Hunt's plans for local avoidable death estimates ‘meaningless’

Nick Black, professor of health services research at the London School of Hygiene and Tropical Medicine, said he did not think the health secretary’s proposal to produce a tailored percentage of avoidable deaths for each hospital in England based on a national estimate could work.

Mr Hunt on Sunday announced an annual review of roughly 2,000 patient case notes to establish “a national rate of avoidable deaths every year”.

The review expands upon research by Professor Black and colleagues, published in 2012, which estimated from a study of 1,000 case notes that there are almost 12,000 preventable hospital deaths in England every year.

However, the government plans to go beyond a national estimate of avoidable deaths by developing a “projected share” of deaths for each hospital.

The Department of Health said this would be done by adjusting the estimated national rate of avoidable deaths “to take account of each hospital’s known safety metrics”.

Hospitals would then be banded “according to the number of deaths estimated locally”, with trust chairs required to write to the health secretary annually on receipt of their figures to update them on progress reducing avoidable deaths.

While Professor Black welcomed the national review and was quoted by the DH in its press release, he told HSJ he questioned the feasibility of the plans for producing estimates for each hospital.

He said: “I don’t understand what the secretary of state is suggesting in terms of in some way modifying that estimated [national] number according to some unspecified safety metric.

“We’re sort of into a chicken and egg [situation] then; if we had an accurate indicator of the safety of a hospital we wouldn’t need to do all this, so I’m intrigued by how that would work and to be honest, I cannot see that it would.”

• ‘Hard evidence’ special measures worked at Keogh trusts, say researchers
• More news and resources on mortality rates

He added that an estimated number of avoidable deaths generated from such a process “wouldn’t have any meaning”.

While there is a wealth of indicators used to measure safety, including patient incident reports, falls and MRSA infections, Professor Black said a different estimate of avoidable deaths would be produced depending on which one was used to adjust the national rate.

“My guess is that whichever one of those you took, you’d come up with a different answer because we know that they don’t all line up together and all point in the same direction,” he said.

Professor Black said the DH’s proposal reflected a “lack of understanding of what data’s available”.

While he did not believe the national review could be used to rank hospitals, Professor Black said it could still give trusts a useful indication of how many avoidable deaths were occurring in their hospitals.

“I would welcome each and every trust board being told, ‘if you were a typical hospital we would estimate that there are, let’s say, 60 avoidable deaths in your trust each year’.

“It may be that just having that exact figure would make you sit up and say, ‘oh goodness, this isn’t just some abstract notion out there for the NHS as a whole’.

“It might help stimulate quality improvement in the trust, and what’s not to like about that?”

The DH said it would publish a “best estimate” of the number of avoidable deaths for each trust by the end of March next year, but did not outline which methodology would be used for calculating it when asked by HSJ.

A spokesman said: “We will continue to work with frontline clinicians, national organisations and academics to support trusts to better understand their levels of avoidable mortality, and, crucially, to take effective action to reduce those levels.”