"Thirteen thousand people died needlessly at the 14 worst NHS Trusts".  It must be true the Telegraph headlined it on Sunday.

It's wrong and based on a what I can only assume is a deliberate misunderstanding of HSMRs. 

Hospital Standard Mortality Ratio calculations give us questions to ask.  HSMRs are indicators of healthcare quality that measures whether the death rate at a hospital is higher or lower than expected. 

The maths are simple; take into account a patient's age, the severity of their illness and other factors, such as whether they live in a more or less deprived area, this forms the basis of working out how many patients we might expect to die at each hospital.  This is then compared with the number who actually die. If the two numbers are the same, the hospital gets a score of 100. If the number of dead is ten per cent less than expected they get a score of 90. If it is ten per cent higher than expected, they score 110.  They are published in quarterly data sets.

As this BMJ article points out; 'The HSMR is complex but cheap and relatively easy to calculate from national or other benchmark data that allow calculation of patients' predicted risks of death. However, there are a number of methodological challenges in their construction'.  Not least the accuracy of coding.  The arguments ebb and flow.  The man who 'invented' HSMRs is Prof SirBrian Jarman.  He says they are a sign-post, where we should look for poor quality care.  He is usually right. 

So, have 14 'failing Trusts' killed 13,000 people since 2005?  No.  All the HSMRs tell us is 13,000 more people died than would have been predicted.  But, 13,000 deaths in 14 Trust, over 8 years means a figure of about two a week.  Don't get me wrong, I am not saying 2 is OK.  Zero is the goal.  What I am saying is this; how do we know if those two deaths were the result of poor medicine or poor care?  Is the Telegraph headline misleading? 

To find out we would have to commission a Clinical Record Review for every person who died in those 14 hospitals since 2005.  A single CRR can take months and Bruce Keogh's merry band has not had the time or resource to do 13,000.

The Telegraph report continues with a litany of Trust failings and foul-ups that I certainly don't intend to defend. 

The CQC is once again mired in questions about why and how they let it all happen.  The truthful answer is Inspectorates don't work and we'll all be back again, next year, 'astonished' to hear another Trust has been responsible for death and mayhem.  The CQC should be honest enough to say they can't do what they've been asked to do, they  can only count the dead.  To pretend otherwise is a fraud on the public.

Bruce Keogh's report is due out later in the week.  I predict he will point out failings.  Hospitals will say; 'That was then, this is now, it's all changed'.  LaLite will slag-off the 'failing NHS', someone will leave (and be rehired) and there will be suggestions of mergers and downgrading that the press, public and MPs will resist.

Why Keogh has allowed his report to 'leak' in this fashion beggars belief.  What has he gained?  Badly written headlines and misleading information for the public.  Who benefits? 

It's obvious; LaLite has manipulated Keogh and grabbed the weekend press.  Select Committee Tory member, David Morris hopped on the bandwaggon and said; "Andy Burnham and his predecessors missed far too many warnings about high hospital death rates. He should take a long hard look at his record and ask himself whether he is really fit for the role of shadow health secretary."

Morris appears to be unaware of the Burnham enquiries into 5 of the hospitals included in the Keogh report which the Coalition did nothing about.

The upshot?  Patients caught in the cross fire of a political battle.  Their worries become the collateral damage.  This morning people will be getting ready to go to these 14 hospitals for treatment.  Most of them will have no choice, their Trust the only show in town.  Are they anxious, reluctant, concerned?  Who cares... headlines and politics are much more important.