Association of traditional leisure time activities during adolescence to later psychiatric morbidity in early adulthood – a systematic review
Article Category: Review
Published Online: Aug 10, 2025
Page range: 82 - 102
DOI: https://doi.org/10.2478/sjcapp-2025-0009
Keywords
© 2025 Johanna Timonen et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The association of leisure activities to the psycho-social well-being of children and adolescents is recognized in the research (1, 2). Leisure time activities have shown to enhance social interaction and the mental health of the individuals involved (3). For example, participation in sport and recreation activities are reported to be associated with better self-esteem and increased pro-social behavior (4). Various sports activities have been documented to reduce contemporary stress, anxiety, and depression (e.g., 5, 6, 7). Various music (8) and arts (9) related hobbies have also been reported as having a positive impact on the current psycho-social well-being of children and adolescents.
Leisure time activities in childhood and adolescence have also been reported to have far reaching impacts on psycho-social health along the life span. They have been shown to increase psychological well-being (10, 11) and reduce psychological symptoms in adulthood (12, 13). A longitudinal population-based study from Norway (14) showed that active free time during adolescence was related to higher life satisfaction and better mental well-being in adulthood, in comparison to those having inactive free time in adolescence. Conversely, Jacka et al. (15) reported that low physical activity under the age of 15 was associated with an increased risk of depression symptoms in adulthood, from the age 20 until the age of 97 years.
Most research on the relationship of leisure time activity during adolescence with later mental health has primarily focused on psychological symptoms or overall mental wellbeing. There is, however, a shortage of studies examining the subject from the standpoint of recognized mental illnesses. Considering the well-documented beneficial psychological impacts of recreational activities on mental health in adolescence, it is reasonable to speculate that this relationship extends to the emergence of psychiatric disorders later in life. To the best of our knowledge, there are no previous systematic reviews that summarize research-based data on the connections between different types of hobbies or leisure activities in youth and the ensuing emergence of distinct psychiatric disorders. This information could be applied to youth-focused interventions aimed at preventing mental illness, with activities focused on leisure activities. With the recent expansion in the amount of time spent on hobbies, as well as the availability and choice of interests, over the last several decades, the importance of leisure activities in society is increasingly recognized (16, 17, 18).
Traditional leisure time activity includes a wide range of activities, such as sports, culture, art, music, travel and social life, which have been passed down from generation to generation and are part of a society’s cultural heritage. (19, 20) “Articles about screen time and digital media consumption were beyond the scope of our review because it was not possible to reliably determine whether the screen time or social media use discussed in a specific study represented a leisure time hobby or a school- or work-related activity, or if it was associated with problematic social-media behavior (21).
This systematic review summarizes characteristics and findings of research examining the influence of several traditional recreational activities throughout youth on subsequent psychiatric problems up until early adulthood.
This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (22). The quality of the studies was assessed according to the modified STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) checklist (23).
Our review included previous studies that analyzed the association of traditional leisure time activities during adolescence to psychiatric morbidity until early adulthood, up to the age of 44 years. The age limits for early adulthood were set at 18–44 years, as defined by Lindemann (24). The articles were filtered by the year of publication (2010–2023). Only articles written in English were accepted for review. The articles were considered to be relevant if they met the following inclusion criteria: (1) study design: longitudinal, cohort study, follow-up study, prospective study, original study; (2) aims: association of leisure time activities in adolescence to later psychiatric morbidity (3) age group: from adolescence to early adulthood. In this study, adolescence was defined as being aged between 13 and 18 years. In Finland adolescent psychiatry services (own psychiatric specialty) starts at age of 13. The upper age limit was chosen because the legal definition of adulthood in most countries is deemed to start at the age of 18 (25).
Our exclusion criteria were as follows: systematic reviews, meta-analyses, and cross-sectional studies. Studies of digital leisure activities and screen time during adolescence were excluded from this review, this was because these types of hobbies form their own entity among all leisure activities and are also difficult to define in a uniform manner due to their complexity, diversity, and rapid evolution (21).
The articles for this review were obtained from the electronic literature databases, primarily on 24th May 2022. An additional literature search was performed on 3rd of November 2023. The literature databases used in our searches were PubMed, Scopus, and EBSCO (including databases CINAHL, Academic Search Ultimate and APA PsycArticles).
Electronic searches were supplemented with manual screening of the reference lists of articles identified, including reference lists of systematic reviews found in searches. Table 1. shows a list of the search terms used for each tier and details of how they were combined.
Search terms combined for each tier
| Tier 1 |
Tier 2 |
Tier 3 |
Tier 4 |
|---|---|---|---|
| “Leisure Activities”[Mesh] OR “leisure activit*”[tw] OR “leisure time”[tw] OR hobby*[tw] OR hobbies[tw] OR sport*[tw] | “Adolescent”[Mesh] OR adolescen*[tw] OR teenage*[tw] OR youth [tw] | “Mental Disorders”[Mesh] OR “mental health disorder*”[tw] OR “mental disorder*”[tw] OR “Psychiatric Disorder*”[tw] OR anxiet*[tw] OR depress*[tw] OR “substance use disorder*”[tw] | “Cohort Studies”[Mesh] OR “Follow-Up Studies”[Mesh] OR “Prospective Studies”[Mesh] OR follow-up*[tw] OR followup[tw] OR “prospective stud*”[tw] OR cohort*[tw] |
The results of the systematic literature searches were examined by screening the title and abstract of each article identified by Covidence. It is a cloud-based tool with the ability to perform title/abstract screening, full-text screening, data abstraction, and quality assessment (
The following information was extracted from each article and summarized in Table 2.
Study characteristics and measures for leisure-time activity in childhood and adolescence and assessment of later psychiatric morbidity up to young adulthood.
| Study-characteristics | Leisure time activity in childhood and adolescence | Later psychiatric morbidity | |||||
|---|---|---|---|---|---|---|---|
| 1. Authors, publishing year, country | 2. The origin of the study material; study design | 3. Number of study participants; study groups, gender | 4. Data collection period, assessment age/school grades for leisure time activity | 5. Instrument for leisure time activity; type of questions | 6. Data collection period, psychiatric diagnoses, and age at assessment | 7. Diagnostic criteria, source or information and treatment setting (if reported) | 8. Confounders |
| Desphande et al. (2020) USA | The National Longitudinal Study of Adolescent to Adult Health (Add Health) study; nationally representative sample of American adolescents | total n=2197, all men
* n=521 adolescents, who participated or intended to participate school football * n=1676 controls, who did not play or intended to play school football, including 610 sport controls (non-collision sport) and 1066 non-sport controls (played no school sport) |
1994–1995, at school grades 7–12 | Questionnaire for leisure activities; participation in base-/softball, basketball, football, swimming, tennis, track, or volleyball) | 2008, depression, anxiety, post-traumatic disorder at age 24–32 years. In addition, single diagnosis of nicotine, cannabis, alcohol dependence/abuse | Self-reported, the Center for Epidemiological Studies Depression scale (CES-D) | subjects' health, behavior, educational experience, family background, and family health history |
| Easterlin et al. (2019) USA | The National Longitudinal Study of Adolescent to Adult Health (Add Health study) | total n=4888 having exposure to ACEs (physical and sexual abuse, emotional neglect, parental alcohol misuse and incarceration, living with a single parent) | 1994–1995, at school grades 7–12 | Questionnaire for leisure activities; one question for participation or intention to participate in any of the following team sports (base-/softball, basketball, cheerleading/dance team, field hockey, football, ice hockey, soccer, swimming, tennis, trac, volleyball, wrestling and other sport. | 2008, depression, anxiety, at age 24–32 years | Self-reported, Positive response to question “having ever received a diagnosis of depression or anxiety) and having positive screen result for current depression assessed with the 10-item subscale of the CES-D-scale. | subjects' health, behavior, educational experience, family background, and family health history |
| Oggenhaug et al. (2015) Norway | Longitudinal population-based case-cohort study in the North-Troendelag region of Norway (the Young-HUNT1 study) |
* 15 adolescents with later diagnosis of schizophrenia (cases) * 18 adolescents with later diagnosis of bipolar disorder (contrast) *120 healthy adolescents matched on age and gender for cases |
1995–1997 (baseline), at age of 13–19 years (mean 16 years) | Young-HUNT1 questionnaire; several questions inquiring active involvement in sports; exercise hours/days per week); participations to team sports (soccer, hand-, basket- and volleyball) and to individual sports (sports other than team sports) | After baseline (exact years not reported), Schizophrenia, bipolar disorder, onset ages not reported | ICD-10, psychiatric diagnoses were identified through the patient administrative system | not mentioned, but discussed in discussion part |
| Koivukangas et al. (2010) Finland | Northern Finland Birth Cohort 1986 study | n=6987 (3367 boys, 3629 girls) | 2001–2002, at the age 15–16 years | Postal questionnaire including one question about how many hours a week, outside of school hours, spent on brisk physical activity (i.e. causing some sweating and shortness of breath). Responses were analyzed in categorized into three groups: inactive (not at all or about half an hour a week), somewhat active (about 1 h to 2–3 h a week), active (at least about 3–6 h or more a week). | 2002–2005, nonorganic psychosis (excluding persons who had developed psychosis before 2002) | ICD-10, psychiatric diagnosis from the Finnish Hospital Discharge Register and antipsychotic reimbursements marked on the Finnish Social Insurance Institute's (FSII) register. Inpatient, outpatient | family's socio-economic status and parent ′s physical activity |
| Sormunen et al. (2017) Finland | Follow up study for Cardiovascular Risk of Young Finns; population-based cohort study | n = 3596; of whom n=3325 controls with no psychiatric diagnoses during follow-up | 1980,1983, 1986, at age 9–18 years | Questionnaire for physical activity assessing engagement in leisure-time physical activities using several questions. ZZ | During follow-up period of 1980–2012, non-affective psychosis, including a subgroup for schizophrenia |
(ICD-8/9/10 psychiatric diagnoses from the Finnish Hospital Discharge Register converted to DSM-IV diagnoses. Inpatient, outpatient |
parents' mental disorders and parent′s physical activity |
| Samek et al. (2015) USA | Sibling Interaction and Behavior Study (SIBS); Longitudinal study | n=1226, (555 men, 671 women) | 1994–1995, after completion of high school usually at age 19 years | The Social Adjustment Interview asking about involvement in extracurricular actives (yes, no) in high school: team sport (varsity, intramural), student government, debate/forensics, clubs, plays/musicals, newspaper. | Exact years not reported, conduct disorder before age of 15 years, adult antisocial personality disorder symptoms (yes, no) in the 12 months preceding the assessment at median age 22.4 years | Structured Clinical Interview for DSM-III-R for Personality Disorders, modified for DSM-IV | parent′s education and marital status |
| Timonen et al. (2021) Finland | The Northern Finland birth cohort 1986 | n=6838 (3298 males, 3540 females) | 2001–2002, at age 15–16 years | Postal questionnaire designed for a cohort study: several questions about the intensity and level of participation in leisure-time activities. Based on these variables, three social activity levels (high, middle, moderate) were formed and analyzed. XX | Up to end of 2018, any major groups of psychiatric disorders (excluding those who had been diagnosed before age 16 years) | ICD-10 psychiatric diagnoses from the Care Register for Health Care (CRHC) Inpatient, outpatient | parents living together, parental educational level, parents' mental health disorders |
| Iverson et al. (2022) USA | The National Longitudinal Study of Adolescent to Adult Health (Add Health) |
n=1762, all men * n=369, play or intend to play high school football * n=952, not played or indented to play football * (n=441 missing information on football playing) |
1994–1995 (Wave I), in school grades 7–12, at median age 15 years | Questionnaire about leisure activities; questions about participating in clubs, organizations, and teams, specifically, in playing football. | 2016–2018 (Wave V follow-up), Depression or anxiety/panic disorder on average at 38 years of age | Self-reported; a question about whether a doctor, nurse of other health care provider ever had told the participant to have or have had depression or anxiety/panic disorder. | subjects' health, behavior, educational experience, family background, and family health history |
| Timonen et al. (2023) Finland | the Northern Finland birth cohort 1986 | n=6709 (3227 males, 3482 females) | 2001–2002, at age 15–16 years | Postal questionnaire; designed for a cohort study: several questions about the intensity and level of participation in leisure-time activities. Based on these variables, three social activity levels (high, middle, moderate) were formed and analyzed. | Up to end of 2018, any and major groups of psychiatric disorders up to age 16–33 years (excluding those who had been diagnosed before the age of 16) |
ICD-10 psychiatric diagnoses from the Care Register for Health Care (CRHC) Inpatient, outpatient |
parents living together, parental educational level, parents' mental health disorders |
| da Silva et al. (2023) Brazil | The 1993 Pelotas Birth Cohort, Brazil | n=3247 (1514 men, 1733 women) | Exact years not reported, at age of 11, 15, and 18 years | The International Physical Activity Questionnaire assessing duration and weekly frequency of recreational moderate- and vigorous-intensity activities. VV | Major depressive disorder, generalized anxiety disorder, at age of 22 yrs | Mini International Neuropsychiatric Interview (MINI) based on DSM-IV and ICD-10 diagnostic criteria, and applied to follow the DSM-V criteria | sex (male and female), skin color (white, black, brown indigenous), family income at birth (in minimum maternal schooling at birth (in complete years). |
| McKercher et al. (2014) Australia | The Childhood Determinants of Adult Health Study, National Longitudinal Study | n=1630 (759 males, 871 females) from 109 schools | 1985, 2004–2006, at age 9–15 years | Questionnaire from the Australian Schools Health and Fitness Survey inquiring about past-week frequency and duration of school and extracurricular sport and exercise. YY | 2004–2006, depression (major depressive and dysthymic disorder) at age 27–36 years | Depression module of the Composite International Diagnostic Interview (CIDI-Auto), | status, education level, main source of income, current occupation, smoking status, and number of live births |
Notes. XX The study categorized leisure time activities based on intensity and level of social activity. Active and non-active activities were identified, and social activity levels were classified as high, middle, or low. High social leisure activity (SLA) included community or sports club participation, middle SLA involved activities considering others but not belonging to a community or sports club, and low SLA comprised activities done alone. Participants were assigned to one of these three mutually exclusive groups. Exclusions were made for those with no hobbies or unanswered questions, resulting in a final sample for analysis.
ZZ Leisure time activities were assessed through questions about common activities, frequency of physical activity, intensity, participation in organized training, and involvement in sports competitions. Participants reported their usual leisure time activities, ranging from indoor reading to outdoor exercise and socializing. Physical activity was frequency gauged by how often individuals engaged in activities for at least half an hour. The intensity of physical activity was measured by breathlessness and sweating levels. Participants also reported the frequency of organized training participation and whether they engaged in sports competitions.
VV Variables describing sufficient level of physical activity (PA) based on WHO guidelines (at least 300 min per week at 11 and 15 years, at least 150 min per week at 18 years) at each age phase. Data was categorized into four PA levels: never sufficiently active at any age period, sufficiently active at one age period, sufficiently active in two age periods, and always sufficiently active. Abbreviations: CI, confidence interval; LTPA, leisure-time physical activity; PR, prevalence ratio; Total PA, total physical activity (leisure time + commuting physical activity).
YY The International Physical Activity Questionnaire (IPAQ) was used to retrospectively re-assess leisure time physical activity (PA) at the age of 15 years and to assess leisure-time PA in adulthood. Four levels of leisure time PA pattern were categorized: persistently inactive, decreasing, increasing, and persistently active.
Characteristics of articles: Author(s), year of publication, country
The origin of the study material; study design
Number of study participants; comparison groups, gender
Data collection period, assessment age/school grades for leisure time activity
Instrument for leisure time activity; type of questions
Data collection period, psychiatric diagnoses, and age at assessment
Diagnostic criteria, Source or information and treatment setting (if reported)
Table 3 summarizes the aims, results, and overall conclusions, which were related to the association of leisure time activities in childhood and adolescence to later diagnosis of psychiatric disorders, assessed up to young adulthood.
Aims and the findings of the reviewed studies regarding the association of leisure time activity in childhood and adolescence to later psychiatric disorders diagnosed by young adulthood.
| Authors, publishing year (country) | Aims of the study addressing leisure time activity in childhood and adolescence and later psychiatric disorders | Detailed results | Overall conclusion based on the results |
|---|---|---|---|
| Desphande et al. (2020) USA | To estimate the effect of playing football during adolescence (at grades 7–12) on diagnosis after age 18 years for depression, anxiety or panic disorders, and PTSD in early adulthood (age 24–32 years). Secondary outcomes for nicotine, cannabis or alcohol abuse were examined. | Study did not find evidence that participation in middle or high school football had a harmful effect on depression or anxiety and substance dependence in early adulthood. | Participating or intending to participate in school football does not appear to be a major risk factor for early adulthood depression, anxiety, or PTSD. |
| Easterlin et al. (2019) USA | To examine among those affected by at least one ACEs, the association of team sports participation during adolescence (at school grades 7–12) to ever having diagnosed for depression or anxiety and to have positive screen results for current clinical depression in adulthood (age 24–32 years). | In total sample, team sports participation in adolescence (compared to non-participants) associated to a decreased likelihood for an adult diagnosis of depression (adj. OR with 95% CI = 0.76, 0.59–0.97) and anxiety (0.70, 0.56–0.89.) | Considering that participation in team sports has been linked to improved adult mental health for individuals with adverse childhood experiences (ACEs), pediatricians may consider recommending team sports involvement for patients with ACEs. Additionally, parents may contemplate enrolling their children with ACEs in team sports. |
| In boys, teams sport participation was associated with a decreased likelihood of having a diagnosis of depression (0.67, 0.46–0.99) or anxiety (0.66, 0.45–0.96), and for showing current depressive symptoms (0.75, 0.56–0.99), while in girls the association was found for diagnosis of anxiety (0.73, 0.54–0.98). | |||
| Okkenhaug et al. (2015) Norway | To investigate the prevalence and pattern of physical activity in adolescence (age 13–19 years) with later diagnosis for schizophrenia (cases) or bipolar disorder (contrasts), and in mentally healthy controls. | The cases with schizophrenia and contrasts with bipolar disorders, compared to healthy controls, reported less physical activity regarding exercise hours being less than one hour per week (53.3%, 61.1%, 29.1%; p < 0.05). | Those who develop schizophrenia might have started to develop a hazardous lifestyle already in their teens. It is important in a preventive perspective to gain more knowledge concerning the relationship between health-related habits and later development of psychiatric morbidity. |
| Participating in team sport was lower in cases with schizophrenia than in controls (53.3%. 79.2%; p=0.048). | |||
| Koivukangas et al. (2010) Finland | To explore the association of the level of physical activity after school hours (age 15–16 years) to later onset of first-episode psychosis | Individuals who later developed psychosis were more likely to be physically inactive during adolescence, adj. OR, (95% CI), 3.3 (1.4–7.9) compared to those without psychosis. | Encouraging physical activity is crucial for individuals at risk of developing psychosis to mitigate the negative impact of physical inactivity on overall health. Further research is recommended to explore the effectiveness of physical activity in treating mental health issues in adolescents. |
| Sormunen et al. (2017) Finland | To examine the association of physical activity levels in childhood and adolescence (age ≤ 18 years) to later development of non-affective psychosis and schizophrenia. | Lower physical activity index (relative risk, RR, 95%CI) (1.26, 1.1–1.5), lower level of common activity during leisure time (1.71, 1.2–2.5), and non-participation in sports competitions (2.58, 1.3–5.3) were associated with higher risk for later non-affective psychosis, compared to controls without no psychiatric diagnoses. | Low physical activity level in children and adolescents is an independent predictor for development of non-affective psychosis. Further research is needed to assess the role and possibilities of early exercise and physical activity intervention as a part of psychosis prevention. |
| Correspondingly, lower physical activity index (1.4, 1.2–1.7), lower common activity during leisure time (1.8, 1.0–3.0), lower intensity of physical activity (1.7, 1.1–2.9), lower frequency of participation in organized training (1.40, 1.1–1.8) and non-participation in sports competitions (4.88, 1.4–17.0) were associated with higher risk for later development of schizophrenia. | |||
| Samek et al. (2015) USA | To analyze whether involvement in high school sports and other extracurricular actives (at age 19 years) associated to adult antisocial behavior (AAB) (at mean age 22.4 years) and has impact on the association between conduct disorder (CD) before 15 years and AAB in young adulthood. | Fewer AAB symptoms (mean) were associated to involvement in in clubs (yes vs. no, 0.61 vs. 0.90, p<0.001) and plays/musicals, 0.57 vs. 0.85, p<0.01) compared with those not involved. | It appears that high school sports involvement may decrease risk for persistent antisocial behavior. It is crucial for future research to continue to identify additional environmental factors that mitigate the relationship between conduct disorder and adult antisocial behavior. |
| The association between CD before 15 yrs and AAB in young adulthood was significantly weaker (χ2=4.13, p=0.04) among those involved in high school sports (β=0.28; p <.001) compared with those not involved in sports (β =0.49; p<0.001). | |||
| Timonen et al. (2021) Finland | To analyze the association of different levels of social leisure time activity (SLA) during adolescence (ages 15–16 years) with the incidence of mental disorders up to young adulthood (the ages of 16–33 years). | Low SLA was associated with increased likelihood of any psychiatric disorder (OR 1.61, 95% CI 1.13–2.30), anxiety disorders (OR 2.07, 95% CI 1.37–3.15) and behavioral disorders (OR 4.12, 95% CI 1.90–8.96). Furthermore, high SLA was associated with decreased likelihood of any psychiatric disorders (OR 0.78, 95% CI 0.66–0.93), substance use disorders (OR 0.53, 95% CI 0.36–0.78), affective disorders (OR 0.72, 95% CI 0.57–0.91) as well and anxiety disorders (OR 0.79, 95% CI 0.63–0.99). Social leisure time activity was not associated with psychotic disorders of the study participants. | Socially active leisure time during adolescence is a protective factor against the development of later psychiatric disorders. Social support of the community in the early phase is one way to prevent development of mental health symptoms into manifest psychiatric disorders. In terms of prevention of mental disorders, these study findings encourage families, schools, and other communities to continue to enhance and develop community-based social leisure time activities for children and adolescents. |
| Iverson et al. (2022) USA | To examine an association of playing high school football (at age 15 years) to depression and anxiety or panic disorder by the age of 37–39 years | Football playing, compared to those who did not play football, was not significantly associated to lifetime history for depression or anxiety/panic disorder. | The results do not find an association between high school football exposure and depression and anxiety later in life. |
| da Silva et al. (2023) Brazil | To evaluate the association of leisure-time and total physical activity (PA) during adolescence (at 11, 15, and 18 years) to major depressive disorder and generalized anxiety disorder in young adulthood (at 22 years). | In crude analyses, all leisure-time and total PA variables assessed in adolescence were negatively associated with major depressive disorder in young adulthood, while corresponding association with generalized anxiety disorder was also present, except in leisure-time PA at 15 years and total PA. In adjusted analyses, active participants to the leisure-time and total PA at 11 years were less likely to have diagnosis for major depressive disorders at 22 years, prevalence ratios (95%CI) being 0.54 (0.11–0.89) and total PA 0.63 (0.41–0.97), while no significant associations was found to generalized anxiety disorder in any of PA variables. | There is a linear trend of leisure-time physical activity throughout adolescence and a lower risk of depression in early adulthood. There was evidence that early adolescence appears to be a sensitive period for the benefits of physical activity on depression in early adulthood. Study reinforces the need for policies to increase leisure-time physical activity levels from early adolescence aiming to decrease the prevalence of depression in youth. |
| The adjusted linear trend analyses showed that the higher level of leisure-time PA throughout adolescence was associated to the lower prevalence of major depressive disorders in young adulthood. | |||
| Mc Kercher et al. (2014) Australia | To examine the association of persistent leisure physical activity (PA) pattern from childhood (age 9–15 years) to adulthood to depression (major depressive or dysthymic disorder) in young adulthood (age 27–36 years). | Males who were increasingly or persistently active, as compared with persistently inactive, had lower risk for depression in adulthood, the relative risk (adj. RR, 95CI%) being 0.31 (0.11–0.92) and 0.35 (0.15–0.81), respectively. | Findings from both prospective and retrospective analyses indicate a beneficial effect of habitual discretionary physical activity since childhood on the risk of depression in young adulthood, given that physical activity is a potentially modifiable risk factor, the public health implications of these findings are substantial. |
| Timonen et al. (2023) Finland | To examine whether adolescence-related social leisure time activity (SLA), per se, is a mediating link in the association between adolescent psychological symptoms and later psychiatric morbidity. | Low SLA was associated with increased likelihood of anxiety disorder (OR 1.92 95% CI 1.25–2.95) behavioral disorder (OR 3.92 95% CI 1.70–9.08). High SLA was associated decreased likelihood of any psychiatric disorder (0.98–2.07) CI 0.68–0.96, affective disorders (OR 0.75 95% CI 0.60–0.95) and anxiety disorder (OR 0.83 95% CI 0.66–1.04). | Socially active leisure time during adolescence is related to better long-term mental health, while socially inactive leisure time associates with the likelihood of later psychiatric morbidity. To prevent psychiatric disorders, it is highly recommended that such leisure time activities are enhanced in society. |
The substantial heterogeneity of the studies regarding participant characteristics, study designs and study outcomes made a meaningful meta-analysis inappropriate.
The modified STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) checklist (23) was used to assess the quality of each article deemed eligible for inclusion in our systematic review (see Supplementary Table 1). Each item was scored from 0 to 1, where 0 = “no, the criteria are not met”, 1 = “yes, the criteria are met”, and a score of 0.5 means that the criterion for the item is partially met. The total quality sum-score of this modified STROBE checklist can vary from between 0 and 27. Two researchers (JT and MN) independently assessed the quality of each article. Possible differences in the quality classification were discussed and a consensus was made together with the whole research group.
Our article selection process, according to PRISMA guidelines, is illustrated in Figure 1. We used Covidence screening and data extraction tool for conducting systematic reviews. The initial literature search identified 1,432 articles. After removing duplicates, 1,095 articles underwent title and abstract screening. After full-text analysis, eight articles were deemed to qualify for the systematic review. An additional search in November 2023 identified 185 new articles, of which three met our inclusion criteria. Thus, 11 articles were included in our systematic review.
PRISMA diagram illustrating the literature search process used to find articles that recorded associations between leisure time activities during adolescence to later psychiatric morbidity in early adulthood.
As summarized in Table 2, a total of 11 articles were included in this systematic review; four from the USA (26, 27, 28, 29), four from Finland (30, 31, 32, 33) and one from Norway (34), Brazil (35) and Australia (36). Four cohort articles had utilized two different Finnish cohorts (30, 31, 32, 33) and one article was based on a Brazilian cohort study (35). Of the six longitudinal articles, three were Add health studies (26, 27, 28), and the remaining studies originated from the Young-Hunt study (34), Sibling Interaction and Behavior Study (29) and the Australian national study (36).
The size of the comparison groups in the reviewed articles varied from 15 to 6,987 participants. Comparison groups were included in three articles (27, 32, 34). In two of these articles’ controls were selected according to the health-related factors i.e., without psychiatric disorder or healthy controls (32, 34) and in one study according to leisure time activity i.e., non-participation in football at school (27).
In two articles baseline age was reported according to the school grade status of participants, i.e., 7–12 school grades corresponding approximately to 13–18 years old (27, 28). The remaining articles reported baseline age of study participants in years, which varied from 9 to 18 years.
The follow-up age for assessment of psychiatric disorders generally varied from between 20 – 44 years. In two articles from the Add-health study follow-up, the age for psychiatric morbidity was reported as 24–32 years of age (27, 28) and in one Add-health study as 34–44 years of age. Two Birth Cohort follow-up articles from Northern Finland included a period from 16 to 33 years of age (30,31) and, in another study from the same cohort, from 16 to 20 years of age (33). In the Brazilian cohort study, psychiatric disorders in the study participants were assessed at the age of 22 (34). In the Cardiovascular Risk of Young Finns cohort study, the follow-up age varied from 18 to 44 years (32) and in the Sibling Interaction and Behavior Study, the mean (sd) follow-up age for setting psychiatric disorders was 22.4 (sd 1.85) years (29). In the Young-Hunt-study (34) follow-up age was from 16 to 19 years. In the Australian national study (35) follow-up age for psychiatric morbidity was between 30–35 years. In nine of the articles the participants represented both female and male genders and in the two other articles, both focusing on American football players, all participants were male (26, 27). Diagnoses of psychiatric disorders were set before the age of 25 in seven (64%) studies, after the age of 25 in three (28%) studies and missing in one study.
The most common confounders were the subjects' health, behavior, educational experience, family background, and family health history. There was one study where confounders were not mentioned (34), but it was discussed in the discussion section.
The data collection periods of the reviewed articles covered the years from 1980 to 2007. Information about the type of leisure time activity was based on the questionnaires in all of the articles under review. Two articles focused purely on American football (26, 27). In one article, leisure time activities were not specified, but were stated in general terms: physical activity, school and extracurricular sport and exercise (35). The remaining eight articles analyzed participation in a wide variety of leisure time activities, including both social and solitary activities, such as basket-ball and dance, or swimming and reading.
In four articles, psychiatric diagnoses were extracted from registers according to ICD-10 classification (30, 31, 33, 34). In one study, primary registry-based ICD-diagnoses were converted to diagnoses according to DSM-IV classification (32) In one article, DSM-IV diagnoses were based on the structured Clinical Interview for DSM-III-R Personality Disorders modified for DSM-IV (29). In one article, ICD-10 diagnoses were based on the Mini International Neuropsychiatric Interview (MINI) (35). Diagnoses in another article were based on the use of the Composite International Diagnostic Interview, CIDI-Auto, version 2.1 (36). In three articles, psychiatric diagnoses were based on self-reports (26, 27, 28).
In the studies included, some assessed psychiatric disorders from major diagnostic groups, while others focused on single psychiatric disorders. In two articles all mental disorder diagnoses according to ICD-10 (all F-codes) were included and analyzed as a whole (30, 31). In two articles, diagnoses of affective disorders (F3-codes) were analyzed as a single group (30, 31) while in five other articles, depression was examined as a single and specific diagnosis of affective disorder (26, 27, 28, 35, 36) and another article focused only on bipolar disorder (34). Diagnoses for anxiety disorders (F4-codes) as a group was examined in five articles (26, 27, 28, 30, 31). Of specific anxiety disorders, generalized anxiety disorder was analyzed in one article (35) and panic disorder in another (26). Post-traumatic stress disorder was analyzed in one article (27).
Psychotic disorders as a diagnostic group (F2-codes) were examined in four articles (30, 31, 32, 33) Of the studies examining single psychotic disorders, one article focused on schizophrenia (34), two examined substance use disorders as a diagnostic group (F1-codes) (30,31) and another looked at nicotine, cannabis, or alcohol dependence/abuse (27). Two further articles included behavioral and emotional disorders as their diagnostic group (F9-codes) (30, 31). The personality disorder (F6-codes) antisocial behavior was examined in one other study (29).
As outlined in Table 2, two of the studies sought to analyze the association between the study participants’ different levels of social leisure time activity to mental disorders in general (30,31). In one study, special attention was paid to the curative effect of team sport activity among those with adverse childhood experiences (ACEs) on the participant’s mental health in adulthood (28). Five articles examined the association of team sport activity to depression (26, 27, 28, 35, 36). One article aimed to evaluate whether involvement in high school sports attenuated the association between conduct disorder during adolescence and subsequent antisocial behavior in adulthood (29). Another article studied the association of physical activity during adolescence to depression and anxiety disorders during young adulthood (35). In three articles, the aims of the study were to examine the association of physical activity to subsequent psychotic disorders (32, 33, 34).
In two articles, leisure time activity in adolescence involving a high amount of social interaction was reported to be associated significantly with lower levels of any psychiatric illness into early adulthood (30, 31). In the diagnostic category of affective disorders, depression was examined in five articles (26, 27, 28, 35, 36) and bipolar disorder in one article (34). Three of the articles evaluating depression demonstrated significant associations between active leisure time and lower levels of depression (28, 35, 36). In two articles, no significant differences in instances of depression were found between those participating in team sports (in this case American football) and non-team sports (26, 27). In addition, two articles considering affective disorders as a diagnostic group, showed that high level leisure time activity was related to a reduced likelihood of diagnoses of affective disorders. In four articles, active leisure time was associated to a decreased likelihood of diagnosed anxiety disorders in early adulthood (26, 27, 28, 30). Three articles reported a significant association between lower levels of physical activity with psychotic disorders being diagnosed later in life (32, 33, 33). High social activity was significantly associated with lower level of substance use disorders in one article (30). Participation in American football, compared to non-participation, was not associated with any differences in the risk of developing post-traumatic disorder or being diagnosed with dependence to or abuse of nicotine, cannabis, or alcohol (27). However, conduct disorder and adult antisocial behavior were found to be significantly less common among those who practiced sports (29). In addition, in one article, leisure time activity was reported to be significantly associated with lower numbers of behavioral disorders (30).
The quality score of the studies reviewed, assessed with the modified Strobe, varied from 23/27 (85%) to 27/27 (100%) (see Supplementary Table 1). This was considered to indicate that the reviewed articles were of good quality.
Far-reaching positive connections between leisure time activity and psychological well-being and symptoms have been widely researched and recognized (13, 14, 37). However, less attention has been paid to the relationship between traditional leisure time activities in adolescence to psychiatric disorders diagnosed later in life.
In this review, we have summarized study findings on the association of different types of traditional leisure time activities during adolescence to subsequent psychiatric morbidity into young adulthood. The time frame of publishing year of the reviewed articles included was set from 2010 to 2023, which is likely to ensure that the results of the review would be compatible to today's hobby practice. We undertook a broad search of the literature which produced, as expected, a vast number of studies addressing possible associations between leisure time activities and current mental health symptoms or disorders. However, the number of studies focusing on subsequent psychiatric morbidity was limited, with only eleven relevant studies meeting the inclusion criteria for this review.
The findings of this literature review supported our hypothesis that active engagement in leisure activities during adolescence is positively correlated with a lower incidence of psychiatric disorders in adulthood. However, given the various factors associated with the mental health status of individuals, the findings of this study should be interpreted with caution. Despite the relatively small number of reviewed studies, certain psychiatric disorders have undergone extensive analysis. We, there, believe the articles included in this review are pertinent in highlighting possible associations between different traditional leisure time activities in adolescence to subsequent psychiatric morbidity in adulthood.
Based on our literature review, the strongest evidence regarding an association of leisure time activities in adolescence to later psychiatric morbidity was found for depression and anxiety disorders, which also were the most commonly studied disorders in this context. There were five studies on depression, two studies considering affective disorders as a group and five studies on specific anxiety disorders. In almost all of these articles, participation in recreational activities in adolescence was connected to a lower prevalence of depression or anxiety in adulthood. For example, the study by Easterlin found that adolescents who participated in team sports had a lower likelihood of being diagnosed with anxiety and depression as adults than adolescents who did not play team sports (28). It was noticeable that, in all the articles reviewed, active participation to leisure time activity in adolescence was not related to an increased likelihood for the development of depression or anxiety later in life. Further, there were two articles (26, 27) specifically studying the possible effects on depression of team sports (i.e., American football), and no differences were found in the occurrence of depression later in life between team sports participants and non-participants.
The quality of the reviewed articles was evaluated to be good. Therefore, we feel justified in concluding that our results support the theory that participation in active leisure activities in adolescence is potentially associated with a decreased likelihood for depression and anxiety later in life. This is an important finding, because these disorders are epidemiologically the most common mental illnesses in the general population (38), and should, therefore, be key targets for preventive actions. Consistent findings of the reviewed studies indicate that different leisure time activities during adolescence can be considered to be important potential tools in preventive initiatives. One example how a society can organize leisure time activity is Finnish model for leisure time activity (
Despite psychotic disorders being epidemiologically relatively rare in the general population, compared to affective or anxiety disorders, nearly half of the reviewed studies had focused on psychoses (30, 31, 32, 33, 34). In addition, studies concerning psychotic disorders were performed on large cohort data with longitudinal follow-up, enabling the examination of these rarer mental health disorders. Our results indicated that recreational activities during adolescence may be beneficial in preventing the development of the most serious psychiatric disorders, such as schizophrenia or other psychosis, later in life. This conclusion is supported by the reverse findings of a methodologically sound study reporting that it found that lower levels of participation in leisure activities and sports competitions were associated with a higher risk of non-affective psychosis later in life (32).
Some of the studies had specifically analyzed bipolar disorder, post-traumatic disorder, and antisocial personality disorders. The findings of these studies suggest that active leisure time is associated with a decreased likelihood for the development of these disorders. However, conclusions should be made with great caution, because research-based knowledge and information accumulated on these specific psychiatric disorders is scarce and further research is required.
A particularly noteworthy finding, demonstrated in practically all of the articles included in our review, was that leisure time activities that include high amounts of social interaction, such as team sports, were associated with lower rates of psychiatric disorders in early adulthood. This finding strongly emphasizes the potential community impact on long-term psychosocial well-being. This finding is in line with the widely accepted theory that social interaction and togetherness have a positive effect on the comprehensive well-being and mental health of young people (11, 39) Social interaction has also been high-lighted as a key factor in the positive impact of different physical activities on current depression (40). The important role of social interaction during adolescence in its influence on psychiatric morbidity later in life is also understandable considering that socialization plays a significant role in the normal growth and development of adolescents (41). The findings of this review emphasized the positive effects of leisure activities that are engaged in within groups and communities. Consequently, it is pertinent to examine whether community and socialization in itself serves as the protective factor, rather than the specific activity. Additionally, further research is needed on the effects of solitary activities on later mental health. Our literature search did not identify any articles focusing solely on individual leisure time activities involving less social interaction, such as reading or swimming, or studies comparing different types of activities. Consequently, it is impossible to conclude with certainty whether solitary leisure activities have the corresponding associations on participants' subsequent mental health as those involving wider social connections. It is also important to consider that the health benefit can also come through via other mechanisms, like physical activity, which has broadly reported to increase positive health outcomes (42).
Evidently, when analyzing long-term associations of leisure time activities on mental health later in life, a long follow-up time would be beneficial. Our review’s finding suggests that longitudinal cohort studies would be the most appropriate way of exploring the relationship between leisure activities and mental health. These types of studies, however, frequently lack comprehensive and detailed information on leisure time activities. To increase study comparability, it would be preferable to utilize an internationally validated measure to examine leisure activities in future research. Therefore, it would be imperative to include this perspective while designing new longitudinal research.
While only relatively recently published articles covering the last 14 years were accepted for our review, the data of these studies dated back over three different decades and varied between the years 1980–2007. However, since our study focused solely on traditional leisure time activities, the information on leisure time activities for those years is, according to our understanding, comparable to similar types of modern-day activities.
It is noteworthy that studies concerning the later development of eating disorders did not emerge in our literature review. It can be considered surprising, because it is widely reported that certain leisure activities, ballet for example, are linked to eating disorders (43, 44). A plausible explanation for the lack of studies on eating disorders in our review is that eating disorders mostly appear during adolescence (45), while our focus was to evaluate the longitudinal impact of leisure time activity to psychiatric morbidity later in adulthood. The lack of research can also be influenced by the poor recognition of adult-onset eating disorders (46) which causes methodological challenges in sample size due to the significant underdiagnosis of eating disorders after adolescence.” However, more research is needed regarding the relationship between leisure activity in adolescence and eating disorders later in adulthood.
This review highlights the scarcity of research investigating the relationship between childhood and adolescent participation in leisure activities and the development of psychiatric disorders late in adulthood. The studies accepted for our review focused mainly on affective, anxiety, and psychotic disorders, indicating that many other diagnostic groups, such as substance use, and personality disorders are less explored. Childhood and adolescence are crucial periods for personality development, underscoring the need for additional research on how significant leisure activities during the pre-adulthood period are in relation to the development of personality and substance use disorders.
Sports were, by far, the most studied leisure activities in this review. Also, leisure time activities that include high amounts of social interaction were emphasized. Further research is required to understand the connections between other types of hobbies and leisure activities and the subsequent development of different psychiatric disorders.
The results of our review were based on studies analyzing the impact of traditional hobbies conducted in adolescence on later psychiatric morbidity, and, therefore, cannot be generalized to digital leisure activities. It is noteworthy that the virtual-reality space (metaverse) in which users can interact with a computer-generated environment and other users is increasingly a part of everyday life (20,47). With recent technological improvements, digital leisure activities have become increasingly important in young people's daily lives (19). However, when reviewing previous studies focusing on adolescents’ leisure time and later psychiatric morbidity for our study, it was not possible to evaluate in a reliable manner whether the digital activity discussed in a specific study indicated a leisure time hobby or a school- or work-related activity, or if it indicated problematic social-media behavior (21). Therefore, defining what digital activities can be qualified as hobbies and examining the relationship between digital leisure time and mental health disorders would be essential topics for future research. Recently published research on the adolescent population also called for longitudinal and randomized controlled trials to explore healthy and balanced use of digital media (48).
The results of the systematic literature searches were examined by Covidence (
The results of our review showed that half of the studies addressing hobbies and leisure time activities in association to later psychiatric morbidity were registry based and derived from national databases from Nordic countries. These databases include nationwide healthcare registers covering primary and special level healthcare visits. In these registers diagnoses are clinically set by doctors and these registry data are documented to provide reliable data for research purposes (49, 50). The articles from the United States, Brazil and Australia had utilized self-reported data on the diagnoses. Self-reports can cause increased likelihood for response bias, i.e. tendencies of study participants, whether accidentally or deliberately, to respond inaccurately or falsely to questions. In methodological point of view, the reviewed articles were heterogeneous with regards of leisure time activities and study design utilized in analyses, and, therefore, the interpretation of findings were drawn with caution., Although the number of articles accepted to this review was limited, we considered those to be consistent, allowing some important conclusions to be drawn regarding the association of traditional leisure activities in adolescence to specific psychiatric disorders emerging later in adulthood. However, the efficacy of the review is compromised by the paucity of studies on the selected subject, the heterogeneity of participant characteristics, the variability in study designs and outcomes, and the lack of comprehensive analysis on the associations between leisure activities and psychiatric disorders which takes into consideration the specific attributes of these activities.
Potential publication bias may exist, but its likelihood is low because if there were studies where the effects of leisure activities on mental health were negative, they would have obviously been published (51).
The objective of this review was to summarize the findings of studies analyzing the association of leisure time activities in adolescence with psychiatric disorders diagnosed later in life of the study participants. Therefore, the databases selected for the literature research primarily focused on medical research since there is a significant overlap between literature databases. This is why PsycInfo was not included in our research, even though it comprehensively covers studies of interest.
Our systematic review focused on the association between traditional leisure time activities during adolescence and psychiatric morbidity in later life. Despite the scarcity of research in this area, our summary of the findings suggests there is a positive association between active participation in leisure activities during adolescence and a decreased likelihood of depression, anxiety disorders and psychotic disorders in adulthood. This review provides important information for community-level decision makers by emphasizing the need to improve community-based leisure activities and make them more accessible and suitable for children and adolescents.