If a national newspaper published that 'slapdash treatment and blunders in the NHS was costing personal injury and lives to the tune of �2.5bn a year' we would expect it to be backed by facts and 100% true. Well, they have and it might be but we really don't know.

The Telegraph, the Indy and the Guardian have all fallen hook, line and sinker for LaLite's story that he does know because he has a new report that tells him it is true. It doesn't and it is not.   

Does any one read these reports?  Am I the only one?   Because a report is written by a posh consultancy group doesn't make it gospel.  I say; get yer common-sense shovel and dig into it.   

The report is from an outfit I have never come across, headed up by former Cabinet Secretary Gus O'Donnell,   boss of Frontier Economics; they 'do' electricity, shops and gas.

The report is a disappointing collection of everyone else's stuff, rewritten and rebadged. I think, with a free afternoon and Google you could have produced the same 'Dodgy Dossier'.  If the NHS is a Rolls Royce this report is O'Donnell trying to flog it a set of retreads. 

O'Donnell goes back to 2000 to the CMO's report, 'An organisation with a Memory'.  O'Donnell's people, to their credit, say; "The current evidence base has limitations and this analysis has therefore focused primarily on setting out a range of plausible estimates..." In other words, they've guessed. They have extrapolated, aggregated and taken a stab at it.  This is the point where O'Donnell should have said there is not enough data and we can't add to what you already know.  I suspect filthy lucre blurred his vision. 

O'Donnell accepts case-study reviews are the gold-standard but confesses; "... the labour-intensive nature of this sort of research means that these studies are typically quite narrowly focused in terms of the locations they cover, and any generalisations from such studies can be methodologically challenging."  

With little else to turn to they major on the work of Charles Vincent done in 2001 based on records from two hospitals of which Vincent himself says; ".... (we) do not claim the results can be generalised more widely across the NHS."

O'Donnell ignores this, gets out his calculator and multiplies Vincent's results by last year's finished consultant episodes and concludes (Page 14) 755,000 might be the number of preventable adverse events; 5%.

O'Donnell gets to the cash by multiplying 755k by (in the words of the report) "The average cost of an inpatient stay (which) was about �3,366".... and concludes �2.5bn.

Elsewhere in the report (Page 14) O'Donnell says; "...the level of preventable adverse events has decreased since the Vincent study". Er, well... where does that leave the �2.5bn? It's not a guess any more it's a punt.  Frankly it is rubbish.

How much do you think the DH paid for this flim-flam?

On page 15 there is a table of scenarios. Get that; 'scenarios'. Not results, evidence, facts or proof. Scenarios; stories, guesses, set-ups. At the bottom of the page it says; "... The lack of systematic evidence about preventable adverse events means that there is uncertainty over such top-down estimates." In other words; we don't really know.

The most honest thing the report says (Page 17) is; "...Taken together, the evidence from the different approaches suggests that preventable adverse events cost the NHS a significant amount of money." Trallah! Thank you. We know.

There are two pages of stuff cloned from Monitor and a number of examples of international good practice that could have been taken from best practice in England.

Honestly? My opinion? It is drivel. The sort of essay a student might write, better.

Is it strong enough for a Cabinet Minister to pray in evidence? Absolutely not. Should the newspapers have presented it in the way they did? Absolutely not. Does it tell us anything we didn't know? Absolutely not. Should the taxpayer stump up for it? Absolutely not.

Is there unsafe care in hospitals? Absolutely yes, globally it is a problem. Should we bust a gut to end it? Absolutely, yes. How to do it? Only one way; the gold standard:

• Case-study reviews of all deaths. Collect real data in real time.  Invest in the technology to do it.
• Medical and clinical teams encouraged to be proud of each other and review their work together: like RAF fighter pilots and support teams do after each mission; no rank, no secrets, no holds barred, no stone unturned.
• Accept mistakes and errors will happen treat them like diamonds under the soles of your feet, dig them up, treasure them, they are priceless; look at every facet and ask; 'what can we learn'.
• Protect the front-line, fund it properly, make it fun to work there and most error-type problems will disappear.

...and Gus, give us our money back.

Have a good weekend.