Our study aimed to enhance understanding of memory decline (MD) in old age by evaluating longitudinal effects of personal and national contributing factors. We used data collected by the Survey of Health, Ageing and Retirement in Europe (SHARE) from 12 European countries and Israel. Our sample included 11,930 retirees aged 50+, interviewed at baseline and four years later. MD was evaluated by the change in the number of recalled words from first to second interview. Except for gender, all of our explanatory variables had a significant unique effect on MD – age, education, type of occupation, European geographical region, early retirement, time elapsed from retirement, reason for retirement, active lifestyle, re-employment, health/function status, depressive symptoms, and decline in physical and mental health – over the four years of the study. Our findings indicate that MD can be postponed by national policies such as those which prolong years of education and participation in the workforce, and by social interventions directed to promote active lifestyles in late life, especially in Mediterranean and Eastern European nations.

Background: We evaluated the combined and differential effects of physical, social and productive activities on cognitive decline and whether they were modified by the presence of the APOE-ε4 allele.

Methods: In a prospective cohort study of 1635 community-dwelling Chinese older adults aged 55 or older participating in the ongoing Singapore Longitudinal Aging Study, physical, social and productive leisure activities were assessed at baseline, and cognitive decline (at least one point drop) in MMSE scores between baseline and follow-up after one year.

Results: Cognitive decline was observed in 30% of the respondents. Controlling for age, gender, education and other risk factors, odds ratios (ORs) were significantly reduced in those with medium (OR: 0.60, 95% CI: 0.45–0.79) and high activity levels (OR: 0.62, 95% CI: 0.46–0.84). A stronger association was shown for productive activity (OR = 0.36), than for physical (OR = 0.78) and social activities (OR = 0.85). These associations showed statistically significant interactions with APOE genotype, being more pronounced in those with the APOE-ε4 allele.

Conclusion: Increased leisure activity, especially productive activities more than physical or social activities, was associated with a lowered risk of cognitive decline. APOE-ε4 genotype individuals appeared to be more vulnerable to the effects of low and high levels of leisure activities.