Memorandum by Professor Carol Brayne
INTRODUCTION
This written contribution draws on experience
as a lead principle investigator in longitudinal studies of older
people over nearly two decades, as well as medical and public
health training. I have consulted colleagues from the main study
for which I am responsible, the MRC funded Cognitive Function
and Ageing Study (MRC CFAS).
KEY FINDINGS
FROM MRC CFAS
We have reported from this multicentre study
in England and Wales that:
prevalence and incidence of disability,
cognitive impairment and dementia rise exponentially after the
age of 65;
there is no down turn at the oldest
age groups;
there is variation across the centres
in disability but not in dementia;
pathology found in the brain after
death at advanced age is found in both those who are cognitively
impaired during life and those who are not and is mostly a mixture
of vascular and Alzheimer's type;
substantial proportions at each age,
including the oldest age groups, report good health despite increasing
health problems;
very high proportions are demented
in the period before death, particularly in the oldest age groups;
potential genetic risk for Alzheimer's
disease appears attenuated by the time most people reach old age
probably because of the high background risk to the whole population.
The bibliography and design of CFAS are described
in more detail at www-cfas.medschl.cam.ac.uk.
FRAMEWORK FOR
ASSESSING POTENTIAL
OF RESEARCH
FINDINGS
It can be helpful to see research findings as
fitting into a prevention framework. Each of these types of prevention
has a different impact at the population level (from highly effective
to minimal) but all are important.
Primary prevention (reducing "risk"
exposure so that pathology does not develop). This type of prevention
has the greatest potential yield if applied to populations. There
are a great deal of observational data from epidemiological studies
which suggest that certain types of lifestyle/opportunity (at
all lifestages) are associated with less chronic disease and health
active life expectancy. Bringing such results into feasible trials
of intervention at appropriate level (individual, social network,
community, high level policy) which generate sound data are far
less commonand tend to be done at individual rather than
community levels, lowering population impact.
The Wanless report has identified such public
health research as an area for active development and this needs
to be carefully encouragedwith a build up of the expertise
necessary and an environment that facilitates research. Such research
has to be built on firm disciplinary research, but then brought
together into the complex intervention framework (well work up
by the MRC). This type of research tends to take a long time and
not result in multiple publications in high ranking journalsso
needs positive nurturing. Current assessment and funding systems
do not encourage such long term high risk investment for researchers,
research groups, universities and funders. Some areas potentially
important to health, but not traditionally researched as such,
have been identified in the imminent ESRC call, but with limited
funding and include the role of built environment (housing, transport,
the way the built environment makes us active or inactive), intergenerational
aspects, civic society, crime and disorder, accessibility and
financial security. The new clinical networks being set up within
the NHS for individual and groups of disorders have the potential
to include only traditional therapeutic trialsit is important
not to miss other types of intervention. Research which aims to
investigate population behaviour at any age could have knock on
effects for each cohort as future health is a function of current
and past health.
Secondary prevention (early detection
and alteration of natural history through early intervention).
This has the potential to modify population health to some extent,
if all the criteria necessary for screening are met. Not many
interventions have been tested on older age groups, and this will
continue to be an area of important research. The MRC has funded
research into a trial of screening in the older population, which
is now yielding results (which, importantly, can have negative
findings as well as positive). Such trials have the same problems
as interventions in primary prevention but their feasibility on
a very large scale is proven, provided that populations can continue
to be recruited through the primary care setting. The UK can provide
amongst the best evidence internationally because of this ability
to recruit from clearly defined populations.
Tertiary prevention (prevention of the
consequences of disease once established). This type of research
has the least potential impact on population health. There is
insufficient follow-up on new and costly introductions of interventions,
once short term trial evidence is released, such as the dementia
treatment drugs, to see what real impact they have on the natural
history of morbidity and ultimately mortality.
DEATH AND
DYING
In the understandable concentration on healthy
ageing and reduction of morbidity at given ages there has been
less attention to the consequences of prevention earlier in life,
and its impact on future populations reaching the oldest age groups.
The relationship of changing health and wellbeing to survival,
with quality of life and death has been less researched but is
bound to receive increasing attention as the whole population
ages. Dementia or severe cognitive impairment is very common in
the period before death in the oldest age groups. This is unlikely
to be prevented in the near future, if ever and we need to understand
this stage of life too. A firmer foundation of research is needed
to prepare for this societal change, including the frail stages
of ageing not just "healthy active" life.
MODELLING POPULATION
HEALTH
Work on modelling population health profiles
including more detail in older age groups given changes in younger
age cohorts is not highly developed and is an area of potential
value. Although many are sceptical about the value of modelling
it is valuable for policy assessment, particularly if carried
out with sensitivity analyses and explicit discussion of uncertainty.
This has been identified as an area by the Wanless report, and
the MRC has asked HEIs to bid for dedicated doctoral studentships.
Continuing support is needed to build up this area of expertise
in the UK.
EVIDENCE BASE
FROM COMPARATIVE
STUDIES
Encouragement of a diversity of approaches to
ageing research is sensible as each can reveal valuable insights
which can be tested from a different angle in another (disease,
system, lifestage, holistic, discipline based approaches, single
and multidisciplinary), not least to provide sound evidence for
modelling. Major gaps for which there are currently no large-scale
research efforts are epidemiological studies examining variation
in health across ethnic groups and cohorts in the oldest age groups.
Such studies are most valuable if they have a robust population
base and if they can have a set of more detailed biomedical and
qualitative studies bolted on. MRC CFAS, amongst others, can provide
a model for this type of approach, which can then be used to address
biological questions (such as why rates of dementia might vary
across populations) as well as policy questions (do we expect
different patterns of ageing as our ethnic populations move into
the older age groups, are future older people likely to be more
healthy or less than now). These questions can only be explored
with stable methodology and population sampling. The opportunity
does exist but has not been taken up.
BARRIERS TO
RESEARCH IN
HUMAN POPULATIONS
There are now major barriers to research beyond
achieving funding including the many levels of permission and
engagement required to work with populations, from research governance
structures (multiple within the NHS), ethical review and the legislative
framework such that more time is spent in these processes than
in the research itself.
September 2004
|