The Association between Leisure Activities and Cognitive Functioning of the Elderly in Hong Kong (HK) and Guangzhou (GZ)

Background: Hong Kong (HK) and Guangzhou (GZ) are two major cities in southern China, and they share very similar ethnic, genetic, health and demographic characteristics. However, the two cities have been run with different social systems over the past few decades. This provided a natural case-control experiment for studying the effect of the social context on cognition. Hence, this study examined the association between leisure activity participation and cognition in the two cities to evaluate the cognitive modulating effects of leisure activities in different social environments.;Objectives of the studies: The main study objectives were to compare the cognitive characteristics and leisure activity participation of the two groups; to examine the association between leisure activity participation and cognitive function and the specific associations in HK and GZ; and to explore the modulating effect of social factors on cognitive function.;Methodology: This was a cross-sectional study. Convenience sampling was used to recruit 557 participants aged 60 years and over without dementia. Of these, 260 were recruited in HK and 297 in GZ. The two groups were recruited with similar demographic characteristics (age, gender and education). Leisure activities were classified as physical, intellectual, social and recreational activities. Leisure activity participation was measured in terms of the total number and total hours of participation per week for each category of activities. A battery of cognitive tests including the Cantonese version of the Mini Mental State Exam (CMMSE), word list learning test, delayed recall test, Category Verbal Fluency Test (CVFT), trail making test, digit cancellation test and Stroop test were used to measure participants' cognitive function.;Differences in the participants' demographic characteristics, cognitive performances and leisure activity participation were computed. A multiple linear regression of cognitive performance on leisure activity was performed, while controlling for other categories of activities and potential confounders that were significantly associated with cognitive function.;Results: The multiple linear regression revealed that living arrangement had a significant positive association with the total number of leisure activities (p=0.01) and total hours of leisure activity participation (p=0.02). Analysis of covariance showed that participants in HK participated in more leisure activities than those in GZ, as measured by the total number of subtypes and hours per week, except total hours of recreational activities per week (p=0.07). No significant differences were found between the cognitive performances of the older persons in the two cities. Pearson's correlation and x2tests were performed to identify the leisure activities and potential confounders that were significantly correlated with cognitive performance. The total number and total hours of intellectual activity were significantly correlated with CMMSE scores (p<0.001 and p<0.001).The total number of subtypes and total hours per week of intellectual activity (p<0.001 and p<0.001), social activity (p<0.001 and p<0.01) and recreational activity (p<0.001 and p<0.01) were significantly correlated with the word list learning test. The total number of physical activities (p<0.01), total number of intellectual activities (p<0.001) and total hours of intellectual activity (p<0.01) were significantly correlated with the delayed recall test. The total number of physical and intellectual activities (p<0.01 and p<0.001), and total hours of intellectual and social activity (p<0.01 and p<0.001) were correlated with the CVFT. The total number of intellectual activities (p<0.01)and total hours of recreational activity (p<0.01)were significantly correlated with the trail making test (p<0.001). The total number and total hours of intellectual activity were significantly correlated with the digit cancellation test (p<0.001 and p<0.001). The total number and total hours of intellectual activity were significantly correlation with the Stroop test (p<0.01 and p<0.001).;Multiple linear regression using the enter method was conducted to measure the association between leisure activities and cognitive performance. The results showed that the total number of intellectual activities was significantly associated with better performance on cognitive tests, including the CMMSE (p<0.001), word list learning test (p<0.001), delayed recall test (p<0.001), CVFT (p<0.001) and digit cancellation test (p=0.01). Total hours of recreational activity was significantly associated with the trail making test (p=0.01). Multiple linear regression using the enter method also revealed that marital status was significantly associated with the CMMSE (p=0.002), word list learning test (p=0.003), delayed recall test (p=0.002), trail making test (p<0.001) and digit cancellation test (p=0.01).;Conclusions: HK participants participated in more leisure activities than GZ participants. However, HK participants did not show better cognitive performance than GZ participants. This finding appears to be inconsistent with previous studies that found that participation in more leisure activities predicted better cognitive functioning. This inconsistency might be explained by socio-demographic differences between the two cities. (Abstract shortened by UMI.).