Director Data Analytics
@HealthFdn
. Ex-Whitehall civil servant & NHSE.
Love ideas, numbers, analysis, debate, challenge. Own views esp when not about health.
The argument that’s often made for why we need an alternative model to the NHS or radical reform is that the NHS is unsustainable. The latest to make the argument is Sajid Javid. But *is* the NHS unsustainable? A Friday🧵
When, on 12 June, Elon Musk made Likes on X private (so that only a post author can see who likes their post) this seemed like a small change. But small changes in complex systems like X can have big, unpredictable effects. 1/7
So, just two months after the changes were made we may be seeing their horrific consequences. It seems incredible that we have allowed a billionaire owner of a social media platform to have such power. 7/7
This subtle change made by Musk seems likely to have fuelled the race riots which have erupted across England. I think it's also likely that Musk foresaw what the effects of his change could be. 6/7
This is speculation about the effects and consequences of the changes to Likes based on what I think is a plausible theory.
The theory needs testing with data, eg to understand whether certain types of tweets (eg offensive, racist ones) now get more likes than other types. 7+1
And, to clarify, I'm not saying Musk foresaw the race-related riots in England. The change to Likes seems motivated by his belief in unrestricted free-speech. But he's a clever man who will have been aware that the change could have far reaching (if unknown) consequences. 7+3
We also don't know to what extent X and engagement with racist posts fueled the riots. There's sure to be more research into this, helping fill gaps in our understanding. 7+2
Long ambulance response times and their devastating impacts on patients and families are frequently in the news. We wanted to get behind the stats and take a whole system view. A🧵based on our new analysis, which featured in a powerful ITV report last night . 1/13
Aside from other benefits, Sure Start reduced chance of hospital admission by 19% at age 11 in poorest areas. This is more than the effect of most NHS programmes whose *goal* is to reduce admissions. A case for rethinking how money is spent?
New IFS report finds
#SureStart
significantly reduced hospitalisations in children in poorer areas.
Gabriella Conti (UCL & IFS) says “Access to Sure Start can help close around half the gap in hospitalisations between rich & poor areas”
@NuffieldFound
NHS E are making A&E performance a competition with million pound cash prizes for the top ranked trusts.
This does not seem like a good incentive system.
Note that letter issued today (12 March) for performance in March.
@RichardJMurphy
I agree with this, but alongside increasing numeracy and real world maths skills I would add more general analytical and critical-thinking skills so people are better at analysing problems and examining claims.
Excellent from
@BBCRosAtkins
on the social care crisis. Concise, clear, fact-based and fair.
His summing up: "social care needs more money and reform and patients need two systems that can act as one"
If you're just getting going, you'll see more stories about the NHS crisis in the news. There are many elements to it - but the social care crisis is crucial to understanding what's happening in hospitals. Here's 6 mins on why.
Deprived areas have experienced many more deaths from Covid than richer areas. As vaccines are being targeted to reduce mortality you’d think that this would mean that more people would be being vaccinated in deprived areas. (1/5)
Our prospective Conservative MP's campaign leaflet, telling us she didn't put sewage in rivers. Pretty sure that this is not a message Comms experts would advise...
Overdue 🧵on social care charging reform. It's been more than 2 weeks since the govt announced delay, but I wanted to let my anger and frustration settle. I’ve spent much time & energy working on this issue so here are my thoughts. 1/22
Rachel Reeves says she will not take forward adult social care charging reform and the cap on care costs.
During the election campaign Wes Streeting said Labour were committed to the cap. 1/3
"One of the things we have committed to is the cap on care costs"
Shadow health secretary Wes Streeting agrees it's not "fair" that a viewer's mother has had to spend £430,000 during the seven years she has been living in a care home
#BBCLauraK
This, from X's director of engineering in May, confirms that the change was intended to incentivise people to like content that might be "edgy", by protecting their public image. 7+4
But a new funding model for the NHS won’t help and would create years of distraction. What’s needed is a longer-term view, a realistic vision and a resourced plan for achieving it. This would also give hope to the public and demotivated staff that things will get better. /end
Almost a year ago, the Lancet published estimates comparing excess deaths across countries. These, which were flawed and misleading, made news here because they suggested the UK had fared much better than had been thought (below from Daily Telegraph).
But having more to spend on other things whilst healthcare is growing as share of GDP depends on economic growth. This has been weak. So it’s our economic model rather than the NHS model that has failed us over the last 15 years.
On BBC R4 Today this morning, Wes Streeting said that patients who can’t get a GP appointment, costing £40, can end up going to A&E, costing £360. Can it be true that 10 minutes of a GP’s time costs so much less than 10 minutes of a hospital doctor’s time?
None of this is to say that the NHS can’t be improved and that improving its efficiency is not important. Not though, that NHS productivity does tend to increase year by year: since 2010 the productivity of the NHS has outperformed the rest of the economy.
The social insurance models of France and Germany are sometimes held up as examples of better. But health systems in both countries spend more and since 2010 have been better funded.
The NHS/govt sees the independent (private) sector as a way of expanding capacity for the 6.5m (and growing) NHS patients waiting for hospital treatment. A🧵on what’s been happening and implications. 1/10
How do you make a slow start to doubling the number of medical school places by 2031, sound impressive?
By saying you're growing capacity exponentially.
So what can be done? No silver bullet, but these things can help:
(1) Invest in hospital capacity eg beds & staff. It's not the time for infrastructure cuts (HMT pls note).
(2) Invest in social care & out of hospital care - so people leave hospital when they are able to. 9/13
As a nation we applaud NHS and care workers for their huge contributions during the pandemic.
But yesterday social care workers were left out of pay increases announced by the government. My thoughts below ⬇️ (1/7)
The opportunity to work for
@HealthFdn
creating this new centre is like a dream come true. I've loved working in partnership with them on the
#ImprovementAnalytics
unit; now I get to join them!
We know how the UK’s health compares to that of other countries* using life expectancy as a measure.
But less discussed is how our levels of health inequality compare? 🧵1/10
* below that of western European countries, ahead of US
There's no doubt the NHS is in crisis and is letting too many patients down. Underinvestment and COVID are partly to blame but the causes go deeper than that. Good solutions are desperately needed but changing the funding model is not one, and would be a costly distraction.
Three years ago Babylon GP at Hand's digital first model (with 13% of appmts conducted F2F) was being promoted by SofS Matt Hancock as the model of the future. In today's new plan for GPs, Govt says <20% F2F likely to be contrary to good clinical practice.
What problem is
@RishiSunak
's fine for missed appmts trying to solve? I think he believes that missed appointments are wasting time (that could be used for other patients). But if someone doesn't turn up, time is only wasted if doctors don't do something else valuable. 1/5
We are hearing horrific stories from patients and families about waits for ambulances, waits in A&E, and waits on floors/corridors for beds. These waits cause pain and suffering and delay treatment and care. How many people are dying as a result? A 🧵1/12
Spending more of our income on something we value is a good thing. We see it elsewhere. For example the share of our income going on culture and recreation is more than 50 years ago – is that unsustainable?!
(3) Invest in community services which can prevent hospital admission.
(4) Increase ambulance capacity eg by doing things which reduce ambulance staff sickness rates (highest across the NHS). But not a fix and note that without improving flow, ambulances will still queue. 10/13
On the eve of the latest
@ONS
weekly deaths publication, here's the Spanish version of
@EdConwaySky
's remarkable chart. 68% excess mortality in month to 18 April.
Here is the story of today's
@ONS
mortality statistics in a single chart. The worst week for UK mortality (all causes) since records began.
The blue area shows the minimum and maximum mortality each week since 2010.
The red line shows you this year so far.
There’s little point in training more doctors if they’re all fighting to get on a Windows XP that is ceaselessly restarting to install important updates.
My piece in today's
@guardian
The govt & Labour see the private sector as key to tackling the 7m+ NHS waiting list. The Elective Recovery Taskforce reports on this soon. What are the lessons from ophthalmology (eg cataracts) & orthopaedics (eg hips) - two specialties where the private sector is biggest? 1/14
Vaccinating all over 2.8m over 80s in England should reduce deaths by about 60%. Vaccinating all 10.4m over 65s should get deaths down by 90%. What's the plan for vaccination and how long will it take to deliver?
The COVID-19 vaccine hopefully marks the beginning of the end of the pandemic. As the vaccine can’t be given to everyone immediately, the govt's advisers, the JCVI, have set out the order in which the first 25 million people will be vaccinated.
Note that the mid 1990s were also a time when the NHS had had a period of low funding growth. Underfunding can allow the narrative that the model is broken to take hold.
BBC reports that excess deaths in 2022 were among the worst in 50 years. This is based on a crude comparison with deaths in 2019 (9% more) – not adjusting for changes in population size and age.
But there are other (in my view better) ways of looking at excess deaths/mortality.
Yesterday the govt announced £200m to get 2,500 people who are medically fit for discharge out of hospitals into care homes. The policy has been criticised by social care experts. I’ve been trying to make sense of the numbers.
40% reduction in emergency admissions from (some) housing improvements (tho mechanisms unclear)
Evaluations of *NHS service interventions* aimed at reducing emergency admissions frequently show no or negative impact, so perhaps the NHS is looking in the wrong place for ideas!
V significant, longitudinal study on long-term impact on
#NHS
admissions of
#housing
improvement. If you're looking for the (demand reduction) case for housing in your
#STP
&
#ICS
here it is (read the devil in the detail though) . Tx
@wouterpoortinga
et al.
Here’s Japan (the differences between the red and blue markers in 2021 and 2022 add up to 111k - their excess deaths estimate). It looks like they’ve estimated a level of expected deaths way below the trend line. Why? The paper doesn't explain. Was there a check on plausibility?
Astonishing that MPs are being asked to vote on the Care Act amendment without any sight of the financial savings and the impact on people. So great that Mel Stride, chair of Treasury Committee has written to Chancellor asking for this info pronto.
Long waits & overcrowding in A&E are often due to a shortage of hospital beds. But 14,000 (1 in 7) beds are occupied by patients fit to be discharged. With the pressure on hospital beds, why is this number so high and why has it been increasing? A 🧵based on new analysis. 1/15
Spending is rising because we are living longer. It’s rising because when we prolong life there are more people with long term conditions (in the past they were dead). These are a the result of the success of public health and healthcare.
Patient flow beyond hospitals is also a big challenge - with an increasing number of patients medically fit for discharge facing delays leaving hospital. Helpful exploration of this issue in this
@NuffieldTrust
QualityWatch analysis. 8/13
Comparing excess deaths across countries is an important way of understanding countries’ relative pandemic performance. The cross-country comparisons in 10 Mar Lancet/IHME excess deaths analysis are quite different to others (eg World Mortality Dataset).
The waiting list was plateauing before the strikes and has been rising since. But this is the usual annual pattern and can't be attributed to strikes. Pre-pandemic, the waiting list was rising year-on-year with a fall Sep-Jan and then a bigger rise.
Children’s & young people’s
#mentalhealth
has been worsening across the UK. Services are expanding but not by enough. Addressing the crisis requires understanding it. A🧵based on work with our Networked Data Lab 1/11
It’s hard to argue that NHS England’s A&E incentive scheme, which rewarded A&E performance in March with capital funding, is a good way of allocating capital. But did it succeed in improving A&E performance? The data and our analysis are in. 1/10
NHS E are making A&E performance a competition with million pound cash prizes for the top ranked trusts.
This does not seem like a good incentive system.
Note that letter issued today (12 March) for performance in March.
Against a long-term trend of rising hospital admissions, policymakers have had an enduring aim of slowing their growth. So it’s remarkable that between 2019 & 2022 admissions fell by 800k (12%). A thread. 1/16
Yesterday's NHS data shows that 59% of over 80s have had 1st vaccine dose but there is huge variation across STPs. 85% of over 80s in Gloucestershire, which may be approaching a ceiling, have been vaccinated compared to 36% in Suffolk and North East Essex. Why this variation? 1/5
Conclusion: protection against social care costs is a missing part of our welfare state. A capped cost model is a cheaper way of offering protection then providing free care to everyone (= cap of £0). In the 11 years since Dilnot no-one has come up with anything better. 21/22
Interesting review by Sir Andrew Dilnot and Michael Blastland of BBC economic coverage. Concludes not wilful bias but journalists' lack of understanding of basic economics that puts impartiality at risk.
The NHS could (like every system) achieve more health gain for the it money spends. E.g. from re-allocation of resources (eg towards primary care and prevention, infrastructure), and from better system design/mgmt (eg ambulances waiting because not enough beds is not efficient).
They did so because they had an emotional response: the results proved they were right (lockdown was bad, govt managed the pandemic well).
We are all prey to this. As
@TimHarford
says, when you see a claim first ask yourself, how does that make me feel?
So, Rule One: SEARCH YOUR FEELINGS.
What we believe, or refuse to believe, is strongly influenced by our emotional reaction. A lot of the statistical claims we see aren’t just data: they are weapons in an argument. Social media thrives on emotion. So do media headlines.
Not a health🧵
Ever since information about new voter ID requirements for elections came through our letterbox I’ve been curious about the rationale. So I’ve studied the Impact Assessment – the official explanation of objectives, costs and benefits. 1/20
An illustration. Assume health is 10% of GDP. Let’s assume GDP doubles over the next 30 years (respectable 2.2% pa) and health grows to 15% of GDP. Health spend will have tripled in size. But GDP available for non-health spend will have increased 1.9 times.
One more thing to add to the social care debate. The Dilnot Commission recs and Care Act legislation were based on people with the same needs reaching the cap at the same time. Under Govt proposals those with less wealth will take longer to reach the cap (and some never will).
JCVI criteria were meant to target vaccines at those at greatest risk of death. But more deprived communities in which covid death rates are higher, have younger populations and therefore lower rates of vaccination per head. It's the inverse care law, designed in.
First, a reminder of those response times stats. Category 1 incidents (life threatening): 7 mins pre-pandemic now 8.5 mins. But cat. 1 rightly prioritised so much bigger increases for other incidents. Cat. 2 emergency (e.g. stroke) up from 22 mins to 41 mins (target=18 mins) 2/13
As the problem won’t go away, Dilnot's proposals may one day be back on the table. But if they are, any weakness (<100% PM support, a lukewarm or fragmented social care sector) will once again be exploited by the Treasury. /end
NAO report on NHS New Care Models published today praises my team's evaluation approach and reports on our analysis of impact. Poll of vanguards put evaluation at the top!
@MichaelLawrie1
Ambulance services have little or no slack. This means small reductions in capacity can have major knock-on effects (as chance of no ambulance being available to respond rapidly increases). Similar to how closing a lane on a busy motorway can slow traffic to a standstill. 6/13
Work by the OECD, referenced in our blog on different funding models, shows there is no clear evidence that social insurance systems are more efficient than tax-funded systems.
Since 2013, the waiting list for planned hospital care has tripled in size to 7.75m and it’s still increasing. When will it peak, and what is the effect of continued strikes by doctors? We’ve just published analysis with scenarios and an interactive calculator. A 🧵1/18
How about capacity? More paramedics, but fewer available ambulances to respond because more waiting outside hospital to hand over patients. In Jul 2019 1 in 50 waited >60mins; in Jul 2022 1 in 10 (& average of 20% of ambulances waiting at any one time). 5/13
Whilst our pre-welfare state social care charging system continues to exist, people will continue to face a care cost lottery - with those with the greatest needs bearing the highest costs.
The cap could have been a first step towards Labour's National Care Service. 3/3
Why do handovers take so long? As with A&E waits, hospital capacity is an issue. Low number of beds per capita, recent reductions in bed numbers with increases in % occupied, mean fewer unoccupied beds. This gives very little flexibility to cope with spikes in demand. 7/13
Of course, at some point the NHS + other public spending may grow to such a large share of the economy that taxation strangles growth. But we’re some way off this (other countries with high economic growth have higher levels of taxation).
The COVID-19 vaccine hopefully marks the beginning of the end of the pandemic. As the vaccine can’t be given to everyone immediately, the govt's advisers, the JCVI, have set out the order in which the first 25 million people will be vaccinated.
@UKHouseofLords
has just voted against the proposed govt amendment to the Care Act, which seeks to change the way that people would progress towards the £86k social cap on social care costs.
A short thread on why this is important and why it's time for a re-think. 1/6
(1) is true. NHS spending has risen from c3% in 1948 to c8% now and is projected to continue to rise. This is a pattern we see in almost all countries (exceptions - countries with totalitarian regimes or in economic difficulty)
@GidMK
Counterfactuals. They are forecasts of what would happen if you do nothing. If you take action, of course reality will be different from the forecast. Doesn't mean the model producing the forecast is wrong.
("Half a million didn't die so Ferguson was wrong")
Statement today on social care funding fills in some of the detail that was missing from the Sept announcement. Lots to digest but what follows are first thoughts from
@so_says_sally
and me. More thoughts and analysis will follow from us both. 1/15
@ShaunLintern
@wesstreeting
Next step will be for NHSE to drop its relentlessly +ve spin, & be more open to challenge.
“We have improved against almost every headline objective of 2023-24 – we have either done what we set out to do, or made meaningful progress towards it.” - 24/25 planning guidance
@sib313
So what's driving the sharp deterioration in response times? Could it be increased demand? Our analysis shows overall number incidents attended hasn’t changed much though the proportion of most serious calls has increased slightly. 4/13
And spending is rising because we want to take advantage of advances in medical science and tech which give us the opportunity to live longer healthier lives. Like in other areas of our lives, when there are innovations which can improve our lives we spend money on them.
Excellent analysis that busts the commonly held belief that social care workers leave for better paid jobs in retail. Only 3% of leave for retail. Far more - 28% - leave for nursing related jobs.
This chart shows occupations social care staff worked in before and after leaving the sector. Sales & retail assistants (inc supermarket workers) are biggest single source but only account for 10% of all entrants. V few (up to 2021) leave for sales & retail (2/7)
@hscwru