The Consultation Document – to be scrutinised by Councillors at a Joint Health Scrutiny Committee meeting on 22nd Feb, 3.45pm at Halifax Town Hall – claims on page 7 that the hospital cuts and changes scheme is needed because the number of patients dying in our hospitals is higher than average.
This is a MYTH
FACT: For the year July 2014 – June 2015, Calderdale & Huddersfield NHS Foundation Trust hospitals had a higher than average figure for the Summary Hospital-level Mortality Indicator (SHMI).
FACT: The SHMI interpretation guidance from the Health and Social Care Information Centre says that the SHMI is not a direct measure of quality of care, because without a detailed case note review, it isn’t possible to determine whether or not a death could have been prevented.
So the FACT that our hospitals had a higher than average SHMI (for one year) can’t be taken to represent the number of avoidable deaths – and NOWHERE does the Health and Social Care Information Centre SHMI interpretation guidance say that a higher than expected SHMI figure is justification for a major shake up of a hospital’s clinical model.
In fact it urges caution and further investigation, starting with the most likely explanations – which it says lie with data errors and patient case mix.
FACT: The Health and Social Care Information Centre Summary Hospital-level Mortality Indicator guidance says that a ‘higher than expected’ SHMI should not immediately be interpreted as indicating bad performance – nor should a ‘lower than expected’ SHMI immediately be interpreted as indicating good performance.
FACT: The Health and Social Care Information Centre Summary Hospital-level Mortality Indicator guidance for hospitals with a higher than expected SHMI is that they should view it as a ‘smoke alarm’ which requires further investigation by the trust, through careful interpretation in conjunction with other indicators and information from other sources (e.g. patient feedback, staff surveys and other similar material).
FACT: In carrying out such further investigation, the Health and Social Care Information Centre Summary Hospital-level Mortality Indicator FAQs recommend that hospitals use a structure such as the pyramid of investigation for special cause variation.
FACT: This pyramid identifies the most likely explanations for a higher than expected SHMI as being “Data” and “Patient case mix” and says that these factors should be investigated first.
The “Data” is likely to be about coding issues and the patient case mix is about identifying whether the hospital is admitting patients who are more ill than a normal patient case mix and so more likely to kick the bucket.
FACT: The Summary Hospital-level Mortality Indicator (SHMI) quarterly report for the period July 2014-June 2015, which shows that Calderdale and Huddersfield NHS Foundations Trusts ’s SHMI is a higher-than- expected “outlier”, defines case-mix as referring to the characteristics of the patients treated by a particular trust:
“In the calculation of the SHMI, the characteristics included are the condition the patient is in hospital for, other underlying conditions the patient suffers from, age, gender and method of admission to hospital, because these impact on the risk of mortality and are outside of the control of the trust.
The severity of the condition is not included, as this information is not captured in the Hospital Episode Statistics (HES) dataset upon which the SHMI is based. This means that if a trust treats a high (or low) proportion of seriously ill patients with a particular condition compared to other trusts, the statistical models used to estimate the expected number of deaths will not take account of this.”
FACT: This same Summary Hospital-level Mortality Indicator (SHMI) quarterly report points out that the SHMI methodology doesn’t make any adjustment for deprivation,although two contextual indicators on deprivation are published alongside the SMHI to support its interpretation. These are:
- Provider spells split by deprivation quintile
- Deaths split by deprivation quintile
FACT: The Summary Hospital-level Mortality Indicator (SHMI) quarterly report shows that our hospitals Trust has a higher than mean average level of deprivation over all finished provider spells AND over all deaths reported in the SHMI.
FACT: The Summary Hospital-level Mortality Indicator quarterly report says that one reason why the SHMI methodology doesn’t make any adjustment for deprivation is:
“because adjusting for deprivation might create the impression that a higher death rate for those who are more deprived is acceptable”.
FACT: The key thing here is that there is a higher death rate for those who are more deprived – and our hospitals have higher than average levels of deprivation both among the patients it admits, and among those who die in the hospital or within 30 days of discharge (which is the deaths that the SHMI measures).
FACT: The other thing that’s notable about the Trust’s most recent SHMI figure is that it’s not a “repeat outlier” with a higher than expected SHMI. The higher than expected figure is just for the year July 2014-June 2015 – the year when the government’s efficiency savings requirement forced the Trust to make cuts that they couldn’t make without damaging patient safety, which is why the Trust ended up in deficit.
RATIONAL QUESTION: Is it possible that the cuts that the Trust did make could have contributed to the higher than expected SHMI?
FACT: At a 29 July 2014 Calderdale Adult Health and Social Care Scrutiny Panel meeting, NHS chiefs made repeated assertions that the cuts were efficiency savings that would not affect the quality or level of service. But at that time, front line staff were saying they couldn’t take any more cuts and preserve patient safety and comfort and their own wellbeing, after services had been pared to the bone following five years of efficiency savings.
FACT: Calderdale Clinical Commissioning Group made assurances throughout the meeting that it had “quality assured” all the cuts and guaranteed that they are all efficiency savings that do not change the amount or quality of the services the hospital provides.
FACT: But this was open to question, because it was unclear which “efficiency savings” they were talking about, since it transpired in the meeting there were two sets of “efficiency savings:
- the ones the Trust thought in April 2014 that it was going make but in July 2014 discovered they couldn’t
- the ones that they were working up at the time of the meeting, that they thought (wrongly, as it turned out) would allow them to cover most of the funding shortfall
FACT: An 18 Feb article in the Nursing Times is headlined:
“Care deteriorating as projected NHS deficit reaches £2.3bn”.
FACT: This reports that a regular survey, carried out by the King’s Fund think-tank, showed that for the first time since it began in 2011, more than half (53%) of finance directors in England said the quality of care in their local area had deteriorated in the past 12 months.
FACT: Royal College of Midwives chief executive Cathy Warwick said:
“Senior midwives are telling us that they have had to cut services due to budget restrictions and cutting staff or holding posts vacant ensures poorer outcomes for women and their babies…the government’s pursuit of savings over safe staffing…seems, at the expense of patient care.”
FACT: One final point to note about the SHMI figures is that even one of the experts on the 2011 panel than came up with this way of measuring deaths in hospital and within 30 days of discharge says that:
“No one, not even its ardent supporters, would maintain that this indicator is an unequivocal marker of hospital quality… there is a risk that the data may be used inappropriately, without regard to the accompanying health warnings, and that some organisations may incorrectly be categorised as providing poor-quality care. This can mislead patients and the public, and erode confidence in NHS staff and organisations.”
RATIONAL QUESTION: Is the Clinical Commissioning Groups’ spin and misuse of statistics on page 7 of the Consultation Document an isolated mistake? Or a more general feature?