The role of physical activity and bilingualism in the development of neurodegenerative disorders: Cross-sectional and neuroimaging evidence

Delaying Alzheimer’s disease (AD), the most common form of dementia, by five years could decrease the global prevalence of AD by 57% and halve the annual economic impact which is currently estimated to surpass US$2 trillion by 2030. Since no treatment or cure for dementia exist, identifying modifiable factors to reduce the incidence of dementia has become a public health priority. Increased physical activity (PA) has been associated with a lower risk of developing dementia in observational studies. Observational studies have also linked bilingualism (the ability to speak two languages) with a delayed onset of dementia but no risk-reduction in dementia in bilinguals relative to monolinguals. Differences in study outcomes in the fields of PA- and bilingualism- related research to methodological limitations including poor measurement of the exposure (PA and language profiles), small sample sizes, and recruitment of participants with different dementia etiologies. The purpose of this thesis was twofold: i) to explore the roles of PA and bilingualism in dementia risk and ii) to inform the development of a randomized controlled trial (RCT) to test whether studying a foreign language combined with increasing PA can improve cognitive performance in seniors who are at a higher risk of developing AD. The aim of Chapter two was to review the available evidence linking PA with the risk of developing dementia as well as to explore the effects of increasing PA on cognition in individuals with dementia. Results showed that aerobic, and high-intensity, habitual PA was associated with improved cognition-related biomarkers and lower dementia risk in epidemiological studies. Experimental evidence showed increasing PA improved cognition-related biomarkers and cognition in preclinical phases of dementia, but not in clinical phases. The findings showed that PA is linked with a lower risk of dementia in epidemiological studies, but experimental studies showed little to no improvements in cognition in participants with dementia following a structured PA program. There was evidence indicating that increasing PA levels in the preclinical phase of AD may result in greater translation impact than in participants at the more advanced clinical stage of AD. Most studies assessed PA with self-report measures questioning the accuracy and precision of exposure and recruited participants with dementia irrespective of aetiology, which makes it problematic to discern whether PA is differentially related to varying dementia aetiologies. The aim of Chapter three was to systematically review the association between bilingualism and the delay in the diagnosis of dementia and AD. Here, we retrieved a total of 20 studies, 15 of which were meta-analysed. Results showed that bilinguals were on average 3.2 (95% CI: 1.5, 4.9) years older than monolinguals at the time of dementia. Moreover, at the time of dementia diagnosis, bilinguals and monolinguals demonstrated a similar level of global cognitive impairment (Hedges’ g = 0.05 95% CI: -0.10, 0.21). Prediction intervals however showed a large dispersion of effect sizes in the meta-analysis comparing monolinguals to bilinguals on the age of dementia diagnosis. To explore possible reasons for the observed dispersion in effect sizes, we conducted subgroup analyses. In one subgroup meta-analysis comparing studies that had recruited participants with dementia to studies that had recruited participants with AD, bilinguals were 4.2 (95% CI: 2.0, 6.2) and 1.7 (95% CI: -1.4, 4.7) years older than monolinguals at dementia and AD diagnosis, respectively. Meta-analytic results combining prospective longitudinal studies showed no risk reduction in dementia among bilinguals compared to monolinguals (Odds Ratio: 0.85; 95% CI: 0.69-1.05). Risk of bias assessment revealed that most studies carried several methodological limitations including poor measurement of participants’ language profiles and small sample sizes. The aim of Chapter four was to explore the underlying mechanisms in the brain that may be responsible for the observed findings in the first systematic review (Chapter three). In this study, we observed that bilinguals compared to monolinguals had greater brain volume in the frontostriatal and frontoparietal circuits. Also, functional neuroimaging studies showed that bilinguals made use of relevant brain areas more efficiently than monolinguals when completing interference cognitive tasks. Results from the cross-sectional studies showed that higher levels of language acculturation were associated with significantly greater verbal and psychomotor speed performance than lower levels of language acculturation. The aim of the Chapter five was to explore the link between language acculturation and cognition in older individuals from ethnic minorities (Hispanic and Asian) living in the United States of America using an epidemiological dataset. In this cross-sectional epidemiological study, we analyzed data from the National Health and Nutrition Examination Survey using a larger sample size than previous studies. We found that higher levels of language acculturation (i.e. speaking the native language and that of the recipient’s country at home) was associated with greater psychomotor speed processing than lower levels of language acculturation (mostly speaking the native language at home) and some, but not all, measures of verbal fluency. Overall, the findings suggest that higher levels of language acculturation are associated with greater cognitive performance in older individuals from ethnic minorities. Overall, the evidence gathered in the previous chapters indicate that i) increasing PA in individuals who are at a higher risk of developing AD might be more useful in improving cognitive performance than in individuals who already have developed AD and ii) bilingualism might render the brain areas typically affected by AD such as the frontostriatal and frontoparietal brain circuits more resilient against neurodegeneration and in turn, delay the onset of AD symptoms and diagnosis. Therefore, because no randomized-controlled trial (RCT) testing the combined effects of increased PA with studying a foreign language currently exist, Chapter five presents a detailed study protocol for an RCT addressing this gap in the literature while addressing the limitations of previous studies in the fields of PA- and bilingualism-based research. The purpose of this thesis was to explore the roles of PA and bilingualism in dementia onset and risk and to inform the development of a randomized controlled trial (RCT) testing the effects of studying a foreign language with increased PA in individuals at a higher risk of dementia. Increasing PA levels are associated with greater cognition in individuals at the preclinical phase of AD rather than in participants with a diagnosis of dementia or AD. Bilingualism was also associated with later age of AD diagnosis on average by 4.7 years. This finding is clinically relevant because a five-year delay in the onset of AD could lower the number of individuals with AD worldwide by 57% and as a consequence, halving the associated economic costs. Moreover, we also showed that bilingualism may be responsible for rendering brain areas typically affected by AD more resilient against neuropathology. Moreover, this thesis revealed that studies in the field of exercise science and bilingualism research in dementia carry important methodological limitations that question the internal validity of these two lines of research. Consequently, the evidence gathered within this thesis led us to propose a novel RCT exploring the effects of increasing PA levels and studying a foreign language in monolingual individuals at a higher risk of AD while addressing the most important limitations of previous research.