Death rates at two Norfolk hospitals higher than expected

Mortality data released across the country was condemned as 'very troubling' by campaigners.

But locally both the James Paget University Hospital (JPUH) in Gorleston and the Norfolk and Norwich University Hospital (NNUH) had more deaths than predicted between April 2016 and March 2017.

At JPUH in 2016/17, it was expected 1,394 people would die - but the actual figure was 1,649. At NNUH there were 3,546 deaths against 3,297 expected.

JPUH was also named by the agency that collects data on the health service, NHS Digital, as one of the ten hospitals they highlighted as having a higher than expected number of deaths.

Peter Chapman, medical director at Norfolk and Norwich University Hospitals NHS Foundation Trust. Photo: NNUH (Image: Archant)

Both trusts exceeded the foreseen number of deaths for the second year running, while the Queen Elizabeth Hospital (QEH) in King's Lynn had fewer deaths than expected - 1,676 when there were 1,707 predicted.

In 2015/16, the measure recorded 1,422 deaths at JPUH (against 1,317 expected), 3,415 at NNUH (3,235 expected) and 1,574 at QEH (1,626 expected).

The data also shows the most common diagnoses patients had before they died - at all three hospitals most deaths were from pneumonia.

JPUH medical director Nick Oligbo said the trust took the results 'extremely seriously' and a full investigation was under way as a priority.

Nick Oligbo, JPUH medical director. Photo: Archant library (Image: Archant)

He said: 'Our patients always come first and therefore it is important to identify any issues. Previous detailed case note reviews by clinical teams have shown no clinical care concerns. We also track our crude mortality, the actual number of deaths compared to the number of people that come to our hospital, which has not changed.'

He said the probe would involved reviewing all documented cases over the last year. But he added: 'This is a really complex issue, with potential differences in how trusts record information. For example septicaemia may be recorded by a clinician as pneumonia and vice-versa. Our population is more elderly with a high incidence of co-morbidities - one or more additional diseases or disorders such as asthma or high blood pressure, mixed with smoking - which can make treatment more complicated in ensuring that each condition is appropriately managed.'

NNUH medical director Peter Chapman pointed instead to standardised hospital mortality data (HSMR) - the other measure of hospital mortality rates.

He said: 'In the last 12 months our HSMR has fallen from 110 to below 100, with fewer observed deaths than expected, which is excellent news for patients. We have done this by really concentrating on patient pathways which mean that patients consistently receive a good standard of care. We are confident that this improvement will continue as we understand that for the latest three months of data available to us our HSMR is below 90.'

'To achieve this improvement, we have made a significant increase in our palliative care provision and improved our recording and coding of patient data. This now gives us the best HMSR of the hospitals in Norfolk.'

QEH medical director Dr Nick Lyons added: 'SMHI data is a useful tool in our continued work to ensure continued patient safety. As always we will continue to use it as part of our quality monitoring process.'

How are mortality rates measured?

The data comes from the Summary Hospital-level Mortality Indicator (SHMI) scheme, which was set up in response to the Mid Staffordshire scandal, where up to 1,200 more patients died over three years than would have been expected.

It predicts how many patients would be expected to die at any particular hospital. And then compares the actual number of deaths at hospital or within 30 days of discharge. Some deaths are excluded, including those in A&E, day cases, stillbirths, or if the patient was subsequently admitted to another hospital within 30 days.

While SMHI compares actual and expected deaths at a particular hospital, a different measure - the Hospital Standardised Mortality Ratio (HSMR) - looks at the number of deaths likely to occur at an average NHS hospital. The SHMI is considered more precise and the government is planning to adopt it as the sole mortality indicator for NHS hospitals.