
Of the studies that implemented a control, most used PE
Petkov, & Olds, 2013), six studies used female-only
as usual as the comparator (N = 13) (Bonhauser et al.,
groups (Dudley, Okely, Pearson, & Peat, 2010; Dunker &
2005; Butzer et al., 2017; Costigan et al., 2016; Dudley
Claudino, 2018; Lindwall & Lindgren, 2005; Lubans
et al., 2010; Dunker & Claudino, 2018; Eather et al.,
et al., 2012; Neumark-Sztainer et al., 2010; Schneider,
2016; Ha et al., 2015; Haden et al., 2014; H€oner & Deme-
Dunton, & Cooper, 2008) and remaining studies used
triou, 2014; Lubans et al., 2012; Lubans, Smith, et al.,
mixed-sex. Inclusion of transgender or gender fluid par-
2016; Schneider et al., 2008; Toulabi et al., 2012). Four
ticipants were not reported by any study. Of studies
studies used no intervention (Bahram et al., 2014; Lop-
reporting ethnicity, samples were predominantly White
era et al., 2016; Roh et al., 2018; Schranz et al., 2013),
(N = 9) (Butzer, LoRusso, Shin, & Khalsa, 2017; Dopp,
four used sedentary activities (e.g. access to health edu-
Mooney, Armitage, & King, 2012; Felver, Razza, Morton,
cation website) (Elnaggar & Shendy, 2016; Felver et al.,
Clawson, & Mannion, 2020; Haden, Daly, & Hagins,
2020; Ho et al., 2017; Kalak et al., 2012), three used a
2014; Hainsworth et al., 2018; Lubans et al., 2012;
wait-list control (Lindwall & Lindgren, 2005; Morgan
Lubans, Smith, et al., 2016; Schneider et al., 2008;
et al., 2012; Smith et al., 2018) and one used a female-
Smith et al., 2018), Asian (N = 2) (Ho et al., 2017; Roh
only PE class (Neumark-Sztainer et al., 2010).
et al., 2018) or African American (N = 1) (Neumark-
Interventions were mainly delivered in schools either
Sztainer et al., 2010). Five studies included overweight/
in replacement of regular PE (N = 15) (Bonhauser et al.,
obese participants (BMI > 85th percentile for age and
2005; Butzer et al., 2017; Costigan et al., 2016; Dudley
sex) (Elnaggar & Shendy, 2016; Hainsworth et al., 2018;
et al., 2010; Eather et al., 2016; Felver et al., 2020;
Lopera et al., 2016; Schranz et al., 2013; Toulabi, Nikoo,
Haden et al., 2014; H€ oner & Demetriou, 2014; Lubans
Amini, Nazari, & Mardani, 2012), and one included ado-
et al., 2012; Lubans, Smith, et al., 2016; Morgan et al.,
lescents that screened positive for depression (Dopp
2012; Neumark-Sztainer et al., 2010; Schneider et al.,
et al., 2012) using the Children’s Depression Rating
2008; Smith et al., 2018; Toulabi et al., 2012), as an
Scale Revised (CDSR-R > 36) (Poznanski & Mokros,
additional sessions (N = 1) (Ha et al., 2015) or as a before
1996).
or after-school programme (N = 3) (Dunker & Claudino,
Eleven studies required low activity levels for eligibility
2018; Ho et al., 2017; Kalak et al., 2012). Other settings
(Bahram et al., 2014; Dudley et al., 2010; Dunker &
included community (N = 6) (Bahram et al., 2014;
Claudino, 2018; Ho et al., 2017; Lindwall & Lindgren,
Choukse et al., 2018; Lindwall & Lindgren, 2005; Lopera
2005; Lubans et al., 2012; Lubans, Smith, et al., 2016;
et al., 2016; Roh et al., 2018; Schranz et al., 2013) and
Morgan et al., 2012; Neumark-Sztainer et al., 2010; Roh
laboratory clinics (N = 3) (Dopp et al., 2012; Elnaggar &
et al., 2018; Schneider et al., 2008), yet definitions
Shendy, 2016; Hainsworth et al., 2018). Interventions
varied, for example less than 60 min per day (Dunker &
often delivered more than type of activity. Frequently
Claudino, 2018), less than once per week (Lindwall
used activities included cardiovascular/aerobic (N = 21)
& Lindgren, 2005). Participants were mainly recruited
(Bahram et al., 2014; Bonhauser et al., 2005; Costigan
from schools (N = 21) (Bahram et al., 2014; Bonhauser
et al., 2016; Dopp et al., 2012; Dudley et al., 2010; Dun-
et al., 2005; Butzer et al., 2017; Costigan, Eather, Plot-
ker & Claudino, 2018; Eather et al., 2016; Elnaggar &
nikoff, Hillman, & Lubans, 2016; Dudley et al., 2010;
Shendy, 2016; Ha et al., 2015; H€ oner & Demetriou,
Dunker & Claudino, 2018; Eather, Morgan, & Lubans,
2014; Kalak et al., 2012; Lindwall & Lindgren, 2005;
2016; Felver et al., 2020; Ha et al., 2015; Haden et al.,
Lopera et al., 2016; Lubans et al., 2012; Lubans, Smith,
2014; Ho et al., 2017; Kalak et al., 2012; Lindwall &
et al., 2016; Morgan et al., 2012; Neumark-Sztainer
Lindgren, 2005; Lubans et al., 2012; Lubans, Smith,
et al., 2010; Roh et al., 2018; Schneider et al., 2008;
et al., 2016; Morgan et al., 2012; Neumark-Sztainer
Smith et al., 2018; Toulabi et al., 2012), yoga (N = 12)
et al., 2010; Roh et al., 2018; Schneider et al., 2008;
(Butzer et al., 2017; Choukse et al., 2018; Dudley et al.,
Smith et al., 2018; Toulabi et al., 2012). However, some
2010; Dunker & Claudino, 2018; Felver et al., 2020;
were recruited from clinical settings (N = 3) (Dopp et al.,
Haden et al., 2014; Hainsworth et al., 2018; Lindwall &
2012; Elnaggar & Shendy, 2016; Hainsworth et al.,
Lindgren, 2005; Lubans et al., 2012; Neumark-Sztainer
2018), community settings (N = 2) (Choukse, Ram, &
et al., 2010; Schneider et al., 2008; Smith et al., 2018)
Nagendra, 2018; H€ oner & Demetriou, 2014) or via adver-
and resistance training (N = 10) (Costigan et al., 2016;
tisements (N = 2) (Lopera et al., 2016; Schranz et al.,
Dunker & Claudino, 2018; Eather et al., 2016; Lopera
2013).

© 2021 Association for Child and Adolescent Mental Health

doi:10.1111/camh.12485 Review of physical activity interventions and mental health 5

et al., 2016; Lubans et al., 2012; Lubans, Smith, et al., based on an Ecological Model of Health (Sallis & Owen,
2016; Morgan et al., 2012; Neumark-Sztainer et al., 2002) and intervened at the environmental and social
2010; Schranz et al., 2013; Smith et al., 2018). Six inter- level, by incorporating sports leaders and resources (Ha
ventions included team sport (Bonhauser et al., 2005; et al., 2015). A Positive Youth Development (PYD) frame-
Ho et al., 2017; Lubans et al., 2012; Lubans, Smith, work (Lerner, 2009) was used to strengthen developmen-
et al., 2016; Morgan et al., 2012; Schneider et al., 2008). tal assets through sport during one study, and coaches
A large proportion were multicomponent (N = 19) with were asked to act as mentors to encourage group discus-
most interventions including education sessions focused sion on goal setting (Ho et al., 2017).
on health (e.g. portion control) (Dunker & Claudino,
2018; Ha et al., 2015; H€ oner & Demetriou, 2014; Kalak Facilitators and training
et al., 2012; Lindwall & Lindgren, 2005; Lopera et al., Interventions were predominantly facilitated by
2016; Lubans et al., 2012; Morgan et al., 2012; schoolteachers (N = 14) (Bonhauser et al., 2005; Butzer
Neumark-Sztainer et al., 2010; Schneider et al., 2008; et al., 2017; Dudley et al., 2010; Dunker & Claudino,
Smith et al., 2018; Toulabi et al., 2012) or behavioural 2018; Ha et al., 2015; Haden et al., 2014; H€ oner & Deme-
skills (e.g. goal setting, team behaviour) (Costigan et al., triou, 2014; Lubans et al., 2012; Lubans, Smith, et al.,
2016; Eather et al., 2016; Ho et al., 2017; Lopera et al., 2016; Morgan et al., 2012; Neumark-Sztainer et al.,
2016; Lubans et al., 2012; Lubans, Smith, et al., 2016; 2010; Schneider et al., 2008; Smith et al., 2018; Toulabi
Morgan et al., 2012). One study included additional et al., 2012) yet the training received varied. Two studies
components focused on journaling and storytelling (Butzer et al., 2017; Haden et al., 2014) provided
(Choukse et al., 2018). A small number (N = 5) 200 hours of certified yoga training, whereas others
addressed psychological topics including self-esteem, offered workshops, professional development days and
body image or stress (Butzer et al., 2017; Dunker & support from the research team (Ha et al., 2015; Lubans,
Claudino, 2018; Felver et al., 2020; Lopera et al., 2016; Smith, et al., 2016). Other interventions were facilitated
Neumark-Sztainer et al., 2010). No identified multicom- by professionals with relevant expertise (N = 10) includ-
ponent interventions addressed mental health literacy ing accredited CrossFit trainers (Eather et al., 2016),
(e.g. developing understanding of mental health, reduc- yoga practitioners (Felver et al., 2020; Hainsworth et al.,
ing stigma and help-seeking). Key messages advocated 2018), sports coaches (Ho et al., 2017), exercise instruc-
by additional components were reinforced using work- tors (Lindwall & Lindgren, 2005; Lopera et al., 2016),
shops, information booklets, parental letters and mobile Taekwondo experts (Roh et al., 2018), exercise psycholo-
applications/websites. gists (Bahram et al., 2014), physical therapists (Elnaggar
& Shendy, 2016) and exercise science students (Schranz
Theory et al., 2013). Three interventions were solely delivered by
Six interventions (Dudley et al., 2010; Dunker & Clau- the research team (Costigan et al., 2016; Dopp et al.,
dino, 2018; Lubans et al., 2012; Morgan et al., 2012; 2012; Kalak et al., 2012), and one yoga camp was facili-
Neumark-Sztainer et al., 2010; Smith et al., 2018) were tated by undergraduate students (Choukse et al., 2018).
underpinned by Social Cognitive Theory (SCT) (Bandura Some multicomponent interventions were delivered by
& Walters, 1977) which suggests that knowledge, out- multidisciplinary teams (N = 6) including dieticians,
come expectations, self-efficacy, perceived facilitators nutritionists and psychologists (Dunker & Claudino,
and goals are required to promote motivation to change 2018; Lopera et al., 2016; Lubans et al., 2012; Lubans,
(Bandura, 2004; Bandura & Walters, 1977). The Physi- Richards, et al., 2016; Morgan et al., 2012; Neumark-
cal Activity Leaders (PALS) intervention (Morgan et al., Sztainer et al., 2010).
2012) encouraged adolescents to become physically
active and set their own goals through barrier Length
identification, encouragement and instruction. Four Five interventions were delivered for under 7 weeks
interventions based on Self-Determination Theory (SDT) (Choukse et al., 2018; Felver et al., 2020; Ha et al., 2015;
(Ryan & Deci, 2000) promoted adherence, autonomy, Kalak et al., 2012; Toulabi et al., 2012), whereas most
competence and relatedness by creating supportive and (N = 12) lasted between 8 and 12 weeks (Bahram et al.,
enjoyable environments (Costigan et al., 2016; Lubans 2014; Costigan et al., 2016; Dopp et al., 2012; Dudley
et al., 2012; Lubans, Smith, et al., 2016; Smith et al., et al., 2010; Dunker & Claudino, 2018; Eather et al.,
2018). Another study followed an ‘adult learning’ 2016; Elnaggar & Shendy, 2016; Haden et al., 2014;
approach (Vella, 1995) which maintains that individuals Hainsworth et al., 2018; H€ oner & Demetriou, 2014; Mor-
are active decision-makers in the learning process. In gan et al., 2012; Smith et al., 2018). Five were delivered
line with this approach, students were involved in between 16 and 20 weeks (Ho et al., 2017; Lopera et al.,
intervention development during one school-based 2016; Lubans, Smith, et al., 2016; Neumark-Sztainer
programme by selecting the sports that were offered et al., 2010; Roh et al., 2018), and the remaining 6 inter-
(Bonhauser et al., 2005). ventions lasted over 6 months (Bonhauser et al., 2005;
The Exercise and Self-Esteem Model (EXSEM) (Son- Butzer et al., 2017; Lindwall & Lindgren, 2005; Lubans
stroem & Morgan, 1989) provided the framework for et al., 2012; Schneider et al., 2008; Schranz et al., 2013).
CrossFit Teens (Eather et al., 2016). EXSEM advocates Frequency and length of sessions also varied. One study
that interventions that develop physical skills and (Costigan et al., 2016) utilised 10-minute bursts of activ-
competences (i.e. fitness, ability) indirectly influence ity three times a week, whereas one 10-day camp deliv-
self-esteem. Therefore, awareness of improvement in ered 8 hours of daily yoga (Choukse et al., 2018). In line
physical competences was reinforced by CrossFit with WHO (Chaput et al., 2020) recommendations, most
instructors through verbal encouragement, goal setting interventions (N = 20) were delivered for at least 60 to
and measurement of goal achievement. One study was 90 min per session (Bonhauser et al., 2005; Choukse

© 2021 Association for Child and Adolescent Mental Health

6 Gabrielle E. Hale et al. Child Adolesc Ment Health 2021; *(*): **–**

et al., 2018; Dopp et al., 2012; Dudley et al., 2010; Dun- et al., 2015). Effect sizes were minor (d < 0.10), and
ker & Claudino, 2018; Eather et al., 2016; Haden et al., numerous negative values suggested that PE as usual
2014; Hainsworth et al., 2018; Ho et al., 2017; H€ oner & experienced larger improvements.
Demetriou, 2014; Lindwall & Lindgren, 2005; Lubans
et al., 2012; Lubans, Richards, et al., 2016; Morgan Self-esteem
et al., 2012; Neumark-Sztainer et al., 2010; Roh et al., Definitions of self-esteem and self-perception tend to be
2018; Schneider et al., 2008; Schranz et al., 2013; Smith used interchangeably (Fox, 1997; King, 1997), and this
et al., 2018; Toulabi et al., 2012). However, due to varia- ambiguity is evident throughout identified studies
tion in number of sessions, total amount of weekly activ- (Table S2). Most measured global self-esteem, whereas
ity ranged from 15 (Ha et al., 2015) to 270 (Haden et al., some assessed self-perception sub-domains of physical
2014) minutes across studies. self-worth and physical appearance. Improvements were
seen across all samples, yet school-based interventions
Risk of bias had greater influence on vulnerable populations (Eather
The highest risk for the 25 studies assessed was selec- et al., 2016; Smith et al., 2018). Whilst comparing PE to
tion bias. Four studies were judged high risk for not CrossFit Teens (Eather et al., 2016) larger effect sizes for
using adequate sequence generators to assign partici- Physical Self-Description Questionnaire (PSDQ; Marsh,
pants to intervention and control groups (Felver et al., 1996) subscales were seen in adolescents ‘at-risk’ of
2020; H€oner & Demetriou, 2014; Lopera et al., 2016; developing mental health problems in the future. Simi-
Schneider et al., 2008), and seven studies provided larly, following Resistance Training for Teens (Smith
insufficient information to determine a judgement of et al., 2018), moderator analysis demonstrated greater
high or low risk (Bonhauser et al., 2005; Eather et al., improvements in overweight/obese participants com-
2016; Kalak et al., 2012; Morgan et al., 2012; Neumark- pared to the full sample. Support for resistance training
Sztainer et al., 2010; Roh et al., 2018; Toulabi et al., was identified during numerous studies (Cositgan et al.,
2012). One study was judged high risk for incomplete 2016; Morgan et al., 2012; Schranz et al., 2013). An RCT
outcome data (Elnaggar & Shendy, 2016) as loss to that compared aerobic (AEP) and resistance-based (RAP)
follow-up was mentioned yet it was unclear whether this HIIT to PE found larger effects for PSDQ Perceived
was adequately addressed during analysis. All other Appearance following RAP (d = 0.32), suggesting that
domains were judged as ‘low’ risk or ‘unclear’ due to resistance activities can have more impact than aerobic
insufficient information being provided (see activity alone (d = 0.10).
Appendix S2). Gender appears to moderate impact, as most studies
that did not identify improvements used female-only
samples (Lindwall & Lindgren, 2005; Lubans et al.,
Effect of interventions
2012; Schneider et al., 2008). The Nutrition and Enjoy-
Quality of life able Activity for Teen Girls (NEAT Girls) intervention
Among studies that measured QOL, overweight/obese (Lubans et al., 2012) did not find significant differences
samples (Elnaggar & Shendy, 2016; Hainsworth et al., and negative effects for PSDQ Perceived Body Fat and
2018; Roh et al., 2018) or adolescents considered physi- Global Self-Esteem favoured PE. There was also low
cally inactive (Bahram et al., 2014; Ho et al., 2017) expe- adherence to NEAT Girls (60.6% attendance), highlight-
rienced significant improvements in at least one ing the need for interventions to address barriers to
subscale (Table S1 found in the Supporting Informa- female physical activity participation. The importance of
tion). One study (Elnaggar & Shendy, 2016) identified ensuring sufficient activity levels are met – through fre-
increases in the Pediatric Quality of Life Inventory quency and adherence – has been advocated repeatedly.
(PedsQL; Varni, Seid, & Rode, 1999) Psychosocial sub- One female intervention evaluated the New Moves Pro-
scale in overweight/obese participants assigned to a gramme (NMP) and identified significant improvements
high-intensity-interval training (HIIT) intervention. in SPPA Physical Self-Worth subscale (Neumark-
Improvements in SF-12v2 Psychological QOL were also Sztainer et al., 2010). However, a later analysis of the
found in inactive adolescents following a PYD sports pro- same intervention did not identify a change in Rosenberg
gramme (Ho et al., 2017) compared to the control group Self-Esteem (RSE; Rosenberg, 1965) scores (Dunker &
that had access to a health education website. Linear Claudino, 2018). Although most components were iden-
regression also identified an association between atten- tical, the latter used a shorter time scale (9 vs 16 weeks)
dance and effectiveness, with greater changes occurring and less weekly activity (2 vs 4 sessions). Adherence was
in those who participated in more sessions (Ho et al., also lower, with less than a third taking part in >70% of
2017). sessions (Dunker & Claudino, 2018).
Improvements were not evident following school-
based interventions that compared impact to PE as Psychological well-being
usual on samples not restricted by weight or sedentary Marginally larger effects for resistance HIIT (Costigan
criteria (Ha et al., 2015; H€
oner & Demetriou, 2014). One et al., 2016) and the Active Teen Leaders Avoiding
study (H€ oner & Demetriou, 2014) observed a QOL reduc- Screen-time (ATLAS) intervention (Lubans, Smith, et al.,
tion that was significant for schools delivering Health 2016) suggested that resistance training has greater
Promoting PE but not those assigned to control. Another impact on well-being than aerobics alone (Table S3).
study investigated the impact of the School-based, However, the Resistance Training for Teens did not
Train-the-trainer, Accessibility of resources, Recre- observe a change (Smith et al., 2018), and the length of
ational (STAR) programme, which embedded rope skip- delivery was considerably shorter than ATLAS (10 vs
ping into PE sessions and provided resources to 20 weeks). Significantly greater improvements in resili-
encourage activity during breaks and after school (Ha ence were also identified following a PYD sports

© 2021 Association for Child and Adolescent Mental Health

doi:10.1111/camh.12485 Review of physical activity interventions and mental health 7

intervention (Ho et al., 2017) yet an association with and offers further insight into intervention effectiveness
attendance emphasises the importance of adherence to in areas such as QOL and psychological distress. Over-
experience maximum benefits. all, reviewed studies present the widespread use of phys-
Taekwondo led to a significant increase in Vigour (Roh ical activity types and additional components in school
et al., 2018) yet evidence for yoga interventions and and community settings. Findings suggest that inter-
mood was unclear. During the Kripalu Yoga in the ventions are predominantly useful in improving QOL
Schools (KYIS) intervention (Butzer et al., 2017), a nega- and psychological well-being. Consistent with previous
tive effect revealed that PE experienced larger improve- research (Biddle & Asare, 2011; Biddle et al., 2019), a
ments after 6 months, whereas yoga led to greater positive impact on self-esteem was highlighted yet find-
increases in Positive Affect following a 12-week interven- ings were mixed. Despite the positive impact on psycho-
tion (Haden et al., 2014). Although length of delivery was logical well-being, reduction in psychological ill-being in
shorter than KYIS, the amount of weekly activity was the general population is less clear.
greater (270 vs 35 min) suggesting that frequency is key Although a recent review identified small-to-moderate
to effective intervention development. However, impact effects on paediatric QOL, most included interventions
of yoga on emotional functioning was evident, with sig- focused on chronically ill samples (Marker, Steele, &
nificant changes emerging following a 10-day camp Noser, 2018). The current findings offer further support
(Choukse et al., 2018) and a recent replication of KYIS and highlight the importance of population, as interven-
(Felver et al., 2020). tions appear most effective on vulnerable adolescents
(Bahram et al., 2014; Hainsworth et al., 2018). Drawing
Psychological ill-being from interventions that did not identify change (Ha et al.,
Outcomes indicative of poor mental health included psy- 2015), there is a need for more trials to explore QOL
chological distress, negative mood, anxiety and depres- using wait-list controls to determine the true effect and
sion (Table S4). CrossFit Teens intervention (Eather frequency of activity required to create impact on healthy
et al., 2016) found small effects using the Strength and samples. Inconsistencies across self-esteem measures
Difficulties Questionnaire (SDQ; Goodman, 2001) on the made it difficult to draw conclusions and studies pre-
full sample, yet larger improvements were seen in those sented a mixed picture. Previous research has found
‘at-risk’. However, most evidence gathered for psycholog- gender differences during adolescence with males dis-
ical distress was weak across various interventions, playing higher self-esteem (McClure, Tanski, Kingsbury,
including yoga (Butzer et al., 2017), running (Kalak Gerrard, & Sargent, 2010; Moksnes & Reindunsdatter,
et al., 2012) and HIIT (Costigan et al., 2016). Evidence 2019), and this discrepancy is evident in the current
gathered for negative mood was also unclear (Butzer review. Most interventions that did not observe change
et al., 2017; Haden et al., 2014; Roh et al., 2018). Taek- in self-esteem used female-only samples, indicating that
wondo (Roh et al.,2018) led to significant reductions in enhancing this psychological factor in females is more
Tension and Depression, yet no other differences were complex. Gender differences can also be owed to self-
apparent (Anger, Fatigue, Confusion) and effect sizes presentation bias in existing instruments (Gentile et al.,
were minor (d < .13), whereas small effects were only 2009; Kling, Hyde, Showers, & Buswell, 1999; Tomas,
found for Confusion and Tension following KYIS (Butzer Oliver, Hontangas, Sancho, & Galiana, 2015), with cer-
et al., 2017). tain domains and individual items favouring males, and
Despite limited studies, clearer support was apparent should be considered during the evaluation of female
for anxiety in comparison with depression (Bonhauser adolescent self-esteem in the future. Nonetheless, given
et al., 2005; Dopp et al., 2012; Lindwall & Lindgren, the higher dropout rates in female-only interventions
2005; Toulabi et al., 2012). A female intervention found (Dudley et al., 2010; Lubans et al., 2012; Morgan et al.,
a small effect on Social Physique Anxiety Scale (SPAS) 2012), associations between adherence and effect (Ho
compared to wait-list control (Lindwall & Lindgren, et al., 2017) and evidence for lower MVPA in the wider lit-
2005) and a mixed-gender sport programme revealed an erature (Farooq et al., 2018), it is vital to ensure interven-
improvement in Hospital Anxiety Depression Scale tion feasibility and address participation barriers to
(HADS; Zigmond & Snaith, 1983) Anxiety subscale but support female engagement. New support was provided
not Depression (Bonhauser et al., 2005). Only one inter- for impact on well-being, resilience and emotional func-
vention displayed reductions in depression, with partici- tioning, yet the small evidence base and heterogeneity
pants displaying clinical levels prior to taking part (Dopp across measures emphasises the need for more research
et al., 2012). At follow-up, most participants (69.2%) in these areas to obtain a clearer overview.
maintained lower levels of depression or demonstrated Limited studies measuring psychological ill-being
further reductions. Although the study utilised a pre– might be due to several exclusion criteria, including
post design on a small sample (N = 13), findings high- samples with serious mental health problems and inter-
light the potential of using supervised and unsupervised ventions delivered in combination with CBT. Earlier
activity to reduce psychological ill-being in clinical popu- reviews have found stronger evidence for physical activ-
lations. ity interventions and depression (Brown et al., 2013;
Spruit et al., 2016) yet, in line with the present findings,
greater effects are apparent in clinical populations (Car-
Discussion ter, Morres, Meade, & Callaghan, 2016; Dopp et al.,
The present review reports findings from 28 physical 2012). Physical activity can also alleviate stress and anx-
activity interventions and evaluates their effectiveness in iety (Haugland, Wold, & Torsheim, 2003; Martikainen
enhancing adolescent mental health. This work comple- et al., 2013); however, the current work advocates the
ments previous reviews (Biddle et al., 2019; Liu et al., need for more research. With reference to the small
2015; Rodriguez-Ayllon et al., 2019; Spruit et al., 2016) number of studies that measured psychological distress,

© 2021 Association for Child and Adolescent Mental Health

8 Gabrielle E. Hale et al. Child Adolesc Ment Health 2021; *(*): **–**

it is difficult to draw clear conclusions due to hetero- mental health; differentiating between mental health
geneity in intervention components and outcome mea- problems and treatments; reducing stigma surround-
sures. However, a greater effect on psychological distress ing mental health problems; and promoting help-
following resistance-based HIIT (Costigan et al., 2016) seeking behaviours and mental health self-care
highlighted the impact of this form of exercise compared (Kutcher et al., 2016; Rampazzo et al., 2015). Despite
to aerobic activity alone via traditional PE. Larger effects both physical activity and mental health literacy
following interventions incorporating resistance training interventions for adolescents gaining considerable
were evident across various outcomes (Costigan et al., attention in recent years (Biddle et al., 2019; Chaput
2016; Eather et al., 2016; Lopera et al., 2016; Neumark- et al., 2020; Kutcher et al., 2016; Rampazzo et al.,
Sztainer et al., 2010; Schranz et al., 2013; Smith et al., 2015), it is evident that the two concepts have not
2018) and support growing interest into the effectiveness been combined within a multicomponent framework
of resistance-based activity and HIIT in enhancing ado- to date. This highlights an interesting pathway for
lescent mental health (Collins, Booth, Duncan, Faw- future investigation, as additional education sessions
kner, & Niven, 2019; Leahy et al., 2020; Smith et al., could be used to address mental health literacy whilst
2014). Given that anxiety and psychological distress emphasising the important link between physical
(American Psychological Association, 2014; Biddle et al., activity and positive mental health. Emphasising
2019; Chiang et al., 2019) is frequently experienced by mental health education could improve effectiveness
adolescents and can intensify during key developmental of existing interventions in enhancing well-being and
stages, it is crucial for more interventions to explore the decreasing problems in adolescents from the general
impact of physical activity interventions in reducing population; hence, further investigation is warranted.
these mental health problems in young people from the
general population.
Conclusion
The impact of physical activity interventions on youth
Strengths and Limitations
mental health is evident, and further support has been
The review employed a rigorous methodology as per provided for the effect of such interventions on QOL and
PRISMA recommendations to ensure appropriate litera- psychological well-being. It was revealed that multicom-
ture were identified and evaluated with scientific preci- ponent interventions have not addressed mental health
sion. Registration with PROSPERO guaranteed that education in the same depth that they cover other health
research questions and inclusion/exclusion criteria topics to date. Future interventions aiming to increase
were clearly defined prior to conducting the search to physical activity and develop adolescent health literacy
eliminate any reviewer bias during study selection. An should consider exploring the impact of inclusion of a
extensive search was conducted across a wide range of mental health component. In light of the recent coron-
databases, and non-English papers were translated avirus disease 2019 (COVID-19) pandemic (WHO, 2020)
where possible; however, only published findings were and anticipated rise in related psychological problems
included. There is a risk that studies have not been pub- (Torales, O’Higgins, Castaldelli-Maia, & Ventriglio,
lished due to nonsignificant or unfavourable outcomes 2020), supporting adolescent mental health is critical.
and not including these may have advocated a publica- Multicomponent interventions focussed on mental
tion bias. There was also variability in intervention health literacy could potentially provide a cost-effective
design and outcomes, due to the aim of gaining a com- and convenient pathway to encouraging physical activ-
prehensive view of intervention impact on mental health. ity, promoting psychological well-being and self-care,
This heterogeneity has partly been responsible for diffi- and preventing the likelihood of more serious problems
culties in drawing clear conclusions across studies and developing in later life and thus is an important area for
made a meta-analysis not possible. Nonetheless, the future inquiry.
review provides a thorough exploration of theory and evi-
dence and highlights directions for future research.
Acknowledgements
Direction for future research This paper provides an overview of physical activity interven-
Numerous interventions were multicomponent with tions for adolescent mental health, incorporating quality of life,
the majority using additional sessions to educate par- self-esteem and psychological well-being. Findings illustrate a
lack of interventions combining physical activity and mental
ticipants on health or behavioural topics. A small
health promotion in a multicomponent framework and identify
number addressed psychological education (Butzer a pathway for future development. Named authors have sub-
et al., 2017; Dunker & Claudino, 2018; Felver et al., stantially contributed to the review. The authors have declared
2020; Lopera et al., 2016; Neumark-Sztainer et al., that they have no competing or potential conflicts of interest.
2010) yet topics were restricted to self-esteem, body
image or stress management and no other areas of
mental health were explored. This gap identifies an Ethical approval
area for future development, as the WHO advocates
the importance of health literacy for improving health No ethical approval was required for this review article.
outcomes (Kickbusch et al., 2013) and arguably
applies to mental as well as physical health. The Correspondence
European Commission maintains that literacy is
essential to enhance youth mental health and inter- Gabrielle E. Hale, School of Psychology and Therapeutic
ventions should address four areas: developing under- Studies, University of South Wales, Pontypridd, CF37
standing of how to promote and maintain positive 1DL, UK; Email: [email protected]

© 2021 Association for Child and Adolescent Mental Health

doi:10.1111/camh.12485 Review of physical activity interventions and mental health 9

Supporting information Chaput, J.P., Willumsen, J., Bull, F., Chou, R., Ekelund, U.,
Firth, J., . . . & Katzmarzyk, P.T. (2020). 2020 WHO guidelines
Additional Supporting Information may be found in the online on physical activity and sedentary behaviour for children and
version of this article: adolescents aged 5–17 years: summary of the evidence. Inter-
national Journal of Behavioral Nutrition and Physical Activity,
Table S1. Effect of physical activity interventions on adoles- 17, 1–9.
cent quality of life. Chiang, J.J., Ko, A., Bower, J.E., Taylor, S.E., Irwin, M.R., &
Table S2. Effect of physical activity interventions on adoles- Fuligni, A.J. (2019). Stress, psychological resources, and
cent self-esteem. HPA and inflammatory reactivity during late adolescence.
Table S3. Effect of physical activity interventions on adoles- Development and Psychopathology, 31, 699–712.
cent psychological well-being. Choukse, A., Ram, A., & Nagendra, H.R. (2018). Effect of resi-
Table S4. Effect of physical activity interventions on adoles- dential yoga camp on psychological fitness of adolescents: A
cent adverse psychological ill-being. cohort study. Journal of Clinical & Diagnostic Research, 12(8),
Appendix S1. NHLBI quality assessment table. https://doi.org/10.7860/jcdr/2018/36839.11872
Appendix S2. Risk of bias summary. Cohen, J. (1992). Quantitative methods in psychology: A power
primer. Psychological Bulletin, 112, 1155–1159.
Cohen, S., Kamarck, T., & Mermelstein, R. (1994). Perceived
References stress scale. Measuring stress: A guide for health and social
Ahn, S., & Fedewa, A.L. (2011). A meta-analysis of the relation- scientists, 10.
ship between children’s physical activity and mental health. Collins, H., Booth, J.N., Duncan, A., Fawkner, S., & Niven, A.
Journal of Pediatric Psychology, 36, 385–397. (2019). The effect of resistance training Interventions on ‘The
American Psychological Association (2014). Stress in America: Self’ in youth: A systematic review and meta-analysis. Sports
Are teens adopting adults’ stress habits. Stress in America Medicine-Open, 5, 29.
Surveys. http://www/apa.org/news/press/releases/stre Costigan, S.A., Eather, N., Plotnikoff, R.C., Hillman, C.H., &
ss/2013/stress-report.pdf. Lubans, D.R. (2016). High-intensity interval training on cog-
Annesi, J.J., Tennant, G., Westcott, W.L., Faigenbaum, A.D., & nitive and mental health in adolescents. Medicine & Science
Smith, A.E. (2009). Effects of the Youth Fit For Life protocol in Sports & Exercise, 48, 1985–1993.
on physiological, psychological, and behavioral factors at Currie, C., Zanotti, C., Morgan, A., Currie, D., De Looze, M.,
YMCA Calgary after-school care sites. Psychological Reports, Roberts, C., . . . & Barnekow, V. (2009). Social determinants of
104, 879–895. health and well-being among young people. Health Behaviour
Bahram, M.E., Akkasheh, G., & Akkasheh, N. (2014). Aerobics, in School-aged Children (HBSC) study: international report,
quality of life, and physiological indicators of inactive male 2010, 271.
students’ cardiovascular endurances, in Kashan. Nursing Diener, E., Wirtz, D., Tov, W., Kim-Prieto, C., Choi, D.W., Oishi,
and Midwifery Studies, 3(2), https://doi.org/10.17795/ S., & Biswas-Diener, R. (2010). New well-being measures:
nmsjournal10911 Short scales to assess flourishing and positive and negative
Bandura, A. (2004). Health promotion by social cognitive feelings. Social Indicators Research, 97, 143–156.
means. Health Education & Behavior, 31, 143–164. Dopp, R.R., Mooney, A.J., Armitage, R., & King, C. (2012). Exer-
Bandura, A., & Walters, R.H. (1977). Social learning theory, Vol. cise for adolescents with depressive disorders: A feasibility
1. Englewood Cliffs, NJ: Prentice Hall. study. Depression Research and Treatment, 2012, 1–9.
Beauchamp, M.R., Puterman, E., & Lubans, D.R. (2018). Physi- https://doi.org/10.1155/2012/257472
cal inactivity and mental health in late adolescence. JAMA Dudley, D., Okely, A., Pearson, P., & Peat, J. (2010). Engaging
Psychiatry, 75, 543–544. adolescent girls from linguistically diverse and low
Beck, A.T., Steer, R.A., & Carbin, M.G. (1988). Psychometric income backgrounds in school sport: A pilot randomised con-
properties of the Beck Depression Inventory: Twenty-five trolled trial. Journal of Science and Medicine in Sport, 13,
years of evaluation. Clinical Psychology Review, 8, 77–100. 217–224.
Biddle, S.J., & Asare, M. (2011). Physical activity and mental Dunker, K.L.L., & Claudino, A.M. (2018). Preventing weight-
health in children and adolescents: A review of reviews. Bri- related problems among adolescent girls: A cluster random-
tish Journal of Sports Medicine, 45, 886–895. ized trial comparing the Brazilian ‘New Moves’ program ver-
Biddle, S.J., Ciaccioni, S., Thomas, G., & Vergeer, I. (2019). sus observation. Obesity Research & Clinical Practice, 12,
Physical activity and mental health in children and adoles- 102–115.
cents: An updated review of reviews and an analysis of Eather, N., Morgan, P.J., & Lubans, D.R. (2016). Effects of exer-
causality. Psychology of Sport and Exercise, 42, 146–155. cise on mental health outcomes in adolescents: Findings from
Bonhauser, M., Fernandez, G., P€ uschel, K., Ya~
nez, F., Montero, the CrossFitTM teens randomized controlled trial. Psychology
J., Thompson, B., & Coronado, G. (2005). Improving physical of Sport and Exercise, 26, 14–23.
fitness and emotional well-being in adolescents of low socioe- Elnaggar, R.K., & Shendy, M.A. (2016). Aerobic interval exer-
conomic status in Chile: Results of a school-based controlled cises versus dietary control induced changes of adipokines
trial. Health Promotion International, 20, 113–122. levels, lipid profile and quality of life in overweight or obese
Brown, H.E., Pearson, N., Braithwaite, R.E., Brown, W.J., & children. International Journal of Physiotherapy, 3(5),
Biddle, S.J. (2013). Physical activity interventions and https://doi.org/10.15621/ijphy/2016/v3i5/117445
depression in children and adolescents. Sports Medicine, 43, Farooq, M.A., Parkinson, K.N., Adamson, A.J., Pearce, M.S.,
195–206. Reilly, J.K., Hughes, A.R., . . . & Reilly, J.J. (2018). Timing of
Butzer, B., LoRusso, A., Shin, S.H., & Khalsa, S.B.S. (2017). the decline in physical activity in childhood and adolescence:
Evaluation of yoga for preventing adolescent substance use Gateshead Millennium Cohort Study. British Journal of
risk factors in a middle school setting: A preliminary group- Sports Medicine, 52, 1002–1006.
randomized controlled trial. Journal of Youth and Adoles- Felver, J.C., Razza, R., Morton, M.L., Clawson, A.J., & Mannion,
cence, 46, 603–632. R.S. (2020). School-based yoga intervention increases adoles-
Carson, V., Hunter, S., Kuzik, N., Gray, C.E., Poitras, V.J., Cha- cent resilience: A pilot trial. Journal of Child & Adolescent
put, J.P., . . . & Tremblay, M.S. (2016). Systematic review of Mental Health, 32, 1–10.
sedentary behaviour and health indicators in school-aged Fitts, W.H., & Warren, W.L. (1996). Tennessee self-concept
children and youth: an update. Applied Physiology, Nutrition, scale: TSCS-2 (p. 118). Los Angeles: Western Psychological
and Metabolism, 41, S240–S265. Services.
Carter, T., Morres, I.D., Meade, O., & Callaghan, P. (2016). The Fox, K.R. (1997). The physical self and processes in self-esteem
effect of exercise on depressive symptoms in adolescents: A development. In K.R. Fox (Ed.), The Physical Self: From Moti-
systematic review and meta-analysis. Journal of the American vation to Well-being (pp. 111–139). Champaign, IL: Human
Academy of Child & Adolescent Psychiatry, 55, 580–590. Kinetics

© 2021 Association for Child and Adolescent Mental Health

10 Gabrielle E. Hale et al. Child Adolesc Ment Health 2021; *(*): **–**

Fox, K.R., & Corbin, C.B. (1989). The physical self-perception Kessler, R.C., Andrews, G., Colpe, L.J., Hiripi, E., Mroczek,
profile: Development and preliminary validation. Journal of D.K., Normand, S.L., . . . & Zaslavsky, A.M. (2002). Short
Sport and Exercise Psychology, 11, 408–430. screening scales to monitor population prevalences and
Garnefski, N., & Kraaij, V. (2006). Cognitive emotion regulation trends in non-specific psychological distress. Psychological
questionnaire–development of a short 18-item version Medicine, 32, 959–976.
(CERQ-short). Personality and Individual Differences, 41, Kickbusch, I., Apfel, F., Apfel, F., Pelikan, J., Pelikan, J.,
1045–1053. Tsouros, A. (2013). Health literacy. The solid facts. World
Gentile, B., Grabe, S., Dolan-Pascoe, B., Twenge, J.M., Wells, Health Organization.
B.E., & Maitino, A. (2009). Gender differences in domain- King, K.A. (1997). Self-concept and self-esteem: A clarification
specific self-esteem: A meta-analysis. Review of General Psy- of terms. Journal of School Health, 67, 68–71.
chology, 13, 34–45. Kling, K.C., Hyde, J.S., Showers, C.J., & Buswell, B.N. (1999).
Goodman, R. (2001). Psychometric properties of the strengths Gender differences in self-esteem: A meta-analysis. Psycho-
and difficulties questionnaire. Journal of the American Acad- logical Bulletin, 125, 470–500.
emy of Child & Adolescent Psychiatry, 40, 1337–1345. Kutcher, S., Wei, Y., Costa, S., Gusm~ ao, R., Skokauskas, N., &
Ha, A., Burnett, A., Sum, R., Medic, N., & Ng, J. (2015). Out- Sourander, A. (2016). Enhancing mental health literacy in
comes of the rope skipping ‘STAR’ programme for schoolchil- young people. European Child & Adolescent Psychiatry, 25,
dren. Journal of Human Kinetics, 45, 233–240. 567–569.
Haden, S.C., Daly, L., & Hagins, M. (2014). A randomised con- Laurent, J., Catanzaro, S.J., Joiner, T.E., Jr, Rudolph, K.D.,
trolled trial comparing the impact of yoga and physical educa- Potter, K.I., Lambert, S., . . . & Gathright, T. (1999). A measure
tion on the emotional and behavioural functioning of middle of positive and negative affect for children: Scale development
school children. Focus on Alternative and Complementary and preliminary validation. Psychological Assessment, 11,
Therapies, 19, 148–155. 326–338.
Hainsworth, K., Liu, X., Simpson, P., Swartz, A., Linneman, N., Leahy, A.A., Mavilidi, M.F., Smith, J.J., Hillman, C.H., Eather,
Tran, S., . . . & Weisman, S. (2018). A pilot study of Iyengar N., Barker, D., & Lubans, D.R. (2020). Review of high-
yoga for pediatric obesity: Effects on gait and emotional func- intensity interval training for cognitive and mental health in
tioning. Children, 5, 92. youth. Medicine and Science in Sports and Exercise, 52,
Hallal, P.C., Andersen, L.B., Bull, F.C., Guthold, R., Haskell, 2224–2234.
W., Ekelund, U., & Lancet Physical Activity Series Lerner, R.M. (2009). The Positive Youth Development Perspec-
Working Group (2012). Global physical activity levels: tive: Theoretical and empirical bases of strengths-based
Surveillance progress, pitfalls, and prospects. The Lancet, approach to adolescent development. In S.J. Lopez & C.R.
380, 247–257. Snyder (Eds.), Oxford Library of Psychology.Oxford Handbook
Hart, E.A., Leary, M.R., & Rejeski, W.J. (1989). Tie measure- of Positive Psychology (pp. 149–163). Oxford, UK: Oxford
ment of social physique anxiety. Journal of Sport and Exercise University Press.
Psychology, 11, 94–104. Lindwall, M., & Lindgren, E. (2005). The effects of a 6-month
Harter, S. (2012). Self-perception profile for adolescents: Manual exercise intervention programme on physical self-perceptions
and questionnaires. Denver, CO: University of Denver. and social physique anxiety in non-physically active adoles-
Harter, S. (2012). Self-perception profile for children: Manual cent Swedish girls. Psychology of Sport and Exercise, 6, 643–
and questionnaires (grades 3–8). Denver, CO: University of 658.
Denver. Liu, M., Wu, L., & Ming, Q. (2015). How does physical activity
Haugland, S., Wold, B., & Torsheim, T. (2003). Relieving the intervention improve self-esteem and self-concept in children
pressure? The role of physical activity in the relationship and adolescents? Evidence from a meta-analysis. PLoS One,
between school-related stress and adolescent health com- 10(8), https://doi.org/10.1371/journal.pone.0134804
plaints. Research Quarterly for Exercise and Sport, 74, 127– Lopera, C., da Silva, D., Bianchini, J., Locateli, J., Moreira, A.,
135. Dada, R., . . . & Junior, N.N. (2016). Effect of water- versus
Higgins, J.P., Altman, D.G., Gøtzsche, P.C., J€ uni, P., Moher, D., land-based exercise training as a component of a multidisci-
Oxman, A.D., . . . & Sterne, J.A. (2011). The Cochrane Collab- plinary intervention program for overweight and obese ado-
oration’s tool for assessing risk of bias in randomised trials. lescents. Physiology & Behavior, 165, 365–373.
BMJ, 343, d5928. Lubans, D., Morgan, P., Okely, A., Dewar, D., Collins, C., Bat-
Higgins, J.P., & Green, S. (2011). Cochrane handbook for sys- terham, M., . . . & Plotnikoff, R. (2012). Preventing obesity
tematic reviews of interventions 5.1. 0. The Cochrane Collabo- among adolescent girls: Outcomes of the nutrition and enjoy-
ration, 33-49. able activity for teen girls cluster randomized controlled trial.
Ho, F.K.W., Louie, L.H.T., Wong, W.-S., Chan, K.L., Tiwari, A., Journal of Science and Medicine in Sport, 15, S332.
Chow, C.B., . . . & Ip, P. (2017). A sports-based youth develop- Lubans, D., Richards, J., Hillman, C., Faulkner, G., Beau-
ment program, teen mental health, and physical fitness: An champ, M., Nilsson, M., . . . & Biddle, S. (2016). Physical activ-
RCT. Pediatrics, 140, e20171543. ity for cognitive and mental health in youth: a systematic
Hoare, E., Milton, K., Foster, C., & Allender, S. (2016). The asso- review of mechanisms. Pediatrics, 138, e20161642.
ciations between sedentary behaviour and mental health Lubans, D., Smith, J., Morgan, P., Beauchamp, M., Miller, A.,
among adolescents: A systematic review. International Jour- Lonsdale, C., . . . & Dally, K. (2016). Mediators of psychological
nal of Behavioral Nutrition and Physical Activity, 13, 108. well-being in adolescent boys. Journal of Adolescent Health,
H€
oner, O., & Demetriou, Y. (2014). Effects of a health- 58, 230–236.
promotion programme in sixth grade German students’ phys- Marker, A.M., Steele, R.G., & Noser, A.E. (2018). Physical activ-
ical education. European Journal of Sport Science, 14(sup1), ity and health-related quality of life in children and adoles-
S341–S351. cents: A systematic review and meta-analysis. Health
Jansen, W., van de Looij-Jansen, P.M., de Wilde, E.J., & Brug, Psychology, 37, 893–903.
J. (2008). Feeling fat rather than being fat may be associated Marsh, H.W. (1996). Physical self description questionnaire:
with psychological well-being in young Dutch adolescents. Stability and discriminant validity. Research Quarterly for
Journal of Adolescent Health, 42, 128–136. Exercise and Sport, 67, 249–264.
Kalak, N., Gerber, M., Kirov, R., Mikoteit, T., Yordanova, J., Martikainen, S., Pesonen, A.-K., Lahti, J., Heinonen, K., Feldt,
P€uhse, U., . . . & Brand, S. (2012). Daily morning running for 3 K., Pyh€ a, R., . . . & R€
al€ aikk€ onen, K. (2013). Higher levels of
weeks improved sleep and psychological functioning in physical activity are associated with lower hypothalamic-
healthy adolescents compared with controls. Journal of Ado- pituitary-adrenocortical axis reactivity to psychosocial stress
lescent Health, 51, 615–622. in children. The Journal of Clinical Endocrinology & Metabo-
Kendzierski, D., & DeCarlo, K.J. (1991). Physical activity enjoy- lism, 98, 619–627.
ment scale: Two validation studies. Journal of Sport & Exer- Martins, J., Marques, A., Sarmento, H., & Carreiro da Costa, F.
cise Psychology, 13, 50–64. (2015). Adolescents’ perspectives on the barriers and

© 2021 Association for Child and Adolescent Mental Health

doi:10.1111/camh.12485 Review of physical activity interventions and mental health 11

facilitators of physical activity: a systematic review of qualita- composition and self-concept of overweight and obese adoles-
tive studies. Health Education Research, 30, 742–755. cent males? A randomised controlled trial. British Journal of
McClure, A.C., Tanski, S.E., Kingsbury, J., Gerrard, M., & Sar- Sports Medicine, 48, 1482–1488.
gent, J.D. (2010). Characteristics associated with low self- Schutte, N.S., Malouff, J.M., Hall, L.E., Haggerty, D.J., Cooper,
esteem among US adolescents. Academic Pediatrics, 10, 238– J.T., Golden, C.J., & Dornheim, L. (1998). Development and
244. validation of a measure of emotional intelligence. Personality
Mental Health Foundation (2018). Statistics. https://www.me and Individual Differences, 25, 167–177.
ntalhealth.org.uk/statistics Shamseer, L., Moher, D., Clarke, M., Ghersi, D., Liberati, A.,
Merrell, K.W. (2011). Social emotional assets and resilience Petticrew, M. & PRISMA-P Group (2015). Preferred reporting
scales. Lutz, FL: PAR. items for systematic review and meta-analysis protocols
Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., (PRISMA-P) 2015: Elaboration and explanation. BMJ, 349,
Petticrew, M., . . . & PRISMA-P Group (2015). Preferred report- g7647.
ing items for systematic review and meta-analysis protocols Smith, J., Beauchamp, M., Faulkner, G., Morgan, P., Kennedy,
(PRISMA-P) 2015 statement. Systematic Reviews, 4, 1. S., & Lubans, D. (2018). Intervention effects and mediators of
Moksnes, U.K., & Reidunsdatter, R.J. (2019). Self-esteem and well-being in a school-based physical activity program for
mental health in adolescents–Level and stability during a adolescents: The ‘Resistance Training for Teens’ cluster RCT.
school year. Norsk Epidemiologi, 28, 1–2. Mental Health and Physical Activity, 15, 88–94.
Morgan, P.J., Saunders, K.L., & Lubans, D.R. (2012). Improving Smith, J.J., Eather, N., Morgan, P.J., Plotnikoff, R.C., Faigen-
physical self-perception in adolescent boys from disadvan- baum, A.D., & Lubans, D.R. (2014). The health benefits of
taged schools: Psychological outcomes from the Physical muscular fitness for children and adolescents: A systematic
Activity Leaders randomized controlled trial. Pediatric Obe- review and meta-analysis. Sports Medicine, 44, 1209–1223.
sity, 7, e27–e32. Sonstroem, R.J., & Morgan, W.P. (1989). Exercise and self-
National Heart, Lung, and Blood Institute. (2014). Quality esteem: Rationale and model. Medicine & Science in Sports &
assessment tool for before-after (pre-post) studies with no con- Exercise, 21, 329.
trol group. Systematic Evidence Reviews and Clinical Practice Spruit, A., Assink, M., van Vugt, E., van der Put, C., & Stams,
Guidelines. Washington, DC: National Institutes of Health. G.J. (2016). The effects of physical activity interventions on
Neumark-Sztainer, D.R., Friend, S.E., Flattum, C.F., Hannan, psychosocial outcomes in adolescents: A meta-analytic
P.J., Story, M.T., Bauer, K.W., . . . & Petrich, C.A. (2010). New review. Clinical Psychology Review, 45, 56–71.
moves—preventing weight-related problems in adolescent Sullivan, G.M., & Feinn, R. (2012). Using effect size—or why the
girls: a group-randomized study. American Journal of Preven- P value is not enough. Journal of Graduate Medical Education,
tive Medicine, 39, 421–432. 4, 279–282.
Popay, J., Roberts, H., Sowden, A., Petticrew, M., Arai, L., Rod- Tennant, R., Hiller, L., Fishwick, R., Platt, S., Joseph, S., Weich,
gers, M., . . . & Duffy, S. (2006). Guidance on the conduct of S., . . . & Stewart-Brown, S. (2007). The Warwick-Edinburgh
narrative synthesis in systematic reviews. A product from the Mental Well-Being Scale (WEMWBS): Development and UK
ESRC methods programme Version, 1, b92. validation. Health and Quality of Life Outcomes, 5, 1–13
Poznanski, E.O., & Mokros, H.B. (1996). Children’s depression Terry, P.C., & Lane, A.M. (2003). User guide for the Brunel mood
rating scale, revised (CDRS-R). Los Angeles, CA: Western Psy- scale (BRUMS). University of Southern Queensland, Aus-
chological Services. tralia, Toowoomba and University of Wolverhampton, Wolver-
Rampazzo, L., Mirandola, M., Davis, R.J., Carbone, S., Mocanu, hampton, UK.
A., & Campion, J. (2015). Mental health and schools: Situation The Whoqol Group (1998). The World Health Organization qual-
analysis and recommendations for action. EC Europa. ity of life assessment (WHOQOL): Development and general
Rodriguez-Ayllon, M., Cadenas-Sanchez, C., Est evez-L
opez, F., psychometric properties. Social Science & Medicine, 46,
Mu~ noz, N.E., Mora-Gonzalez, J., Migueles, J.H., . . . & 1569–1585.
Esteban-Cornejo, I. (2019). Role of physical activity and Tomas, J.M., Oliver, A., Hontangas, P.M., Sancho, P., & Gali-
sedentary behavior in the mental health of preschoolers, chil- ana, L. (2015). Method effects and gender invariance of the
dren and adolescents: A systematic review and meta- rosenberg self-esteem scale: A Study on Adolescents. Acta De
analysis. Sports Medicine, 49, 1383–1410. Investigaci o n Psicol
o gica, 5, 2194–2203.
Roh, H.T., Cho, S.Y., & So, W.Y. (2018). Taekwondo training Torales, J., O’Higgins, M., Castaldelli-Maia, J.M., & Ventriglio,
improves mood and sociability in children from multicultural A. (2020). The outbreak of COVID-19 coronavirus and its
families in South Korea: A randomized controlled pilot study. impact on global mental health. International Journal of Social
International Journal of Environmental Research and Public Psychiatry, 66, 317–320.
Health, 15, 757. Toulabi, T., Nikoo, M.K.N., Amini, F., Nazari, H., & Mardani, M.
Rosenberg, M. (1965). Rosenberg self-esteem scale (RSE). (2012). The influence of a behavior modification interven-
Acceptance and commitment therapy. Measures package, tional program on body mass index in obese adolescents.
61, 18. Journal of the Formosan Medical Association, 111, 153–159.
Rush, A.J., Trivedi, M.H., Ibrahim, H.M., Carmody, T.J., Arnow, Varni, J.W., Seid, M., & Rode, C.A. (1999). The PedsQLTM: Mea-
B., Klein, D.N., . . . & Keller, M.B. (2003). The 16-Item Quick surement model for the pediatric quality of life inventory.
Inventory of Depressive Symptomatology (QIDS), clinician Medical Care, , 37, 126–139.
rating (QIDS-C), and self-report (QIDS-SR): A psychometric Vella, J. (1995). Training through Dialogue. Promoting Effective
evaluation in patients with chronic major depression. Biologi- Learning and Change with Adults. The Jossey-Bass Higher
cal Psychiatry, 54, 573–583. and Adult Education Series. San Francisco, CA: Jossey-Bass.
Ryan, R.M., & Deci, E.L. (2000). Self-determination theory and Viera, A.J., & Garrett, J.M. (2005). Understanding interobserver
the facilitation of intrinsic motivation, social development, agreement: The kappa statistic. Family Medicine, 37, 360–
and well-being. American Psychologist, 55, 68–78. 363.
Sallis, J.F., & Owen, N. (2002). Ecological models of health von Hippel, P.T., Powell, B., Downey, D.B., & Rowland, N.J.
behaviour. In I.K. Glanz, B.K. Rimer & K. Viswanath (Eds.), (2007). The effect of school on overweight in childhood: gain
Health behavior and health education: Theory, research and in body mass index during the school year and during sum-
practice (pp. 462–484). Hoboken, NJ: Jossey Bass Publish- mer vacation. American Journal of Public Health, 97, 696–
ers. 702.
Schneider, M., Dunton, G., & Cooper, D. (2008). Physical activ- WHO. (2020). Coronavirus disease (COVID-19) – World Health
ity and physical self-concept among sedentary adolescent Organization. Retrieved 9 May 2020, from https://www.who.
females: An intervention study. Psychology of Sport and Exer- int/emergencies/diseases/novel-coronavirus-2019.
cise, 9, 1–14. Wilson, D.B. (2013). Practical meta-analysis effect size calcula-
Schranz, N., Tomkinson, G., Parletta, N., Petkov, J., & Olds, T. tor. https://campbellcollaboration.org/escalc/html/Effec
(2013). Can resistance training change the strength, body tSizeCalculator-Home.php

© 2021 Association for Child and Adolescent Mental Health

12 Gabrielle E. Hale et al. Child Adolesc Ment Health 2021; *(*): **–**

World Health Organisation. (no date). Adolescent development. Yu, X.-N., Lau, J.T.F., Mak, W.W.S., Zhang, J., Lui, W.W.S., &
https://www.who.int/maternal_child_adolescent/topics/ad Zhang, J. (2011). Factor structure and psychometric proper-
olescence/development/en/ ties of the Connor-Davidson Resilience Scale among Chinese
World Health Organisation (2009). Interventions on diet and adolescents. Comprehensive Psychiatry, 52, 218–224.
physical activity: What works summary report. Zigmond, A.S., & Snaith, R.P. (1983). The hospital anxiety and
World Health Organisation. (2018, February 23). Physical activ- depression scale. Acta Psychiatrica Scandinavica, 67, 361–
ity. https://www.who.int/news-room/fact-sheets/detail/ 370.
physical-activity.
Yeun, E.J., & Shin-Park, K.K. (2006). Verification of the profile
of mood states-brief: Cross-cultural analysis. Journal of Clini- Accepted for publication: 14 May 2021
cal Psychology, 62, 1173–1180.

© 2021 Association for Child and Adolescent Mental Health