The global population aged 65 and older is rapidly expanding, posing significant challenges to healthcare systems due to increased prevalence of social and psychological issues like loneliness, social isolation, and worsening mental health. These factors are linked to higher rates of physical multimorbidity and mortality. While advancements in healthcare have extended lifespans, healthy life expectancy often lags behind, resulting in a growing burden of long-term mental health conditions. Exploring cost-effective strategies to improve mental health and wellbeing in older adults is crucial for the sustainability of global health and social care systems. Engagement in psychosocial activities, particularly hobbies, is gaining international attention as a potential solution. Hobbies—leisure activities pursued for pleasure (arts, crafts, reading, sports, gardening, volunteering etc.)—offer benefits through various pathways. They engage imagination, creativity, and cognitive function, and provide opportunities for self-expression and relaxation. Participation in hobby groups also offers social support, potentially reducing loneliness and social isolation. Many countries are promoting hobbies as a means of improving mental health, particularly among older adults. Existing research, while demonstrating positive links between hobbies and improved mental wellbeing (including reduced depressive symptoms, better subjective well-being and eudemonic well-being), suffers from limitations. Studies often focus on single nations, lack harmonized measures, and examine specific hobby subcategories, making cross-cultural comparisons and policy implications unclear. This study aimed to address these limitations by harmonizing hobby engagement and mental wellbeing measures across 16 countries, exploring the relationship, direction of association, and country-level variation.

A large body of research indicates a positive relationship between hobby engagement and various aspects of mental wellbeing in older adults. Meta-analyses of observational and interventional studies consistently show that hobbies, including nature-based activities and volunteering, are associated with a reduced risk of depressive symptoms. Studies examining specific hobby types, such as community groups, arts, and social clubs, similarly report lower rates of depression in adults aged 50+. Meta-analyses of leisure activities, including dancing, nature-based activities, and gardening, also demonstrate benefits for positive wellbeing dimensions like happiness and life satisfaction. However, existing research has limitations. First, studies are often country-specific, making it difficult to generalize findings. Second, diverse definitions and methodologies across studies hinder cross-cultural comparisons. Third, many studies focus on specific hobby types, neglecting the common underlying mechanisms that contribute to mental wellbeing across different hobbies. This study aimed to overcome these limitations by using harmonized measures across multiple countries.

This study employed a multinational meta-analysis design using longitudinal data from five large-scale studies: the English Longitudinal Study of Ageing (ELSA), the Japan Gerontological Evaluation Study (JAGES), the US Health and Retirement Study (HRS), the Survey of Health, Ageing and Retirement in Europe (SHARE), and the China Health and Retirement Longitudinal Study (CHARLS). The analysis included 93,263 participants aged 65+ from 16 countries. To ensure comparability across datasets, measures of hobby engagement and mental wellbeing were harmonized. A binary indicator of hobby engagement (yes/no) was created, along with standardized measures for four mental wellbeing outcomes: depressive symptoms, self-reported health, happiness, and life satisfaction. Nine time-varying covariates were controlled for (age, partnership status, household size, employment status, household income, housing tenure, long-standing health conditions, ADL difficulties, and IADL difficulties). Fixed effects models were used to examine within-individual changes in hobby engagement and mental wellbeing, controlling for all time-constant factors. These country-specific results were then pooled into multinational meta-analyses using a random-effects model. To investigate the directionality of the association, ordinary least squares (OLS) regressions were conducted, examining the association between hobbies at Time 1 and outcomes at Time 2, controlling for baseline outcomes and covariates. Sensitivity analyses were performed using multiple imputation for missing data, expanding the age range to 55+, stratifying by gender, and restricting to retired individuals. Multilevel models explored the moderating effect of country-level factors (prevalence of hobby engagement, GDP per capita, world happiness index, life expectancy, and Gini index) on the associations. The Arellano-Bond estimator was used on the ELSA dataset to further confirm directionality findings.

The prevalence of hobby engagement varied substantially across the 16 countries included in the study, ranging from 51% in Spain to 96% in Denmark. Fixed effects models and multinational meta-analyses showed a consistent association between hobby engagement and improved mental wellbeing. Specifically, having a hobby was significantly associated with: * Fewer depressive symptoms (pooled coefficient = −0.10; 95% CI = −0.13, −0.07) * Higher self-reported health (pooled coefficient = 0.06; 95% CI = 0.03, 0.08) * Greater happiness (pooled coefficient = 0.09; 95% CI = 0.06, 0.13) * Higher life satisfaction (pooled coefficient = 0.10; 95% CI = 0.08, 0.12) These associations were consistent across different statistical approaches and remained after controlling for potential confounders. Analyses examining directionality suggested a temporal relationship, with increased hobby engagement predicting subsequent improvements in mental wellbeing. Country-level factors explained less than 9% of the variance in these associations. However, hobby engagement prevalence correlated positively with the world happiness index score, country wealth, and life expectancy, and negatively with the Gini index. Meta-regressions showed a weak positive correlation between the world happiness index score and life satisfaction effect sizes and a positive correlation between life expectancy and self-reported health effect sizes. Sensitivity analyses, including multiple imputation for missing data, changing the age range, stratifying by gender, and restricting to retired individuals, yielded largely consistent results. The association between hobbies and outcomes did not vary by gender or retirement status.

This study provides robust evidence for the consistent association between hobby engagement and improved mental wellbeing in older adults across diverse cultural and national contexts. The findings support the promotion of hobbies as a cost-effective strategy to enhance mental health in aging populations globally. The relatively small variance explained by country-level factors suggests that the benefits of hobby engagement are relatively universal. The consistent association with life satisfaction is particularly notable, given its independence from objective health status and its potential for improvement even among individuals with long-standing health conditions. The temporal analyses strongly suggest that hobby engagement is not merely a consequence of good mental health but also a potential driver of improved wellbeing. The observed variation in hobby engagement rates across countries may reflect differences in access, societal barriers, or cultural norms. Addressing these inequalities is crucial for maximizing the potential benefits of hobbies and promoting health equity.

This large-scale multinational study provides compelling evidence for the widespread benefits of hobby engagement for mental wellbeing among older adults. The findings are remarkably consistent across diverse cultural contexts, highlighting the importance of prioritizing equitable access to hobbies to improve population mental health and support healthy aging globally. Future research should investigate the specific types, frequency, and duration of hobbies that maximize benefits, the potential role of hobby engagement in displacing less healthy activities, and the influence of individual-level factors on the hobby-wellbeing relationship. Natural experiments, such as changes in leisure or retirement policies, offer further opportunities to explore causal effects.

As an observational study, this research cannot definitively establish causality. While measures were harmonized, variations in question wording and response options across datasets might have introduced some subtle biases. The reliance on self-reported measures introduces potential recall bias. The relatively small number of studies for some countries could limit the power of some analyses and increase uncertainty in results. The study did not explore other potentially relevant intraindividual factors, and future studies with more interview waves may wish to explore this variability.