Mid Staffs report demands stats overhaul

Tags: A Audit coding Dr Foster Engagement Information Centre Inquiry Mid Staffs Monitor statistics

24 Feb 2010

An independent report into the scandal at Mid Staffordshire NHS Foundation Trust has called for an “impeccably independent and transparent source” to be set up to produce comparative statistics for the NHS.

The exhaustive report by Robert Francis QC, a medical law expert, devotes a chapter to coding and mortality statistics. It expresses concern that trusts may be coding in different ways and that different bodies place different interpretations on the resulting data.

It says “there is surely a need to ensure that there is uniformity of [coding] practice across the country, so that the public can be assured that like is being compared with like” and that “there may be a case for considering whether a public service should be tasked with producing [comparative] statistics.”

Coding and statistics played a key role at Mid Staffordshire, where the board initially blamed coding errors for apparently high death rates revealed by Dr Foster’s monitoring unit in summer 2007.

The unit alerted the Healthcare Commission to its findings, and the Commission launched an investigation in summer 2008.

The Commission concluded that more patients died than would have been expected over a three year period, primarily because of “shocking” care for those admitted through A&E.

But it also uncovered poor standards on a number of wards, with patients left unwashed, unfed and without access to medicines.

The government subsequently set up the Francis inquiry, which today published a report backing the Commission’s findings.

“What this investigation has uncovered cannot be expressed in statistics – indeed, an over-reliance on figures has been one of the reasons for the trust’s failings,” it concludes.

“If anything, the scale of the deficiencies uncovered is greater than that of the Healthcare Commission’s report. Shock is the appropriate reaction to many of the experiences the inquiry has been told about.”

The 450 page report includes oral evidence and letters from the relatives of patients, one of whom described how her husband was repeatedly left unwashed and unfed and with his medication out of reach.

“On numerous occasions when I arrived on the ward, he was lying in faeces,” she told the inquiry. “His wounds smelled dreadfully and needed to be cared for properly.

“I had to keep asking for his [pain relief] syringe drivers to be refilled when they emptied. He did not deserve to end his life in such an undignified manner and in such distressing circumstances.”

However, one of the key findings of the report is that the trust failed to act effectively on complaints from patients and relatives and on staff concerns. Instead, it said a common response was to “refer to data, often of a very generic type, such as star ratings, CNST levels and so on.”

The report is clear that it does not want to downplay the importance of data, and indeed calls for greater staff engagement with audit, mortality and morbidity reviews, most of which were absent at Mid Staffordshire.

However, it says: “benchmarks, comparative trust ratings and foundation status do not in themselves bring to light serious and systemic failings.” Other criticisms of the trust include weak management and poor professional standards.

The report presents the pros and cons of stripping the trust of foundation status, saying the health secretary must decide whether to ask Monitor to look at doing this.

Other recommendations include improving training and audit at the trust, improving its complaint and incident handling, reviewing the roles of commissioning, supervisory and regulatory bodies in picking up problems, and promoting a culture of openness at all trusts, not least by ensuring their boards meet in public.

On coding and the production of statistics it says: “In view of the uncertainties surrounding the use of comparative hospital mortality statistics in assessing hospital performance and the understanding of the term ‘excess’ deaths, an independent working group should be set up the Department of Health to examine and report on the methodologies in use.

“It should make recommendations as to how such mortality statistics should be collected, analysed and published, both to promote public confidence and understanding of the process and to assist hospitals in using such statistics as a prompt to examine particular areas of patient care.”

Link: Independent inquiry into care provided by Mid Staffordshire NHS Foundation Trust January 2005 - March 2009 Volume One.

Opinion and analysis: More on the Healthcare Commission's findings.

Lyn Whitfield

Mid Staffs death rates - poor care not poor data (17 Mar 2009) (17)
Data recording may explain trust’s high death rate (18 Mar 2008)

Readers Comments

Data Validation

gerard@careprovider.com

25 Feb 10 23:54

Independent data checking was called GP fundholding.

Trusts hated it at the time because it forced them to look at their data for small proportion of their activity. GPs knew what patients had had done to them as they saw the patients day to day, and could challenge hospital invoices. Successful lobbying in 1996/7 brought that wonderful data checking system to an end when fundholding was abolished in 1999. Its time has come again.

Clinical Coders do not kill patients

26 Feb 10 07:33

Clinical Coders cannot possibly be to blame for the deaths at Mid Staffordshire NHS Trust, they never see a patient !!!

Their role is to reflect the episode of care as written by the Clinician in the case notes and to put this into a coded format.

All Clinical Coders code for Morbidity analysis not Mortality!

The Clinical Coders at this Trust have a good standard of coding and recieve excellent training to ensure they work to the national standards as set by NHS Classifications Service and are audited at least twice a year.

Their main problem is the inconsitency within the case notes, the illegible writing and the lack of a good summariastion of the episode of care from the clinicians.

Stop blaming the Coders!

How about political policy validation?

26 Feb 10 11:35

While we're at it let's see the evidence that achieving Foundation Trust status has positive effects on services for patients and their outcomes.

Stats are too late to save patients' lives

rsarson@blueyonder.co.uk

28 Feb 10 12:10

Forget whether the coding is right or wrong, it amazes me as a patient that stats are the only way that a hospital is judged. The stats come too late. Look at Staffs; Dr Foster high-lighted high mortality in November 2007, based on data collected for the previous year. It took the NHS until March 2009 to admit there was something wrong, and another year to produce the Francis report. How many people died unnecessarily in those three and a half years?

Stats just take too long, and NHS administrators try to rubbish them if they are unfavourable. Some other way is needed to judge failing hospitals. Like using your eyes and nose, and seeing how staff react to probing questions. Easy, really. Frank Dobson used to say that he could judge a failing hospital in five minutes flat.