Table 3

Characteristics of included studies

Published literature
Authors Country	Numbers of participants	Participant description	Intervention/comparison	Well-being outcomes and measures used Measurement times	Study design	Limitations (risk of bias)
Akandere and Demir45 Turkey	n=120	Gender: 50% female Age: 20–24 years Ethnicity: NR	Dance training intervention 90+20 min warm-up and cool down 3x week for 12 weeks Comparison: no intervention	Depression (Beck Depression Scale) Before and after 12 week dance intervention	RCT	Only one measure used Small population Sample already had dance knowledge Participant details not clearly reported Baseline levels of depression differ in groups
Amorose et al 52 USA	n=93	Gender: female Age: 13–18 years (M=15.78 years) Ethnicity: mostly Caucasian (90.6%) AI: members of a competitive club volleyball programme in Midwestern United States	Followed a cohort of female adolescent volleyball players through a season of competitive volleyball games. Approximately 4 months Comparison: time (before vs after)	Need satisfaction Sport competence (5-item subscale of the Intrinsic Motivation Inventory) Need for autonomy (6-item scale: Hollembeak and Amorose 2005) Need for relatedeness (10-item Richer and Vallerand’s Feelings of Relatedeness Scale) Well-being Self-esteem (10-item Rosenberg’s Self-Esteem Scale) Burnout (15-item Athlete Burnout Questionnaire) 1–2 weeks before competitive season starts and postseason (1–2 weeks before the last official game/~4 m after start of season)	Cohort	Sample bias: one club in Western USA, one sport. All females. Mostly Caucasian Selection bias: only those that agreed to volunteer. Dropout not reported Study design: no control group. Only 2 time points looked at Did not assess social contextual factors, eg, coaching behaviour
Kanojia et al 46 India	n=50	Gender: female Age: 18–20 years Ethnicity: NR AI: study conducted in the Department of Physiology, Lady Hardinge Medical College and Smt. Sucheta Kriplani Hospital, New Delhi, India	Yoga 35–40 min 6x week for the duration of three menstrual cycles Comparison: no intervention	Anger (16-item questionnaire) Trait anxiety (40-item questionnaire) Depression (10-item questionnaire) Subjective well-being (50-item questionnaire) Questionnaires were developed by the Defense Institute of Physiology and Allied Sciences At the beginning and after completion of three menstrual cycles	RCT	Dropout not reported Recruitment methods not reported Not possible to double blind Consistent findings
Kim and Kim47 Korea	n=277	Gender: 48% female Age: 17–22 years (M=20.6 years) Ethnicity: NR AI: Korean high school (n=45) and undergraduate students (n=232) volunteers	One of four exercise sessions: aerobic exercise, body conditioning, hip-hop dancing and ice skating One-off 40 min session+10 min warm-up and cool down	Mood (Subjective Exercise Experiences Scale: measuring three dimensions; positive well-being, psychological distress and fatigue) Before and after the exercise session	RCT	Data based on one session only
Li et al 48 China	n=222	Gender: 82.5% female Age: 18–25 years (M=20.78 years) Ethnicity: NR AI: college students recruited from college in China	Baduanjin exercise 1 hour 5x week for 12 weeks Comparison: usual exercise	Self-esteem (Self-esteem Scale (SES)) Mood/mindfulness (Profile of Mood States (POMS) scale) QoL (WHOQOL-BREF) Stress (Chinese Perceived Stress Scale) Self-symptom intensity (SCL-90 scale) Baseline (before start), at the end of the intervention (week 13), 12-week follow-up (week 25)	RCT	Not blinded Participants recruited from one medical university Greater proportion of female participants Small effect size Excellent protocol adherence No significant loss to follow-up
Lindgren et al 49 Sweden	n=110	Gender: female Age: 13–19 years (average=15.3) Ethnicity: NR AI: physically inactive students from secondary schools in low socioeconomic status areas	Empowerment-based exercise intervention programme 45 min moderate exercise+15 min discussion (topics such as healthy lifestyles were addressed) 2x week for 6 months Comparison: waiting list	Self-efficacy (Swedish version of a 10-item General Self-efficacy Scale) Behaviour changes (Social Barriers to Exercise Self-efficacy Questionnaire) Once at the start of the programme and once at end (6 months)	RCT	Small sample size High dropout rate
Noggle et al 50 USA	n=51	Gender: 61% female in Yoga group, 47% female in control Age: average age 17 years (grades 11 and 12) Ethnicity: 92.2% white, 3.5% Hispanic, 2.1% African-American, 1.4% multirace and 0.8% Asian AI: students at a public high school in rural western Massachusetts.	A Kripalu-based yoga programme of physical postures, breathing exercises, relaxation and meditation 30 min 2–3x week for 10 weeks (28 yoga sessions total) Comparison: physical education as usual 30–40 min 2–3x week for 10 weeks	Mood (POMS-Short Form) Affect (Positive and Negative Affect Schedule for Children) Stress (Perceived Stress Scale) Positive psychology (Inventory of Positive Psychological Attitudes) Resilience (Resilience Scale) Anger (State Trait Anger Expression Inventory-2TM) Mindfulness (Child Acceptance Mindfulness Measure) One week before and after	RCT	Small sample size. Would have been ideal to randomise individually but being in a school setting required allocation at the classroom level Moderate attendance at the yoga classes
Staiano et al 51 USA	n=54	Gender: 55.6% female Age: 15–19 years Ethnicity: African-American AI: overweight and obese students from an urban public high school	Exergame (EG) intervention—students encouraged to play the Nintendo Wii Active game. Two EG groups: cooperative EG worked with a peer to expend calories and earn points together; competitive EG participants competed against a peer 30–60 min per school day in a lunch-time or after-school programme for 20 weeks Comparison: regular daily activities	Self-efficacy (Exercise Confidence Survey) Self-esteem (Rosenberg Self-Esteem Scale) Peer support (Friendship Quality Questionnaire) Baseline, T2 (10 weeks), T3 (20 weeks)	RCT	Sample bias: small sample from one school and some attrition