By me and Steve Laitner
I've (GF) blogged a lot on the myth that is the "demographic timebomb", and set out a set of data and arguments to make up case argument wise. See references at the bottom.
Here are our collected summary thoughts on why getting the narrative wrong on this is harmful
Lessons in demography
Despite popular conceptions, the age distribution of the population shifts very slowly from one year to the next. The population does grow, but the distribution of age shifts slowly. This slow growth in the age distribution of the population limits the magnitude of the impact on utilisation.
Many perceive that the impact of ageing during the next ten years will be higher than it was for the previous ten years; this perception is incorrect.
Strunk et al estimate that ageing increased inpatient utilisation by 0.35% per year from 1995 to 2005. One can also assess the changing impact of ageing by comparing annual increases in inpatient utilisation for selected one-year periods. For 1994–95, 2004–05, and 2014–15, the ageing factors are 0.33%, 0.63%, and 0.80%, respectively
Ageing therefore accounts for a relatively small portion of the growth in hospital spending projected for the next decade. One USA estimate suggested only 11.8 percent of the total increase in inpatient spending from 2005 to 2015. (Ref 10 from Strunk)
In this country demand is going up year on year – c4%. The population age profile is shifting considerably more slowly than this.
To prove "the ageing population timebomb myth", you need to:
look at the data and who is actually using healthcare, and other factors
A&E – generally NOT older people
it's important to consider the a) absolute number b) length of stay (perhaps older people for LOS but for admissions) and c) where the increases are in utilisation d) absolute numbers in a population, especially in e) a fixed capacity system
– you then need to demonstrate that it's AGEING per se that's driving this, not morbidity or demand side failures.
In general, the effect of ageing effect on use of inpatient services will be small, but it will have a larger impact on use by patients with certain types of medical conditions that are more concentrated among the elderly.
Finally, and arguably critically, although the AGE of the population is increasing is the burden of disease increasing (NO, what limited data there is would suggest not) or is it just that 72 or the new 70 etc and its just numbers on the clock and mean nothing (probably)
Given all that, are you sure you can demonstrate the ageing timebomb? No, we thought not.
In a nutshell …..Cost, and cost growth, is driven by
- morbidity – incident and prevalent disease. Incident disease is CONSIDERABLY more important here
- Low value expensive technology,
- over diagnosis and over treatment,
- unrealistic expectations,
- degradation of the demand management functions,
- over use of supply sensitive care.
- Preference sensitive care choices – an assumption is often made that people want the same care at 90 as they do at 40 – they don’t.
- End of life intensive tx.
- Proximity to death – whether that's at 69 or 99, it doesn’t really matter doesn't matter.
These drive health cost inflation and NOT "the ageing population".
A number of blogs on this are referenced at the bottom. They are fully referenced.
And also even IF the ageing population is a reason for increasing demand on healthcare surely the improving health of the population which allows people to live longer should be REDUCING the demand on healthcare for slightly younger population
…and if technology means that people are living longer shouldn’t technological advances enable us to LOWER the cost of healthcare such as through non face to face means
How much is “living with disease” life increasing (if at all)?
This IS, however, proving a very difficult zombie to kill especially as far as popular media and NHS policy are concerned.
Its a fantastic cop-out - “its not our fault its those bloody citizens having the audacity to live longer”!
Here are 5 reasons why it matters that we get this narrative right, and matters a lot.
- If we focus on “the ageing population” the wrong response becomes more likely. If we are fatalistic, and accept it's an inevitable consequence of an ageing population, we will prepare wrongly by building bigger hospitals to cope with demand, not preventing demand. Hospitals are certainly not what the older population want and in most cases need, consider them as expensive hotels that sometimes add value, but often they don't and can do harm. – as a subset of this, if we get the narrative wrong we will loose focus on healthy ageing (in old money that's called health promotion)
- We focus on what we CANT do much about, and we don't focus on what the problems driving demand actually are, ones that we CAN do something about – this becomes a sort of Moral Hazard and we just plead for more money as always. Take for example the recent Lancet commission on dementia. One of the points made in the accompanying editorial was that Dementia prevalence- halved if we delayed onset by 5y, furthermore this may be achieved with more assertive and ambitious mid-life interventn
- Stigmatising old people. It's ageist.
- It weakens the intergenerational, whole society compact that underpins the establishment of the NHS. Why should I pay for all those old folk who are expensive to care for.
- A reliance on "the ageing population" hypothesis entrenches (as if it needed further entrenchment!) inequity in resource allocation as it (unfairly) further weights age per se over real drivers of need. Given that 1) the distribution of resource in healthcare between hospital and out of hospital is in itself a determinant of health and 2) the distribution of resource within primary care is inequitable – this 'ageing timebomb hypothesis' makes inequality worse.
Change your narrative please.
Critique of these ideas
Of course this is all an unproven hypothesis.
It's worth saying that despite this, feedback from our GP friends on twitter – and in real life – has been consistently along the lines of ….. "yep, that about reflects my reality"
One person summarised the ideas as a bit fast and loose
One of us (GF) wasn't very happy with that. We referred the person to the assiduously referenced blogs at then bottom of this one, and pointed out we were indeed writing s blog not a PhD thesis!
Another suggested that the focus on multi morbidity was a bit medical model.
Fair critique. We could call it multi morbidity, we could call it loss of functional ability or ability to adapt, we could even call it health and well being.
We maybe should characterise it as all of the above.
However our point is that those things are not driven by "the ageing process" (which is obviously not modifiable but by extraneous events that often are, but we choose to ignore them at both an individual and a societal level.
Muir Gray has been writing about this for years now – #sod70
This takes us back to place we give to the concept of "healthy ageing" in policy terms versus "preparing for the demographic timebomb". Guess which one gets more airplay in public consciousness?
Lastly on demonstrating the hypothesis with data.
Tricky, may not be a doable thing. We don't think we're clever enough to do it.
- We think the analytic question is on of whether a robust AGE SPECIFIC measure of morbidity is changing over a long period.
- That measure would need to be adjusted for perverse screening programmes that lead to overdiagnosis, over treatment trends, insulated from the impact of changes in supply or demand side factors in the health and care system – i.e. Pure morbidity or functional ability / impairment.
- This would test whether 72 is indeed the new 70 or whether 68 is the new 70. Ideally this would also need to be data that can be cut along socio economic deprivation and other lines
Until such time as such analysis is available, our contention is that "the ageing pop" remains lazy shorthand for not addressing much more difficult issues
Further reflection and feedback from others
Is the focus on multi morbidity the right focus.
Our original blog said that it was multiple morbidity that is driving cost, and that morbidity a proxy for disability and also proximity to death
Alf Collins consistently points out that morbidity is a proxy for disability, which is a proxy for functional impact (see Commonwealth Fund)
See here – Lancet editorial What is health? The ability to adapt:
This has obvious implications for how we take forward the next steps.
Huber and colleagues went close to this in their 2011 article called "how should we define health" http://www.bmj.com/content/343/bmj.d4163
A subset of this is the "ahhh but it's all about frailty, and that's definately related to age" line
Not in our view.
There are many old folk who are "frail", no doubt. There are many old folk who are definately NOT frail.
There are plenty of younger folk who are also frail.
Our view is that frailty is not an "age thing" per se, but about the accumulation of stuff earlier on in life, and is full of modifiable risk factors,
See here on the framing of frailty – helpful https://britishgeriatricssociety.wordpress.com/2017/06/26/framing-the-narrative-of-frailty-differently-will-help-to-promote-wellbeing/amp/
Individual vs population approach to frailty and resilience
Of course individuals live in a social context. Many told me that it's actually about the social context in which individuals live. This leads to the question of whether we can build the resilience of communities to help individuals in those communities. Of course, is the obvious answer.
See this poster
Communities Fit for Frailty:
The Relationship between Community Resilience, Frailty, and the Use of Care
Wider still – social factors,……. is the need to take into account other changes in social structure are making life rather more difficult – no longer can rely on nuclear families to look after our old rellies, changing job market, melting of welfare state as we know it – and many other things all have profound effect
There are many who take a different line
Eg see here http://www.bmj.com/content/357/bmj.j2759.short?rss=1
Can this be summed up as:
Some older people stay for and well and active
We need to think of the latter as well as the former
The paper in the lancet was interesting.
Forecasted trends in disability and life expectancy in England and Wales up to 2025: a modelling study
The study is excellent and warrants a careful read. The cncousyiom line in the abstract – sadly only bit most people ever read – says "The number of older people with care needs will expand by 25% by 2025, mainly reflecting population
ageing rather than an increase in prevalence of disability. Lifespans will increase further in the next decade, but a quarter of life expectancy at age 65 years will involve disability."
There, said it – population ageing.
Actually the study is one that puts together disease indecence, prevalence, life expectancy, healthy life expectancy and the number of people of a certain age. All projected into the future.
The sheer increase in numbers of people over 65 (25% in 10 years) must be an issue but this is more demographic than increases life expectancy is it not, and an economic and political issue and offset to some extent by less younger people.
The increase in years with disability (the major cost driver) is small -about 0.7 years increase in 10 years?
I still insist that neither account for 4% per year on year increase in NHS demand and we must get out of that Moral Hazard cul-de-sac quickly.
"Population ageing" doesn't even get close. You've got to distinguish net population growth, changing in disease profile and ageing per se.
There were a bunch of other limitations and caveats. Picked out in both the main paper and the editorial. Obviously nobody likes long lists of caveats – folk have short attention spans and get bored. Suffice to say caveats are important!
That starts with really understanding the causes of increased demand and what we need to do about it!
It is interesting to note the recommendations of the authors
increased capacity in formal social care (making good cuts of last 7years?)
improved support for informal social care arrangements
enhanced interventions against predictable risk factors
Simple, but rather unpalatable stuff as we continue with our narrative around broken hospitals, must fix hospitals etc etc etc
So, all up……we (STILL) can't robustly, or with any credibility, pin the current problems in health and social care on the ageing pop etc
It's NOT about the ageing population – https://gregfellpublichealth.wordpress.com/2016/11/18/is-it-the-ageing-population-need-demand-or-supply/
It's NOT about the ageing population – volume 1 https://gregfellpublichealth.wordpress.com/2016/11/18/is-it-the-ageing-population-need-demand-or-supply/
the ageing population" conundrum volume 2 -https://gregfellpublichealth.wordpress.com/2016/03/25/the-its-all-about-the-ageing-population-conundrum-where-next/
prevention – ultimately futile in broad economic terms as folk will likely live longer and get dementia….discuss - https://gregfellpublichealth.wordpress.com/2016/08/24/prevention-lets-not-bother-as-itll-never-save-money-how-much-money-will-you-save-in-our-stp-by-next-week-fell-12/
The epidemiology of multi morbidity https://gregfellpublichealth.wordpress.com/2017/03/09/on-the-epidemiology-of-multi-morbidity/
The transformation issues we don't talk about – https://gregfellpublichealth.wordpress.com/2016/11/18/the-transformation-that-are-missing-from-current-iterations-of-thinking/
The GP 5YFV and the deep end. The importance of inequality. https://gregfellpublichealth.wordpress.com/2016/12/10/the-gp-5-year-forward-view-the-importance-of-inequality-and-the-deep-end/
Using routinely collected data to demonstrate where the NHS is going wrong https://gregfellpublichealth.wordpress.com/2016/12/09/using-routinely-collected-data-to-demonstrate-where-the-nhs-is-going-wrong/
Strunk B, Ginsburg P, Banker M. The Effect Of Population Aging On Future Hospital Demand. Health Affairs 25, no.3 (2006):w141-w149. http://content.healthaffairs.org/content/25/3/w141
Preparing for later life today – The Lancet – http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(17)31996-7/fulltext?rss=yes