Introduction

Leisure activities are defined as enjoyable activities that individuals engage in during their free time, when they are free from the demands of their duties or obligations [1], and they are not limited to simple hobbies, but include a very wide range of activities, including social activities, participation in art or sports activities, outdoor activities such as fishing and hiking, gardening, reading, and watching television [2]. In particular, as people retire from work and social roles in old age, older adults have relatively more free time, and leisure activities are receiving attention as a key factor that determines quality of life in old age. Recently, as a result of the aging of the baby boomers and the overall improvement of the educational and economic levels of older adults [3], there has been growing interest in leisure activities, and types of leisure activities have also been becoming more diverse.

In older adults, participation in leisure activities can act as an important mechanism for mitigating negative emotions, such as helplessness and depression resulting from the loss of social roles or social isolation, and maintaining physical and mental health [1]. In addition, as a result of participation in leisure activities, through social contact, older adults can gain opportunities for self-realization as well as the improvement of self-esteem, and reconstruct the meaning of life in old age [1]. A number of previous studies have reported that in older adults, engagement in leisure activities positively influences various health outcomes, such as a decrease in depression [4], prevention of cognitive decline and dementia [2,5], improvement of self-rated health status [6,7], improvement of quality of life [7], and a lower risk of mortality [8].

However, leisure activities may have various health effects, because the required physical and mental functions and the mechanisms through which they contribute to health may vary depending on the nature of each type of leisure activity [9]. For example, leisure activities that involve physical activities can reduce the risk for cardiovascular disease [10], diabetes [11], and cognitive decline [12], and also have a positive impact on alleviation of depressive symptoms not only through biological mechanisms, such as the increase of neuroplasticity and neurochemical changes, but also through psychosocial mechanisms, such as self-efficacy and a sense of control [13,14]. On the other hand, sedentary leisure activities, such as reading or cultural activities, may have limited physical effects [15], but they have been reported to be effective in maintaining cognitive function and preventing dementia through a high level of cognitive stimulation [16], and contribute to the decrease of mortality rates [17]. In addition, leisure activities involving social interaction have been shown to have a significant impact on promoting mental health [18], preventing social isolation [19], and alleviating cognitive decline [20] through various socio-psychological pathways, such as enhancing social support and alleviating stress [21].

As the health benefits of leisure activities can vary depending on the nature of leisure activities as described above, it is important to examine the effects of leisure activities by type of activity. For example, a longitudinal study of Chinese older adults by Cai et al. [22] reported that the survival rate was highest in the group that engaged in both physical activities and cognitive stimulation activities, but the group that engaged only in cognitive stimulation activities showed the lowest survival rate. In addition, Kobayashi et al. [8] found that while engagement in physical leisure activities as well as group-based leisure activities showed a significant positive association with lowered mortality, cultural leisure activities, such as reading and listening to music, had no significant relationship with mortality. Domestic studies have also shown that physical activity is significantly associated with life satisfaction [23] and depression [24,25] in older adults. On the other hand, some recreational activities have been found to have a negative impact on depression. These findings of prior studies suggest that leisure activities do not always have a positive influence, and that the impact of leisure activities on health outcomes may vary depending on the type of leisure activity.

Nevertheless, previous studies were mainly focused on health outcomes according to the frequencies of specific types of leisure activities [2,4,14,16,17] or on long-term and comprehensive indicators such as mortality and life satisfaction [7,8], and there has been a relative lack of research to comparatively analyze health outcomes according to types of leisure activities [6]. Although some domestic studies analyzed the relationship between leisure activity types and mental health [23-25], their analyses were limited to samples from particular regions [23-25] or to some health indicators such as life satisfaction [23] and depression [24,25].

Therefore, in order to complement the above-mentioned limitations of these previous studies, this study used multidimensional indicators encompassing the physical, mental, and cognitive health of older adults using data from the 2000 National Survey of Older Koreans, which is considered to be representative of the entire elderly population in Korea. Self-rated health is a representative indicator predicting life satisfaction and mortality [26], and depression and cognitive function are key factors that determine quality of life and functional independence in old age, and thus have significant implications regarding policies for older adults. In particular, by setting self-rated health, depression, and cognitive function as dichotomous variables, this study aimed to provide reference points for policy-based interventions, and to investigate the different degrees of contribution of different types of leisure activities to health risks, because these effects of types of leisure activities are considered to have high policy applicability.

Methods

Study design

This study is a cross-sectional, secondary data analysis study using data from the 2020 National Survey of Older Koreans, and attempted to classify the types of leisure activities of Korean older adults, and examine the associations between each type of leisure activity and major health outcomes (self-rated health, depression, and cognitive function) in older adults.

Participants and data collection

This study was conducted using raw data from the 6th National Survey of Older Koreans (2020), which was jointly conducted by the Ministry of Health and Welfare and the Korea Institute for Health and Social Affairs. The National Survey of Older Koreans has been conducted every three years since 2008. The survey data is official statistics approved by the Ministry of Data and Statistics (formerly, Statistics Korea), and the validity of the sample design and the reliability of the content of the survey have been verified.

The data was collected from a sample of community-dwelling older adults aged 60 years or older selected nationwide by a multi-stage stratified cluster sampling method, and 10,097 people from a total of 969 enumeration areas participated in the survey. Data collection was conducted using the Tablet-PC Assisted Personal Interview (TAPI) method by 169 trained interviewers from September 14 to November 20, 2020. Weights were applied to ensure that the survey results reflect the demographic distribution of the population, and all data were used after receiving approval from the Institutional Review Board (IRB) of the affiliated institution.

In this study, data from 7,806 people were finally used for analysis, excluding 2,291 people, who included those who did not respond to questions about leisure activities (those who did not respond about the first and second primary leisure activities), those with missing values for variables related to depression and cognitive function, and those with responses from proxy respondents.

Measurements

1. Dependent variables: Self-rated health, depression, and cognitive function

The main health outcome variables of this study were self-rated health, depressive symptoms, and cognitive function. Self-rated health was assessed with a single question about one’s usual health status, and respondents were asked to respond as ‘Very good,’ ‘Good,’ ‘Fair,’ ‘Poor,’ or ‘Very poor.’ Regarding poor self-rated health, self-rated health was set as a dichotomous variable by dichotomizing responses about self-rated health status by classifying ‘Poor’ and ‘Very poor’ as ‘Poor’ (0 = Good or Very good, 1 = Poor or Very poor). Self-rated health assessed using a single item has been shown to be a valid indicator predicting mortality in previous studies across multiple countries [26].

Depression was measured using the South Korean Geriatric Depression Scale-Short Form (SGDS-K), which contains 15 items. Each item was scored by giving 1 point for ‘Yes’ and 0 points for ‘No,’ and negative questions were reverse-scored. The total scores range from 0 to 15 points, and a higher score indicates a higher level of depression. A score of 5 or higher points is classified as ‘mild depression’ and a score of 10 or higher points as ‘severe depression’ [27]. In this study, scores of 5 or higher points were classified as ‘depression (1)’ and scores of less than 5 points were classified as ‘normal (0).’ Regarding the reliability of the scale, the value of Cronbach’s α was 0.83.

Cognitive function was assessed using the Korean-Mimi Mental State Examination (K-MMSE). K-MMSE scores, which indicate cognitive ability, range from 0 to 30 points, and they are adjusted scores obtained by adjusting the raw scores according to educational level to reduce distortion of cognitive test results due to differences in educational level, and more accurately evaluate actual cognitive function. In the case of K-MMSE scores, 24 points or higher is classified as normal, 19 to 23 points as mild cognitive impairment, 10 to 18 points as moderate dementia, and 9 points or less as severe dementia [28]. However, this study dichotomized cognitive function scores by classifying a K-MMSE score of 23 points or lower as ‘poor cognitive function (1)’ and a K-MMSE score of 24 points or higher as ‘normal (0).’

2. Independent variables: Types of leisure activities

Types of leisure activities were examined using a question that asks respondents to select their first and second main leisure activities performed in the past year. Types of leisure activities were divided into 7 main categories (cultural and artistic viewing, participation in culture and arts activity, watching sports, sports participation activity, hobbies & recreational activity, rest activity, and social and other activity) and 22 subcategories. In order to explore the types of leisure activities of older adults, the first and second-ranked items of the main leisure activities can be used, and this method has the advantage of allowing comprehensive analysis of complex combinations of the responses regarding multiple leisure activity items [29]. In other words, it is possible to identify potential leisure activity types and lifestyle characteristics of the elderly that are difficult to identify using responses about a single main leisure activity, and to explore the relationships of leisure activities with health, social participation, and quality of life indicators. Therefore, many previous studies suggested classifying leisure activity types and deriving their characteristics by analyzing two or more main leisure activities that respondents prefer or participate in [29]. Since the responses to the first and second main leisure activities were all categorical variables, for cluster analysis, they were converted into continuous variables by changing the responses of categorical variables into dummy variables by assigning 1 if the response was ‘Yes’ for the specific activity presented for each variable and assigning 0 otherwise, and then cluster analysis was conducted [30].

3. Control variables

To explore the association between leisure activity types and health independently from the impact of other variables, a total of nine sociodemographic and health behavioral variables were included in the analysis as control variables, based on previous studies, and the nine control variables are as follows: gender, age, marital status, coresidency, residential area (urban, rural), education level, household income, economic activities, smoking, and drinking. For household income, this study used equivalent household income, a measure of household income adjusted for household size.

Data analysis

Types of leisure activities were explored using the K-means cluster analysis method. Since the first and second main types of leisure activities of older adults were categorical variables, as many dummy variables as the number of values for each variable were generated, and cluster analysis was performed by converting categorical variables into continuous variables. In the case of qualitative data, K-means cluster analysis can be performed by converting qualitative data into quantitative data through dummification. This cluster analysis method allows calculation of differences between the characteristics of each variable value, and the accuracy of the cluster analysis results has been verified in previous studies [30]. In cluster analysis, initial cluster centroids were assigned to each reference variable, and an iterative, centroid-based clustering process was repeated until a pre-defined, optimal number of clusters based on the distance between cluster centroids was achieved. After the clusters were derived, the distribution of the reference variables of the clusters was comparatively evaluated to examine the differences between the clusters and evaluate the validity.

To examine the significant differences in the distribution of leisure activity types according to demographic and socioeconomic characteristics and health-related characteristics, cross-tabulation analysis, the Chi-squared test, analysis of variance (ANOVA), and post-hoc tests were conducted. To investigate the degree to which each type of leisure activity contributes to the risk of poor self-rated health, depression, and poor cognitive function among older adults, which was one of the main objectives of this study, logistic regression analysis was used as the main analysis method. Multicollinearity between independent variables in multiple logistic regression analysis was assessed using the variance inflation factor (VIF), and all variables had a VIF value of 0.79 or less, indicating that there was no multicollinearity. The goodness-of-fit of the logistic regression model was assessed using the Hosmer-Lemeshow goodness-of-fit test, and the model was considered a good fit because the difference between the observed and expected values was not significant. In addition, to test the sensitivity and robustness of the main analysis results, univariate analysis of covariance (ANCOVA) controlling for covariates was additionally conducted for the scores for self-rated health, depressive symptoms, and cognitive function, and the results of ANCOVA are presented as supplementary data (Supplementary Table S1). This is because violations of the prerequisites (normality and homoscedasticity) identified in the ANCOVA analysis reduced the reliability of the F-test and p-value. Therefore, logistic regression analysis, the primary analysis method used in this study, was thought to be a more statistically robust approach to data asymmetry and heteroscedasticity.

Finally, the publicly available data of the National Survey of Older Koreans did not include the primary sampling unit (PSU), so it was not possible to conduct a design-consistent analysis of complex sample survey data. Therefore, although weighting was limitedly applied, this study applied weights to reduce the bias of a point estimator when analyzing research data, and applied robust standard error inflation (1.23 to 1.43 times) using Kish’s Deff formula to compensate for the possibility of underestimation of the variance. We also presented recalculated confidence intervals and testing results according to the median scenario (p=0.02). This method is widely used in international surveys and health and social panel studies [31-33]. All analyses were performed using IBM SPSS Statistics 21.0.

Ethical considerations

This study is a secondary data analysis using data from the 2020 National Survey of Older Koreans. The research data is public data from which personal information has been removed. This study was conducted after receiving an exemption determination from the Institutional Review Board of the affiliated institution of the researcher (IRB No. MNUIRB-250215-SB-003-01).

Results

Socioeconomic characteristics and prevalence of poor self-rated health, depression, and poor cognitive function of korean older adultsThe left panel of Table 1 shows the demographic, socioeconomic, and health-related characteristics of older adults aged 65 and older in this study. The participants were a total of 7,806 older adults. Out of the total participants, 42.7% were male and 57.3% were female. The average age of the participants was 72.97 years (±6.34). In terms of marital status, 68.3% were currently married, 28.7% were widowed, and 2.9% were categorized as other cases, such as divorced, separated, or unmarried. The most common form of coresidency was ‘couple only’ (47.6%), followed by ‘living alone’ (27.1%) and ‘living with kids’ (23.4%). In terms of education level, high school accounted for the largest proportion (30.4%), followed by elementary school at 29.7%, middle school at 23.9%, uneducated at 9.4%, and college or over at 6.7%. The average equivalent household income per year was 19,980,500 (±25,844,600) won, and 37.7% of the total participants were engaged in economic activities.

With respect to health behaviors, the proportions of smokers and drinkers were 12.4% and 84.4%, respectively. 18.3% of the total participants self-rated their health status as ‘poor’ or ‘very poor.’ The ‘risk group for depression’, which had a score of 5 points or higher on the SGDS-K, accounted for 30.4%. The average K-MMSE score was 24.23 (±5.11) points, and 29.3% of the total participants were classified as having ‘poor cognitive function’ with a score below the cutoff point (≤24 points), as shown in the left panel of Table 1.

As a result of testing the differences in the scores for self-rated health, depression, and cognitive function according to the type of leisure activity, the ‘religious activity-oriented’ group had a lower score of self-rated health and a higher score of depression, compared to the other four leisure activity type groups (p<.001), and the ‘religious activity-oriented’ group also had a lower score of cognitive function than the ‘various physical activity-oriented’ group and the ‘social activity-oriented’ group (p<.001). In the supplementary ANCOVA controlling for covariates (Supplementary Table S1), the ‘religious activity-oriented’ group showed the most unfavorable health status among the five leisure activity type groups.

Types of leisure activity of korean older adults

Based on older adults’ responses about their first- and second-priority leisure activities, K-means cluster analysis was performed, and a total of five types of leisure activities were derived (Table 2). In Cluster 1, for the first-priority leisure activity, the proportion of religious activities was highest at 38.9%, and for the second-priority leisure activity, there were almost no responses, so Cluster 1 was named ‘religious activity-oriented.’ In the case of Cluster 2, for the first-priority leisure activity, various physical activities, including sports (10.1%), hiking (13.5%), and walking (33.0%), were the most common type of activity accounting for 56.6%, and for the second-priority leisure activity, walking (23.0%) was the most common activity, so Cluster 2 was named ‘various physical activity-oriented.’ As for Cluster 3, the most common activities reported as the first-priority leisure activity were walking (42.3%) and rest activity (19.9%), and the second-priority leisure activity was social activity (100.0%), so Cluster 3 was named ‘walking and rest-oriented.’ In the case of Cluster 4, for the first-priority leisure activity, the proportion of gardening activity was highest at 61.2%, and for the second-priority leisure activity, rest (49.1%) was ranked first, so Cluster 4 was named ‘gardening and sedentary activity-oriented.’ For Cluster 5, the first-priority leisure activity was social activity (100.0%), so Cluster 5 was named ‘social activity-oriented.’ Among the total participants, the proportion of the various physical activity-oriented group was highest at 46.7%, followed by the religious activity-oriented group (33.2%), the walking and rest-oriented group (8.0%), the gardening and sedentary activity-oriented group (7.8%), and the social activity-oriented group (4.4%).

The right panel of Table 1 shows the differences in the distribution of types of leisure activities according to sociodemographic and health-related characteristics. There were significant differences in types of leisure activities according to gender, age, marital status, coresidency, residential area, education level, household income, and engagement in economic activities. In males, the proportions of the various physical activity-oriented group (49.6%), the social activity-oriented group (41.1%), and the walking and rest-oriented group (38.2%) were relatively higher. Meanwhile, in females, the proportions of the gardening and sedentary activity-oriented group and (67.0%) and the religious activity-oriented group (63.4%) were relatively higher (p<.001). In the 65-69 age group, the various physical activity-oriented group and the gardening and sedentary activity-oriented group took up a high proportion, but in the ≥ 80 age group, the proportion of the religious activity-oriented group was relatively higher (p<.001). In terms of marital status, among married older adults, the various physical activity-oriented group took up the largest proportion, followed by the walking and rest-oriented group, the social activity-oriented group, and the gardening and sedentary activity-oriented group in descending order, but among widowed older adults, the proportion of the religious activity-oriented group was highest (p<.001). Regarding coresidency, older adults living alone showed a high proportion of the religious activity-oriented group, those living with the spouse had a high proportion of the various physical activity-oriented group, and those living with kids showed a low proportion of the social activity-oriented group (p<.001). Regarding residential area, among older adults living in an urban area, the proportions of the various physical activity-oriented group and the gardening and sedentary activity-oriented group were relatively higher, but among those living in a rural area, the religious activity-oriented group and the gardening and sedentary activity-oriented group took up a high proportion (p<.001). In terms of education level, among older adults with an education level of college or over and those with an education level of high school, the proportions of the various physical activity-oriented group and the garden and sedentary activity-oriented group were high, while the proportion of the religious activity-oriented group was high among those with an education level of elementary school or lower. Among older adults in the higher household income group (the first and second quartiles in this study), the gardening and sedentary activity-oriented, various physical activity-oriented, and social activity-oriented groups showed a high proportion, whereas the proportion of the religious-activity oriented group was highest in the lower household income group (the third and fourth quartiles in this study) (p<.001). Among those currently engaged in economic activities, the various social activity-oriented group had a high proportion, while the proportions of the walking and rest-oriented and the religious activity-oriented groups were high among retired people (p<.001).

In addition, there were also differences in the distribution of types of leisure activities according to health-related characteristics. Regarding smoking status, the proportion of the social activity-oriented group was high among older adults who were smokers, while the proportions of the religious activity-oriented and gardening and sedentary-activity-oriented groups were high among non-smoking older adults (p<.001). The score for self-rated health and the score for cognitive function were higher in the social-activity oriented, the various physical activity-oriented, and the walking and rest-oriented groups than the religious activity-oriented group, but the score for depressive symptoms was highest in the religious activity-oriented group (p<.001).

Association between types of leisure activity and poor self-rated health, depression, and poor cognitive function Table 3 shows the results of logistic regression analysis to examine the differences in the risk of poor self-rated health, depression, and poor cognitive function according to the type of leisure activity. To increase the clarity of analysis results, the ‘religious activity-oriented’ group, which showed the most vulnerable health status, was set as the reference group, based on the results of Table 1 presented above and the results of ANCOVA controlling for covariates (Supplementary Table 1). The main research results are as follows. In Model 1 (simple regression analysis), the odds ratios for poor self-rated health, depression, and poor cognitive function were significantly lower in the various physical activity-oriented, walking and rest-oriented, garden and sedentary activity-oriented, and social activity-oriented groups than the religious activity-oriented group. In the results of Model 2 (regression analysis after controlling for covariates), the risk of poor self-rated health was significantly lower in three other leisure activity type groups, compared to the ‘religious activity-oriented’ group. In other words, compared to the ‘religious activity-oriented’ group, the social activity-oriented group had an odds ratio of 0.46 for the risk of poor self-rated health (95% CI=0.29-0.72), the walking and rest-oriented group had an odds ratio of 0.55 (95% CI=0.39-0.77), and the various physical activity-oriented group had an odds ratio of 0.71 (95% CI=0.59-0.85). Meanwhile, although the ‘gardening and sedentary activity-oriented’ group had an odds ratio of 0.80 for poor self-rated health, compared to the ‘religious activity-oriented’ group, there was no statistically significant difference between the two groups. With respect to the risk of depression, the risk of depression was statistically significantly lower in all of the other four leisure activity type groups, compared to the ‘religious activity-oriented’ group. More specifically, the odd ratio for the risk of depression was lowest in the ‘social activity-oriented’ group (OR=0.40, 95% CI=0.28-0.57), followed by the ‘walking and rest-oriented’ group (OR=0.41, 95% CI=0.31-0.54), the ‘gardening and sedentary activity-oriented’ group (OR=0.47, 95% CI=0.36-0.62), and the ‘various physical activity-oriented’ group (OR=0.48, 95% CI=0.41-0.56) in ascending order. Meanwhile, in the risk of poor cognitive function, as a result of performing regression analysis after controlling for covariates (Model 2), all of the other four leisure activity type groups did not show a statistically significant difference, compared to the ‘religious activity-oriented’ group.

Discussion

This study attempted to classify the types of main leisure activities of Korean older adults, and explore how each type of leisure activity contributes to physical, mental, and cognitive health in Korean older adults by using data from the 2020 National Survey of Older Koreans. The results of cluster analysis showed that Korean older adults’ leisure activities can be categorized into five types: ‘religious activity-oriented,’ ‘various physical activity-oriented,’ ‘walking and rest-oriented,’ ‘gardening and sedentary activity-oriented,’ and ‘social activity-oriented.’ Among the five types, the ‘various physical activity-oriented’ type (46.7%) and the ‘religious activity-oriented’ type (33.2%) were the most common types, while the ‘social activity-oriented’ type (4.4%) was the least common one. The main research findings of this study are discussed below.

As a result of exploring differences in health risks according to the type of leisure activity, the ‘various physical activity-oriented’, ‘walking and rest-oriented,’ and ‘social activity-oriented’ groups had a lower risk for poor self-rated health and depression, compared to the ‘religious activity-oriented’ group. These results are consistent with previous studies reporting that physical and social activities have a positive effect on perceived health status [6,7] and depression [4,14,18,24] among older adults. In particular, the ‘social activity-oriented’ and ‘walking and rest-oriented’ groups showed lower odds for poor self-rated health and depression than the ‘various physical activity-oriented’ group, and this result is presumed to be an effect of the fact that social activities are included as a major component in these leisure activity types. In fact, Table 2 shows that in the case of the ‘social activity-oriented’ group, 100% of the first-priority activity was social activities, and for the second-priority leisure activity, walking (25.0%) took up a high proportion. Additionally, in the case of the second-priority activity of the ‘walking and rest-oriented’ group, the proportion of social activities was 100%, supporting the contribution of social interaction to reducing the risk of poor self-rated health and depression. These results are consistent with previous studies [34,35] reporting that the positive impact of social leisure activities on self-rated health and depression increases with age. Leisure activities that involve social interaction have been reported to enhance a sense of belonging and social identity, and thereby have the effects of alleviating depression [36] and improving mental health [37,38]. In addition, leisure activities involving social interaction increase accessibility to health-related information, thereby promoting healthy lifestyles and preventing unhealthy and high-risk behaviors, and also play the role of protecting older adults from social isolation in addition to promoting physical health [39]. However, according to a survey conducted in 2023, among Korean older adults, rest (96.5%) was the most common activity among leisure and cultural activities, the average number of social participation activities, including socializing and religious activities, was 1.1, the level of social participation activities was low, and the content of social participation activities was limited [40]. Therefore, to ensure that older adults can form diverse social relationships and actively participate in leisure activities, it is necessary to develop programs for promoting their participation in leisure activities, provide spaces for leisure activities, and expand social support.

With respect to the degree of contribution to health outcomes, the differences in the degree of contribution to depression between types of leisure activities were more noticeable than the differences in the degree of contribution to poor self-rated health and poor cognitive function. In other words, compared to the reference group, the ‘religious activity-oriented’ group, all of the other four leisure activity types, including the ‘gardening and sedentary activity-oriented’ group, showed a significantly lower odds ratio for depression. This result is consistent with prior studies reporting that activities that can be defined as leisure activities positively contribute to reducing depression in older adults, although the degree of contribution varies among diverse leisure activities [41]. Since leisure activities involve a variety of positive elements, including creative expression, aesthetic pleasure, cognitive stimulation, activation of the human senses, emotional stimulation, and social interaction, they act as one of the main factors that contribute to the improvement of mental health in older adults [42]. However, all leisure activities do not have the same level of impact on health, and their effects on health may be different depending on the nature and context of the leisure activity. In this study, when the ‘religious activity-oriented’ group was used as the reference group, the risk of depression was found to be lower in all of the other four leisure activity type groups, and this research result is consistent with previous studies that found that religious activities have a negative or minimal impact on depression, compared to other activities [43,44]. However, there is inconsistency in research findings across studies regarding the relationship between religious activity and depression, and some studies found that participation in religious activities had a more positive effect in reducing depressive symptoms than physical or social activities [41]. Therefore, it is necessary to accumulate empirical data by conducting further empirical research on the impact of leisure activities on depression, taking into account the characteristics and context of various leisure activities, including religious activities.

Although there were significant differences between the religious activity-oriented type and other types of leisure activities in the case of the degree of contribution to the risk of poor self-rated health and depression, multiple logistic regression analysis controlling for covariates did not show any significant differences in contribution to the risk of poor cognitive function between different types of leisure activities. This finding is in contrast to the results of previous studies that reported the positive effect of regular exercise on the improvement of cognitive function [45-47], the contribution of gardening activity to cognitive stimulation [48,49], and the positive effect of active social participation on the maintenance of cognitive function [50-51]. This disagreement in research findings is thought to require a comprehensive consideration of the following factors. First, it is necessary to consider the difference in the sensitivity of analysis methods. While the logistic regression analysis in this study explored dichotomous risk using the clinical cutoff (MMSE≤24), the results of the sensitivity analysis of the supplementary ANCOVA (Supplementary Table 1) revealed that when continuous cognitive function scores were used, the ‘various physical activity-oriented’ group and ‘social activity-oriented’ group showed significantly higher adjusted mean scores than the ‘religious activity-oriented’ group. The analysis using a continuous variable for cognitive function detected subtle differences in the level of cognitive function according to leisure activity types, but the significant effects of types of leisure activities were not found when clinical risk for a dichotomous outcome based on the clinical cutoff was analyzed. Therefore, the results of the logistic regression analysis in this study are thought to have significance in that this logistic regression analysis clearly provided ‘risk’ information with high policy applicability, and it was a statistically robust approach more suitable for the research purpose than comparing the means of continuous variables. Second, there is a need to consider the lack of quantitative information on leisure activities. One of the limitations of this study is that it focused on categorizing types of leisure activities, and thus failed to consider quantitative indicators such as the frequency and intensity of leisure activities. Because the cognitive effects of leisure activities may vary depending on the quantity and quality of participation in these activities, there is a possibility that the effects of activities may not have been fully reflected. Third, there is a possibility that reverse causality may have diluted the effects of leisure activities. In other words, it cannot be ruled out that older adults already experiencing cognitive decline may choose simple or sedentary activities with a relatively lower cognitive load. In particular, in the case of the ‘gardening and sedentary activity-oriented’ type, although gardening provides cognitive stimulation, this positive cognitive stimulation effect may have been diluted as rest activities accounted for a high proportion in the ‘gardening and sedentary activity-oriented’ type in this study [52]. Fourth, it is required to consider qualitative differences in social activities and the specific nature of the measurement domains of cognitive function [53]. The analysis results about the effects of types of leisure activities may vary depending on the specific nature of social activities or the measured domain of cognitive function. For example, a prior study reported that personal meetings with close friends were not associated with cognitive decline, but group activities such as social clubs or class reunions showed a significant association with delayed cognitive decline [54]. Therefore, to overcome the limitations described above, in future studies, there is a need to clearly elucidate the causal relationship between the types of leisure activities and cognitive function through a longitudinal analysis that comprehensively considers the frequency and intensity of gardening and sedentary activities as well as physical and social activities.

This study empirically classified the types of leisure activities of Korean older adults by K-means cluster analysis using the 2020 National Survey of Older Koreans,’ which used a sample that was representative of the elderly population nationwide in Korea. In particular, this study has academic significance in that it comprehensively explored the differences between different types of leisure activities in the contribution (odds ratio) to the risk of three dichotomous health outcomes: poor self-rated health, depression, and poor cognitive function. The results of this study have practical value in that they provided a comprehensive understanding of leisure activities for promoting the health of older adults and can serve as empirical fundamental data for the development of diverse leisure programs and creation of an environment for establishing policies. Nevertheless, this study has several methodological limitations as follows. First, since this research is a cross-sectional study, it has limitations in identifying clear causal relationships between the types of leisure activities and health outcomes. Therefore, there is a possibility that different types of leisure activities may have different impacts on health, and it is also difficult to rule out the possibility of reverse causality, which means that the current health status or functional level of individual respondents may have led to the selection of specific leisure activities. Second, the quantitative characteristics of participation in leisure activities were not considered in this study. The actual health-promoting effects of leisure activities can vary significantly depending not only on the type of leisure activity but also on the frequency and intensity of leisure activity. However, this study has measurement limitations in that it did not include the quantitative indicators in the analysis. Therefore, in future studies, it is necessary to apply a longitudinal study design to overcome the limitations of cross-sectional studies and clearly identify causal relationships, and to conduct a more in-depth exploration of the relationships between the types of leisure activities and health outcomes by comprehensively including not only the types of leisure activities but also quantitative indicators such as the frequency and intensity of leisure activities in the analysis.

Conclusions

This study classified the types of main leisure activities of older adults, and explored the degree of contribution of each type of leisure activity to the risks of dichotomous health outcomes, that is, poor self-rated health, depression, and poor cognitive function by multiple logistic regression analysis using data from the National Survey of Older Koreans. The analysis results showed that when the ‘religious activity-oriented’ group was set as the reference group, the ‘social activity-oriented,’ ‘walking and rest-oriented,’ and ‘various physical activity-oriented’ groups had a significantly lower odds for poor self-rated health. In addition, all the four leisure activity types, including the ‘gardening and sedentary activity-oriented’ type as well as the above-mentioned three types, were found to provide a significant protective effect against the risk of depression. On the other hand, in the degree of contribution to the risk of poor cognitive function (MMSE≤24), none of the four leisure activity types showed a significant difference from the ‘religious activity-oriented’ type. These results clearly suggest that leisure activities involving social interaction as well as a variety of physical activities such as walking are key protective factors for improving overall health in older adults. Thus, based on the above research findings, this study suggests the following directions of policy-based interventions to improve the physical, mental, and cognitive health of older adults. First, there is a need to strengthen social networks and mental health. In other words, it is necessary to enhance social networks by activating leisure spaces and programs that promote social interaction, such as hobby classes at senior welfare centers, support for social or hobby clubs, and intergenerational activities, rather than simply recommending leisure activities. Second, measures to ensure sustainable participation in physical activity are needed. To this end, community-based exercise programs, such as stretching exercises for older adults, groups for group walking, and hiking clubs, should be expanded to encourage people to continuously participate in physical activity. Third, customized support for leisure activities should be provided. Considering the observed significant differences in the distribution of types of leisure activities according to socioeconomic characteristics such as gender, age, and education level, it is suggested that it is necessary to establish a strategy for developing and supporting customized leisure activities that reflect sociodemographic characteristics.

Supplementary Materials

References

• 1. Adams KB, Leibbrandt S, Moon H. A critical review of the literature on social and leisure activity and well-being in later life. Ageing and Society. 2011;31(4):683–712. https://doi.org/10.1017/S0144686X10001091Article
• 2. Mao C, Li Z, Lv Y, Gao X, Kraus VB, Zhou J, et al. Specific leisure activities and cognitive functions among the oldest-old: The Chinese longitudinal healthy longevity survey. The Journals of Gerontology: Series A. 2020;75(4):739–746. https://doi.org/10.1093/gerona/glz086Article
• 3. Jeong KH. An Outlook of baby boomers’ old age life. Health and Social Welfare Forum. Research Report. Sejong: Korea Institute for Health and Social Affairs; 2012. Report No.: 2012.05.N187_04.
• 4. Bone JK, Bu F, Fluharty ME, Paul E, Sonke JK, Fancourt D. Engagement in leisure activities and depression in older adults in the United States: Longitudinal evidence from the Health and Retirement Study. Social Science & Medicine. 2022;294:114703. https://doi.org/10.1016/j.socscimed.2022.114703Article
• 5. Barragan GA, Lucumi D, Lawlor B. Association of leisure activities with cognitive impairment and dementia in older adults in Colombia: A SABE-based study. Frontiers in Neurology. 2021;12:629251. https://doi.org/10.3389/fneur.2021.629251ArticlePubMedPMC
• 6. Tomioka K, Kurumatani N, Saeki K. Cross-sectional association between types of leisure activities and self-rated health according to gender and work status among older Japanese adults. Journal of Epidemiology. 2019;29(11):424–431. https://doi.org/10.2188/jea.JE20180108ArticlePubMedPMC
• 7. Chick G, Hsu YC, Yeh CK, Hsieh CM. Leisure Constraints, leisure satisfaction, life satisfaction, and self-rated health in six cities in Taiwan. Leisure Sciences. 2015;37(3):232–251. https://doi.org/10.1080/01490400.2014.967897Article
• 8. Kobayashi T, Tani Y, Kino S, Fujiwara T, Kondo K, Kawachi I. Prospective study of engagement in leisure activities and all-cause mortality among older Japanese adults. Journal of Epidemiology. 2022;32(6):245–253. https://doi.org/10.2188/jea.JE20200427ArticlePubMedPMC
• 9. Nilsen C, Agahi N, Shaw BA. Does the association between leisure activities and survival in old age differ by living arrangement? Journal of Epidemiology and Community Health. 2018;72(1):1–6. https://doi.org/10.1136/jech-2017-209614ArticlePubMed
• 10. Nocon M, Hiemann T, Müller-Riemenschneider F, Thalau F, Roll S, Willich SN. Association of physical activity with all-cause and cardiovascular mortality: a systematic review and meta-analysis. European Journal of Cardiovascular Prevention and Rehabilitation. 2008;15(3):239–246. https://doi.org/10.1097/HJR.0b013e3282f55e09ArticlePubMed
• 11. Aune D, Norat T, Leitzmann M, Tonstad S, Vatten LJ. Physical activity and the risk of type 2 diabetes: A systematic review and dose-response meta-analysis. European Journal of Epidemiology. 2015;30(7):529–542. https://doi.org/10.1007/s10654-015-0056-zArticlePubMed
• 12. Bherer L, Erickson KI, Liu-Ambrose T. A review of the effects of physical activity and exercise on cognitive and brain functions in older adults. Journal of Aging Research. 2013;2013(1):657508. https://doi.org/10.1155/2013/657508ArticlePubMedPMC
• 13. Erickson KI, Gildengers AG, Butters MA. Physical activity and brain plasticity in late adulthood. Dialogues in Clinical Neuroscience. 2013;15(1):99–108. https://doi.org/10.31887/DCNS.2013.15.1/kericksonArticlePubMedPMC
• 14. Wassink-Vossen S, Collard RM, Oude Voshaar RC, Comijs HC, de Vocht HM, Naarding P. Physical (in)activity and depression in older people. Journal of Affective Disorders. 2014;161:65–72. https://doi.org/10.1016/j.jad.2014.03.001ArticlePubMed
• 15. de Rezende LF, Rey-López JP, Matsudo VK, do Carmo Luiz O. Sedentary behavior and health outcomes among older adults: A systematic review. BMC Public Health. 2014;14:333. https://doi.org/10.1186/1471-2458-14-333ArticlePubMedPMC
• 16. Fancourt D, Steptoe A. Cultural engagement predicts changes in cognitive function in older adults over a 10 year period: Findings from the English Longitudinal Study of Ageing. Scientific reports. 2018;8:10226. https://doi.org/10.1038/s41598-018-28591-8ArticlePubMedPMC
• 17. Väänänen A, Murray M, Koskinen A, Vahtera J, Kouvonen A, Kivimäki M. Engagement in cultural activities and cause-specific mortality: Prospective cohort study. Preventive medicine. 2009;49(2-3):142–147. https://doi.org/10.1016/j.ypmed.2009.06.026ArticlePubMed
• 18. Kawachi I, Berkman LF. Social ties and mental health. Journal of Urban Health. 2001;78(3):458–467. https://doi.org/10.1093/jurban/78.3.458ArticlePubMedPMC
• 19. Ejiri M, Kawai H, Fujiwara Y, Ihara K, Watanabe Y, Hirano H, et al. Social participation reduces isolation among Japanese older people in urban area: A 3-year longitudinal study. PLoS One. 2019;14(9):e0222887. https://doi.org/10.1371/journal.pone.0222887Article
• 20. Yang X, Xu XY, Guo L, Zhang Y, Wang SS, Li Y. Effect of leisure activities on cognitive aging in older adults: A systematic review and meta-analysis. Frontiers in Psychology. 2022;13:1080740. https://doi.org/10.3389/fpsyg.2022.1080740ArticlePubMedPMC
• 21. Teo AR, Choi H, Valenstein M. Social relationships and depression: Ten-year follow-up from a nationally representative study. PLoS One. 2013;8:e62396. https://doi.org/10.1371/journal.pone.0062396ArticlePubMedPMC
• 22. Cai J, Hu T, Zhou L, Jiang H, Gao Y. Effects of leisure activities and general health on the survival of older people: a cohort study in China. Frontiers in Public Health. 2023;11:1273074. https://doi.org/10.3389/fpubh.2023.1273074ArticlePMC
• 23. Lee HS, Shin WW. The Effect of Types of Leisure activities of elders on life satisfaction. Journal of Korea Culture Industry. 2016;16(3):157–165.
• 24. Kim JO, Um IS. The effects of leisure types and social supports on elderly depression. Journal of the Korean Home Economics Association. 2007;45(4):29–42.
• 25. Lee K, Nam SI. Types of leisure lifestyle among urban older adults: Analysis of differences in leisure satisfaction and depression by type and factors influencing typology. Korean Journal of Gerontological Social Welfare. 2024;79(3):241–267. https://doi.org/10.21194/kjgsw.79.3.202409.241Article
• 26. Reinwarth AC, Wicke FS, Hettich N, Ernst M, Otten D, Brähler E, et al. Self‑rated physical health predicts mortality in aging persons beyond objective health risks. Scientific Reports. 2023;9 13(1):19531. https://doi.org/10.1038/s41598-023-46882-7ArticlePubMedPMC
• 27. Yesavage JI, Sheikh JA. 9/Geriatric depression scale (GDS): Recent evidence and development of a shorter version. Clinical Gerontologist. 1986;5(1-2):165–173. https://doi.org/10.1300/J018v05n01_09Article
• 28. Park HJ. Cognitive diagnostic tests for dementia in primary care. Korean Journal of Family Practice. 2024;14(3):116–120. https://doi.org/10.21215/kjfp.2024.14.3.116Article
• 29. Rommerskirch-Manietta M, Purwins D, Van Haitsma K, Abbott K, Rodrigues-Recchia D, Roes M. Assessing preferences for leisure activities of people receiving adult day services: A study protocol for concept mapping and psychometric testing. BMJ Open. 2021;11(11):e055069. https://doi.org/10.1136/bmjopen-2021-055069Article
• 30. Henry D, Dymnicki AB, Mohatt N, Allen J, Keelly JG. Clustering methods qualitative data: A mixed methods approach for prevention research with small samples. Prevention science. 2015;16(7):1007–1016. https://doi.org/10.1007/s11121-015-0561-zArticlePubMedPMC
• 31. United Nations. Designing household survey samples: Practical guidelines. [Internet]. New York: United Nations Statistics Division; 2005 [cited 2025 May 2]. Available from: https://unstats.un.org/unsd/demographic/sources/surveys/handbook23june05.pdf
• 32. Heeringa SG, West BT, Berglund PA. Applied survey data analysis. 2nd ed. New York: Chapman & Hall/CRC Press; 2017. p. 590.
• 33. Lohr SL. Sampling: Design and analysis. 3rd ed. Boca Raton, FL: Chapman & Hall/ CRC Press; 2022. p. 674.
• 34. Lee HY, Jang SN, Lee S, Cho SI, Park EO. The relationship between social participation and self-rated health by sex and age: A cross-sectional survey. International Journal of Nursing Studies. 2008;45(7):1042–1054. https://doi.org/10.1016/j.ijnurstu.2007.05.007ArticlePubMed
• 35. Van Willigen M. Differential benefits of volunteering across the life course. The Journals of Gerontology. Series B. 2000;55(5):S308–S318. https://doi.org/10.1093/geronb/55.5.s308Article
• 36. Cruwys T, Haslam SA, Dingle GA, Haslam C, Jetten J. Depression and social identity: An integrative review. Personality and Social Psychology Review. 2014;18(3):215–238. https://doi.org/10.1177/1088868314523839ArticlePubMed
• 37. Dadswell A, Wilson C, Bungay H, Munn-Giddings C. The role of participatory arts in addressing the loneliness and social isolation of older people: A conceptual review of the literature. Journal of Arts and Communities. 2017;9(2):109–128. https://doi.org/10.1386/jaac.9.2.109_1Article
• 38. Ryu J, Heo J. Relationships between leisure activity types and well-being in older adults. Leisure Studies. 2018;37(3):331–342. https://doi.org/10.1080/02614367.2017.1370007Article
• 39. Kawachi I, Berkman L. Social cohesion, social capital, and health. In: Berkman L, Kawachi I, editors. Social Epidemiology. 3rd ed. New York: Oxford University Press; 2000. p. 174–190.
• 40. Kim S. Leisure activities and digital technology use among older Koreans. Health and Welfare Policy Forum. 2024;11:24–37. https://doi.org/10.23062/2024.11.3Article
• 41. Roh HW, Hong CH, Lee Y, Oh BH, Lee KS, Chang KJ, et al. Participation in physical, social, and religious activity and risk of depression in the elderly: A community-based three-year longitudinal study in Korea. PLoS ONE. 2015;10(7):e0132838. https://doi.org/10.1371/journal.pone.0132838ArticlePubMedPMC
• 42. Dunphy K, Baker FA, Dumaresq E, Carroll-Haskins K, Eickholt J, Ercole M, et al. Creative arts interventions to address depression in older adults: A systematic review of outcomes, processes, and mechanisms. Frontiers in Psychology. 2019;9:2655. https://doi.org/10.3389/fpsyg.2018.02655ArticlePubMedPMC
• 43. Min J, Ailshire J, Crimmins EM. Social engagement and depressive symptoms: do baseline depression status and type of social activities make a difference? Age and Ageing. 2016;45(6):838–843. https://doi.org/10.1093/ageing/afw125ArticlePMC
• 44. Hackney C, Sanders G. Religiosity and mental health: A meta-analysis of recent studies. Journal for the Scientific Study of Religion. 2003;42(1):43–55. https://doi.org/10.1111/1468-5906.t01-1-00160Article
• 45. Larson EB, Wang L, Bowen JD, McCormick WC, Teri L, Crane P, Kukull W. Exercise is associated with reduced risk for incident dementia among persons 65 years of age and older. Annals of Internal Medicine. 2006;144(2):73–81. https://doi.org/10.7326/0003-4819-144-2-200601170-00004ArticlePubMed
• 46. Sofi F, Valecchi D, Bacci D, Abbate R, Gensini GF, Casini A, Macchi C. Physical activity and risk of cognitive decline: A meta-analysis of prospective studies. Journal of Internal Medicine. 2011;269(1):107–117. https://doi.org/10.1111/j.1365-2796.2010.02281.xArticlePubMed
• 47. Buchman AS, Boyle PA, Yu L, Shah RC, Wilson RS, Bennett DA. Total daily physical activity and the risk of AD and cognitive decline in older adults. Neurology. 2012;78(17):1323–1329. https://doi.org/10.1212/WNL.0b013e3182535d35ArticlePubMedPMC
• 48. Park SA, Lee AY, Park HG, Lee WL. Benefits of gardening activities for cognitive function according to measurement of brain nerve growth factor levels. International Journal of Environmental Research and Public Health. 2019;16(5):760. https://doi.org/10.3390/ijerph16050760ArticlePubMedPMC
• 49. Corley J, Pattie A, Deary IJ, Cox SR. Gardening and cognitive ageing: Longitudinal findings from the Lothian Birth Cohort of 1921. Journal of Environmental Psychology. 2024;97:102361. https://doi.org/10.1016/j.jenvp.2024.102361Article
• 50. Cunha C, Voss G, Andrade R. Is formal social participation associated with cognitive function in middle-aged and older adults? A systematic review with meta-analysis of longitudinal studies. Behavioral Sciences. 2024;14(4):262. https://doi.org/10.3390/bs14040262Article
• 51. Piolatto M, Bianchi F, Rota M, Marengoni A, Akbaritabar A, Squazzoni F. The effect of social relationships on cognitive decline in older adults: an updated systematic review and meta-analysis of longitudinal cohort studies. BMC Public Health. 2022;22(1):278. https://doi.org/10.1186/s12889-022-12567-5ArticlePubMedPMC
• 52. Fancourt D, Steptoe A. Television viewing and cognitive decline in older age: Findings from the English Longitudinal Study of Ageing. Scientific Reports. 2019;9(1):2851. https://doi.org/10.1038/s41598-019-39354-4Article
• 53. Gow AJ, Corley J, Starr JM, Deary IJ. Which social network or support factors are associated with cognitive abilities in old age? Gerontology. 2013;59(5):454–463. https://doi.org/10.1159/000351265ArticlePubMed
• 54. Kim D, Arai H, Kim S. Social activities are associated with cognitive decline in older Koreans. Geriatrics & Gerontology International. 2017;17(8):1191–1196. https://doi.org/10.1111/ggi.12861Article

Figure & Data

References

Citations

Citations to this article as recorded by