Print This PostIn my last blog I gave a background to population studies and an overview of why they have been an important part of dementia and ageing research. In this second part I wanted to give an example of a population study I have had a lot of experience with, The Cognitive Function and Ageing Studies (CFAS), and some examples of the most important and exciting findings to have come out of it.
The Cognitive Function and Ageing Studies
CFAS are two multi-centre population-representative cohort studies they are identically designed and sampled two decades apart, CFAS I in 1991 and CFAS II in 2011. Specifically, to investigate dementia epidemiology over time. They recruited around 7500 people aged 65 or above from Newcastle, Nottingham, and Cambridgeshire in the UK, from both the community and from institutions such as care homes. CFAS also had a brain donation programme with around 550 brains donated. It can currently be accessed via the Dementia Platforms UK archive.
Dementia Prevalence and Incidence
Once CFAS II had been completed it was possible to compare how dementia prevalence had changed over time and a generation of older people. If using CFAS I age and gender-specific estimates of dementia prevalence, it would be projected that 8.3% (or 884,000) of the population aged 65 and over would have dementia in 2011. However, CFAS II showed a prevalence of only 6.5% (or 670,000) – a decline of 1.8%. The odds ratio of having dementia in CFAS II vs CFAS I was 0.7, meaning dementia risk had declined in the population over time. The estimates on prevalence were surprising at the time given that the consensus was that prevalence was expected to increase, however when the CFAS II 2-year follow-up was analysed there was a 20% drop in incident (new cases) dementia compared to CFAS I. This paper was also interesting in that it showed that by far the biggest driver in the decrease was seen in men, whose cardiovascular health had improved significantly.
Mild Cognitive Impairment (MCI)
As well as surprising dementia results, population studies have been key in the often controversial evolution of pre-clinical dementia, most often termed MCI. After establishing cohort effects in dementia prevalence, it was becoming increasingly important to see if MCI showed the same decrease in prevalence over time. A study from the cohort showed that when using the most common clinical criteria for MCI prevalence was unchanged over 20 years, however when measuring cognitive impairment generally in the population, prevalence had increased over time. This was again backed up when CFAS was combined with a number of international population studies forming the COSMIC consortium, which reiterated the instability of MCI diagnostic criteria. In different studies prevalence ranged from 3% – 12%. Studies like this have implications for how we approach dementia clinically, if MCI is to be used to identify those in the population at high risk of dementia, then when diagnostic criteria are taken out of the clinical setting and applied to the population it produces highly unstable estimates.
Neuropathology
As well as the more clinical and social side of dementia research, population studies also have an important role in more traditional lab based dementia research. Again using CFAS as an example the donations made to the brain bank have been able to give insight on dementia pathology on a level that can be generalised to the population.
The epidemiological neuropathology perspective ensures an unbiased assessment of the prevalence of different brain pathologies and their associations with late-life dementia. Traditional case-control studies generally focus on younger clinical cohorts, while population-based studies like CFAS provide a broader understanding of cognitive impairment and its relation to ageing.
A brief summary of the key population neuropathology findings have been:
There is a High Prevalence of Pathology in the Ageing Brain:
Studies, including CFAS, have shown a high prevalence of pathological changes in the ageing brain. Alzheimer’s disease pathology and vascular pathology are particularly predominant.
Increasing Burdens of Pathology is Associated with Dementia:
As the burdens of pathologies such as Alzheimer’s disease and vascular pathology increase, so does the risk of dementia. Neurofibrillary tangles are most strongly correlated with dementia in the population.
There is Overlap in Dementia Pathology:
There’s a significant overlap in brain pathology between individuals with and without dementia. This overlap is particularly evident at intermediate burdens of pathology and becomes even more pronounced at the oldest ages.
Emergence of Additional Pathologies:
Recently discovered pathologies, such as limbic-predominant age-related TDP-43 encephalopathy, ageing-related tau astrogliopathy, and primary age-related tauopathies, also contribute to late-life dementia.
These findings provide a nuanced understanding of dementia pathology within the broader community. The high prevalence of neuropathological changes and its inconsistent correlation with dementia raises questions about disease definitions, suggesting the existence of other factors that modulate cognitive outcomes.
Population approaches highlight that late-life dementia is a complex disorder, often characterized by mixed pathologies. This complexity underscores the need for a comprehensive understanding of these processes throughout an individual’s lifespan. Gaining a deeper insight is crucial for identifying the best research avenues to minimize the risk of avoidable clinical dementia syndromes in later life.
Population studies are vital for providing a holistic understanding of cognitive impairment and dementia in ageing populations. These insights highlight the complexity of dementia, emphasizing the need for continued research on its various pathologies and the potential factors that could mitigate its risks.
I hope you have enjoyed these two parts on population studies and dementia. I used CFAS as an example simply because I am most familiar with it. I definitely encourage anybody to go out and see what population studies are out there and whether the wealth of data could be of use to your own research!
Dr Connor Richardson
Author
Dr Connor Richardson is a Neuro-epidemiology Research Associate in the Newcastle University Population Health Sciences Institute. Connor is the research statistician for the Cognitive Function and Ageing studies (CFAS) multi-centre population cohort. His research interest lies in using advanced statistical modelling and machine learning to measure dementia risk. Connor blogs about his research, Equality, Diversity and Inclusion and sometimes his Pomapoo’s.
Dr Clíona Farrell
Dementia Researcher
Adam Smith
Rebecca Williams
