Interesting NHS data no-one is looking at
Last week, John Appleby, chief economist at the King's Fund, wrote a cry of help.
One of the most interesting and important set of statistics in the NHS is being neglected. And he's right to be annoyed.
The statistics - known as proms - are respected, cheap to collect and provide useful information.
But, since their initial proof of concept, he says, their use has progressed "barely at all".
I, for one, think this is a terrible shame because they - uniquely in the health service - relate to real care quality.
Proms stands for "patient reported outcome measures".
Before some of the more popular forms of elective surgery, patients fill in a form explaining how they feel.
Then, after the operation, they fill them in again. Using that, we can say who is good at getting patients to feel better.
Why is this so important?
Well, most statistics on this kind of care are about mortality. Quite often, we struggle to say more than "well, they didn't die".
But using proms, we can very easily see whether care quality is rising. And we can also do fancier stuff.
For example, we can see whether GPs - newly empowered by the Andrew Lansley reforms - now commission more operations from very good providers and fewer from the very bad.
If his market-based reforms are working, they should be.
So far, that is not very apparent.
The share of hip replacement operations commissioned from good providers rose marginally from 10.2% to 11.5% in 2013-14 - the first year of the reforms.
But it did not fall at the worst providers, sticking at about 10%.
That is actually a very tiny amount better than before the reforms - but we'll have to wait to see if we get any significant movement.
We can also see evidence of excess treatment within proms.
Overtreatment means unnecessary surgery and drugs. It's both expensive and genuinely terrible for patients.
For example, take hernias - that's when muscles near your groin tear, allowing a portion of your intestines to poke through.
It is a condition that affects men, in particular. There were 60,000 groin hernia repair operations in England in 2013-14.
It has historically been treated with surgery - usually by shoving a small piece of plastic mesh between the muscles and intestine.
The principal idea is to prevent dangerous complications known as "strangulation" or "obstruction".
But that is now reckoned rare for men with few symptoms.
In 2006, the Journal of the American Medical Association advised that "watchful waiting" was advised for men "with asymptomatic or minimally symptomatic" hernias.
Yet according to proms, about a fifth of hernia repairs are for men who report no or minimal discomfort on the proms test.
Furthermore, 85% of them report feeling worse after the treatment.
That's about £50m of possibly unneeded treatment right there, hidden in a survey.
Tim Hughes, a doctor at Kirkbymoorside in Yorkshire, has recently been using proms to encourage hernia patients to consider whether they really need their treatment - and most decide not to have it.
We should be watching proms to see how that progresses in the coming years.
We can also use proms to see whether private providers operating inside the NHS are "cherry picking" - that is to say, getting easier patients.
Because the NHS has an inflexible pricing structure, they often get as much money for treating easy cases as slightly harder ones.
You can use proms to do more sophisticated analysis, but - for a start - the average groin hernia repair patient, at the average NHS provider, has a health score of 77 out of 100.
At private providers, it's 81.
That's not much, but enough to say they do tend to take slightly easier cases.
But the private providers also do slightly better by them, since the improvement in scores tends to be a bit higher at the private providers.
They get an average improvement of 8.8 points. For state providers, it's 7.8.
Proms is perhaps the most interesting data on the NHS that no-one is looking at.
Given how sterile much of our discussion of the health service is, Prof Appleby is right: that's a real shame.