Top doctor baffled as death figures put Northampton General Hospital in the worst grouping

editorial image

A DEATH audit, set to take thousands of hours, is being set up by Northampton General Hospital (NGH) to figure out why its mortality figures are worse than average.

Dr Foster Intelligence’s high-profile Hospital Guide 2011, ordered by the Government, claims to show how many more deaths each hospital has than is expected.

For a number of years, NGH bosses have been among those questioning the number of predicted deaths attributed to their trust statisticians, which then leads to alarming figures of unexpected deaths.

The report published today – which also covers areas such as orthopaedic surgery – puts the hospital in the worst grouping.

After complex sums, the hospital’s number of deaths was given a score of 115 against the national average of 100.

But Dr Sonia Swart, NGH’s medical director, said the hospital was again struggling to understand why so few deaths were expected when its mix of patients appears unremarkable, which then affects how the hospital rates against others in England.

She said: “I’m concerned about it’s because it’s my job to be concerned and I’m very frustrated.

“But I don’t think that, because the way things are calculated puts us worse than the average, it actually makes us unsafe.”

Dr Swart has consistently pointed out the actual numbers of deaths at NGH were among the lowest in the East Midlands. Between 2010 and 2011 there were 1,072 deaths out of 27, 685 patients.

However, Dr Foster believed there should only have been 964 deaths.

The hospital has set up a group of doctors to analyse the paper records of every patient who dies, usually around 1,000 a year, in far more detail.

Doctors admit NGH’s recording of the masses of data that is passed to Dr Foster can be poor or inconsistent and is seeing if the expensive detailed analysis work it has already done to improve key areas – such as aortic aneurysms – can be widened.

It will also seek to find out why the predicted deaths were so low.

Dr Swart said: “The indicators were never designed to compile league tables for patients.

“They were designed to alert clinicians that there may be a problem so they could investigate.”