Who is, and is not, invited to Cameron's emergency NHSbill summit? A data visualisation.

- Edit
- Delete
- Tags
- Autopost
Who is, and is not, invited to Cameron's emergency NHSbill summit? A data visualisation.

David Cameron is holding an emergency summit about the troubled NHS bill on Monday. There has been a lot of chatter on twitter about who is, and is not, invited. It's notable, for example, that the elected head of the Royal College of GPs has not been invited: the NHS bill is all about putting power in the hands of GPs, but they have such serious concerns about the bill that they've called, with regret, for it to be dismissed.
I've put the main players into a table, to see if the invitation list it tells a story. Let me know if you think it does: I'm still not sure what statistical methods would be best to analyse this data, and extract the signal from the noise?

(Apologies if I've left organisations out. Also. I'm sure this invite list will change slightly over the coming weekend, if only enough to avoid the lolz on Monday.)
- 17 February 2012
- Views
- Permalink
- Tweet
3 months ago killhopelaw (Twitter) responded:

No surprises there

3 months ago killhopelaw (Twitter) responded:

Ben I've heard about your book it's on my must read list sorry didn't realise

3 months ago SceptiGuy (Twitter) responded:

I think the RCN, RCGP and BMA should hold their own summit and not invite Lansley.

3 months ago AndyM54 (Twitter) responded:

That is disgusting, just by disagreeing to something disagreeable with you miss out on a meeting with Cameron.

3 months ago Guy Hornsby responded:

Quelle surprise. Odds on them doing a poll and reporting that everyone is bang up for it.

3 months ago paulstpancras (Twitter) responded:

Cameron, Lansley, their Management Consultants sorry CCGs and Focus Groups look likely to be present.

3 months ago onegpprotest (Twitter) responded:

The Royal College of Paediatrics may be heading in direction of opposing the bill. A number of eminent paediatricians signed the letter to the Lancet http://bit.ly/xsxQL3 and they have called an EGM to be held shortly.

3 months ago nottRobin (Twitter) responded:

Yes it's not like the organisations that have been invited are 100% for the bill. There should hopefully still be some good points raised at the summit.

3 months ago InterUncut (Twitter) responded:

Hardly surprising of such a despicable lying Tory Government. Don't they realise we can see right through their charade?

3 months ago MikeTonge (Twitter) responded:

I think I may have spotted a pattern there, but I'm not sure.

3 months ago Ben Goldacre responded:

yeah me too. chi squared test? but we don't have more than 5 expected in each cell yet.

3 months ago djb_ptigga (Twitter) responded:

Can we see that as a Venn diagram?

3 months ago Ben Goldacre responded:

i'm keen that others should use this raw data to create their own visualisations, but it's going to be a *very* complex venn diagram. do you have some specialist software? or maybe a biro and an envelope?

3 months ago Guy Hornsby responded:

The maths is beyond me i'm afraid, and the diagrams would be too complicated.

3 months ago RealRiffler (Twitter) responded:

You might want to change "Tolerate the NHS bill" to "Haven't called for the bill to be dropped... yet" as that seems to be the way things are going.
And the emergency meeting that is going to lead to the bill being dropped is the one that doesn't include an invite for Lansley.

3 months ago mark_mcguire (Twitter) responded:

Hi Ben
A little bird drew me here (I caught your re-tweeted tweet).
This table really helps to make the situation clear. It would be even better, perhaps, if you also made use of an appropriate visual metaphor. Maybe interest groups represented as individuals admitted through a door (or turned away). Visuals can help to support stories to increase the "get it" factor. I don't know the background story behind this issue, so I'm not sure what scene might help to explain what you are revealing. Your table is certainly a lot more effective than a paragraph reporting on who is attending. Listing who is NOT welcome as well as who IS welcome already pulls aside the curtain. Now, what scene could be staged that would set up an equivalent or resonant narrative? I guess I'm thinking of a cross between a political cartoon and a graphical display. In any case, I applaud your efforts and I'll investigate your other work.
Thanks.
Mark

3 months ago DinahMay (Twitter) responded:

Cameron et al - we despair at their blatant dismissal of the e-petition and now they put up this farce. Why aren't the benches opposite shrieking and raging? I've done my diagram and funnily enough it resembles a big foot (I think it's a cameron foot) crushing 3 letters - nhs

3 months ago iftmaster responded:

'Tolerate...'? 'Aren't calling for...' would be better and wouldn't dilute the point.

3 months ago zut responded:

Ben,
The Royal college of Radiologists had an EGM yesterday. Motions were passed against the bill. They don't seem to be invited either. What a surprise!
http://www.rcr.ac.uk/docs/about/pdf/Post_EGM_16212.pdf

3 months ago john_jameson (Twitter) responded:

Ben, you've missed McKinsey. Our government seems to think that they know more about how the NHS should be run than any of these professional organisations.

3 months ago Farctum (Twitter) responded:

And that's without the professional groups whose patients are most likely to be directly impacted by yet more unnecessary organisational change eg Podiatrists, Physiotherapists, SALT, OT, children's services etc etc

3 months ago ChangeTS (Twitter) responded:

Looks like we're seeing (yet again) Group-think in action. Whether consciously or unconsciously - gathering a group of people who will not offer you the rigorous challenge necessary when making important decisions.

3 months ago Antinomodes (Twitter) responded:

Ok - as Ben Goldacre says, Chi-squared - but for 'homogeneity' or 'goodness of fit'? They seem to be a very heterogeneous bunch, with not much goodness to the fore, and as for fitness, who can say...?

3 months ago PaulG22 (Twitter) responded:

Maybe if you presented the data a bit more visually it would be easier for us to interpret? Perhaps one of the 3D bar charts in Excel or even a scatter graph?

3 months ago midalgleish (Twitter) responded:

This is disgusting. Another example of David Cameron and his evil self serving manipulative party doing what they want and ignoring the people.

3 months ago Paul Sampson responded:

Invited to neither, so I guess it balances out, are the user group, the Royal College of Patients and the patron, the Royal College of Taxpayers.

3 months ago Jessica Classom liked this post.

3 months ago BarryTCat (Twitter) responded:

Hey Ben Goldacre and fans,
a summit for the uninvited has been called for 12-4pm, Monday, outside 10 Downing Street. A petition- similar to the ad hoc letter to the Telegraph from disgruntled Doctors -is being organised alongside this.
Please come, http://www.facebook.com/events/191859057582900/ and help spread the word.
Not many people follow me on twitter (and with good reason), so I would appreciate the word being spread. No need to re-tweet or namecheck me or any of that jazz. Just help attendance please.
Thanks

3 months ago Dirk vom Lehn liked this post.

3 months ago Mike liked this post.

3 months ago ian_hiles (Twitter) responded:

No mention of the public being invited. Surely the health of the public being such a driving force of the bill, we would be top of Cameron's list? A casual viewer might think public opinion wasn't welcome in Number 10.

3 months ago John Keogh liked this post.

3 months ago CyborgDave (Twitter) responded:

I *think* I've been able to work through the complex mathematics involved and come up with the following Venn diagram, which may help further visualise the dynamics of the situation : http://www.twitpic.com/8llmo7

3 months ago storagechat liked this post.

3 months ago andrewlavender responded:

Gosh, Ben. It's astonishing how many eminent groups have decided not to support a bill simply because they're not invited to a summit.
(Or were you trying to derive causality from correlation, one of your favourite no-nos i thought? :-) )

3 months ago R0b1et (Twitter) responded:

I'd suggest a Heidke’s skill score. Quite a good case really... if we consider DC's trying to predict a supporter/rejector and compare with observations, we find a=4, b=c=0 and d=5. So the Heidke’s skill score =1 (or 100%). So Cameron was at least skilled, even if biased.

3 months ago R0b1et (Twitter) responded:

I'd suggest a Heidke’s skill score. Quite a good case really... if we consider DC's trying to predict a supporter/rejector and compare with observations, we find a=4, b=c=0 and d=5. So the Heidke’s skill score =1 (or 100%). So Cameron was at least skilled, even if biased.

3 months ago iainstrachan responded:

Well, I'm jealous. As a confusion matrix this tells me that the sensitivity is 100%, the specificity is 100%, the Positive Predictive Power is 100% and the Negative Predictive Power is 100%.
I wish my datasets were as clean as this! Or perhaps it shows that the hypocrite that governs the country is a highly skilled cherry-picker!

3 months ago supremelydisappointing liked this post.

3 months ago lordbile responded:

class

3 months ago Reg Edward responded:

Anaesthetists, the biggest department in every hospital , where are they?

3 months ago rfielding responded:

Fisher's Exact test and I'll tell you now, it's p is a lot smaller than 0.0001

3 months ago Unionbuster (Twitter) responded:

Where is the Patients voice?
Sod the unions & employees they always want the status quo . .

3 months ago neuropsych responded:

Yes it's Fisher's Exact Test, which gives a precise answer when the numbers in the cells of the contingency table are too small for using the Chi squared statistic.
The (two-tailed) figure for p is 0.0079 (statistically significant at the 0.01 level).
Which means that the probability is 0.0079 (1 in 127) that this selection of inclusions and exclusions happened by chance.
Of course the number of exclusions is not limited to 5, but any higher number in that cell just decreases the probabality that this was chance. Likewise any higher number in the top left cell does the same. Only some numbers in the bottom left or top right cell (ie health professional groups who tolerate the bill and are not invited, or those who want it dropped but are invited) would change the probablility upwards and make this look less like a clear-cut discrimination on grounds of attitude to the bill.

3 months ago neuropsych responded:

correction: increases the probablity that this was not chance. doh!

3 months ago neuropsych responded:

Sorry, ignore that correction, I'm just confusing myself - you get the idea

3 months ago mc_hankins (Twitter) responded:

Clearly a 'goodness of fit-up' test is required.

3 months ago colinwforster (Twitter) responded:

Also not invited are all those private corporations who will benefit from the Bill and the management consultancies who wrote it.

3 months ago shenting responded:

Looks like we have a really excellent screening test here!

3 months ago jvfreeman responded:

Really liked this table. The main modification I would suggest would be to scale the text for each organisation by the size of the membership, that way you will be able to see a little more clearly the scale of support/opposition to the bill. Obviously, you will need to be creative about the BMA given that it is the main trade body representing all doctors.
ps. Technically speaking the p-value shows, given the data you have, how likely is it that these data have arisen by chance alone i.e. how likely is it that the pattern we see in the table has happened by chance alone - you need that caveat about 'given the data we observe' otherwise it doesn't make sense. Thus a small p-value indicates that this is pretty unlikely!

3 months ago jvfreeman responded:

Though not to be too much of a pedant, in the case of Fisher's exact test it is the probability of observing a table at least as extreme as the one you have observed

3 months ago Call_111 (Twitter) responded:

Nice to see the list. I think it important to keep away people with vested interest and make sure he has enough information to make GPs work efficiently. BMA is useless any way and is not directly involved in offering patient care but a club.

3 months ago studentpara responded:

College of Paramedics?
Ben Goldacre's Space

I like stats. I'm a doctor. I write about dodgy scientific claims in my spare time.
This isn't my main blog. Find me here:
Blog:
www.badscience.net/
Book:
http://www.amazon.co.uk/Bad-Science-Ben-Goldacre/dp/000728487X/?tag=bs0b-21
Tweet:
www.twitter.com/bengoldacre
TED talk:
http://www.ted.com/talks/ben_goldacre_battling_bad_science.html

bengoldacre - secondary blog

Who is, and is not, invited to Cameron's emergency NHSbill summit? A data visualisation.

Ben Goldacre's Space

About Ben Goldacre

Archive

Follow Me