Data blindness

Mid Staffordshire NHS Foundation Trust

When the Healthcare Commission issued its report on Mid Staffordshire NHS Foundation Trust this week, one of the first questions that journalists asked was how its high death rates and terrible medical care could have gone unnoticed for so long.

The report says they didn’t. However, the trust had poor information and clinical governance systems, became bogged down in disputes about coding, and almost wilfully ignored what its complaints and patient feedback sites were telling it. The Commission is now calling for better data systems across the NHS that, crucially, capture patient experience. Lyn Whitfield reports.

If the scandal at Mid Staffordshire NHS Foundation Trust is going to make a lasting impact on the NHS, it should make it by changing health service’s attitude to data collection.

This week’s report  by the Healthcare Commission, which catalogues “appalling” failures in the trust’s emergency services alongside terrible stories of poor medical care, opens its national-level recommendations with a call for all trusts to improve coding and the use of statistics in clinical and board decision making.

Crucially, given that concerns about the quality of care delivered at Mid Staffordshire was bogged down for years in debates about coding, it says boards should also be looking for information that “properly captures the experience of patients, so non-executives can scrutinise and challenge the care received by patients.”

Never good

The Healthcare Commission’s report shows that Mid Staffordshire NHS Foundation Trust had been in trouble for some time. As early as 2002, its predecessor, the Commission for Healthcare Improvement, carried out a clinical governance review and identified high levels of emergency admissions, low levels of nursing staff and clinical coding as problems.

Two years later, the Healthcare Commission gave it one star in its first star ratings. In the annual health check that replaced the star ratings, the trust steadily improved its use of resources rating, but remained “fair” for quality of services.

The Commission’s report says the trust was overly focused on stabilising its finances and becoming a foundation trust. “It lost sight of what should have been its main priority, to provide high quality of care to all of its patients.”

A national outlier

In 2007, a few months before the trust achieved foundation status, the Dr Foster Hospital Guide raised concerns about apparently high levels of mortality at Mid Staffordshire. The guide found the trust had a three year Hospital Standardised Mortality Ratio of 125, against the 100 baseline.

From July 2007 to March 2008, the Dr Foster Intelligence Unit triggered four further high mortality outlier alerts, while the Healthcare Commission, which had launched its own programme to identify and follow up apparently high mortality rates in 2007, generated three.

“Each of the alerts was analysed to establish whether the apparently high rates of mortality could be ruled out because of errors in coding or data quality issues, or whether concerns about the quality of care remained,” the report says.

“In a number of alerts, poor coding was apparent. However… we calculated that the chance of this number of alerts, within an individual trust, all being false alarms was extremely low (probability less than 0.001).”

In March 2008, the Commission launched an investigation. Yet the trust continued to insist that it was facing coding problems.Trust chief executive Martin Yeates issued a statement saying this and reassuring local people that “if we thought the trust was unsafe we would have taken action. There is no cause for immediate alarm.”

Poor coding, worse clinical governance

The trust had certainly had problems with coding. A report by CHKS that was commissioned in 2007 found problems in its coding department. The Healthcare Commission’s report says: “The coding manager at the time had been on long-term sick leave and the rest of the team was working part time.”

In addition, clinicians “had little understanding of the need to make notes clear for coders” and the coders were “reluctant to approach them about unclear notes.” The trust finally appointed a new coding manager in July 2007 and matters improved, although a move from manual to electronic data collection in June 2008 suffered “teething problems."

Meanwhile, the trust was poor at using the information it had. Clinical audit was poor and the Commission’s report says clinicians “hardly used” a system introduced in 2007 that would have enabled them to monitor their own performance.

When Dr Foster raised its concerns, the trust first focused on coding issues and then asked clinicians to review the case notes of patients who had died. “The general conclusions… were that the deaths were predictable and no problems with care were identified.”

Complaints and pressure groups

Yet these debates about coding and monitoring were hardly going on in isolation. Non-clinical staff plainly knew something was wrong. In the 2006 staff survey, just 27% said they would be happy to be treated at their own trust, compared with 45% nationally.

And the trust received a large number of complaints from patients – some of whom, indeed went to their local MP to try and get action, formed a pressure group and finally triggered an investigation by the local primary care trust.

The Commission’s report says people complained about “patients not being fed, call bells not being answered, patients being left in soiled bedding, medication not being administered, charts not being completed, and a general disregard for privacy and dignity.”

Yet the board either did not know about them or saw them “grouped into and effectively lost in categories such as ‘communication’ or ‘quality of care’.”

More recently, the trust would also have been able to see comments left by patients on NHS Choices and www.patientopinion.co.uk. Even a quick review of the comments left on NHS Choices shows that people were flagging problems with long waiting times in A&E, uninformed and locum staff, and generally poor conditions.

Last May, one patient wrote: “The wards and corridors need attention to remove the stains of ingrained dirt and body fluids (if that’s what the stains are?!). The nurses appear rushed in doing their work… the clinician never turns up on time… there is little opportunity to ask what I need to.”

Appalling, shocking, failure at every level…

Senior figures in the NHS appear to have been genuinely shocked by what the Healthcare Commission found during its investigation, with health secretary Alan Johnson, NHS chief executive David Nicholson, and NHS medical director Sir Bruce Keogh all issuing apologies.

The Commission concluded that between 400 and 1,200 patients may have died than would have been expected to over a three year period, mostly because of the “shocking” care delivered to people admitted through A&E. But it also found a catalogue of horrors in the theatres and general wards.

The Commission’s report lays out stories of paralysed patients being told to use buzzers they could not reach, patients getting no help to eat or drink, “over half of patients [on one ward] not getting the antibiotics they were prescribed within 24 hours” and others left “screaming in pain.”

Horrifically, an internal audit by the trust last January found that more than half (55%) of patients on one 38 bed ward had pressure sores – yet only four had any pressure damage when they were admitted to hospital.

Further horrors are likely to come to light, since Johnson has ordered two further reviews of the trust. Professor Sir George Alberti, the Department of Health’s national clinical director for urgent and emergency care, will review its emergency work.

And Dr David Colin-Thome, the NHS’s national director for primary care, will investigate standards at the trust from 2002 to 2007 "to determine how the obvious failings were allowed to continue for so long."

Time for action

A national inquiry has also been launched into the effectiveness of the NHS “early warning” systems. This will be conducted by the National Quality Board, a new body due to hold its first meeting on 30 March. Meanwhile, Nicholson has issued a “dear colleague” letter telling trust chief executives and boards to take note of the report and its findings.

These include the recommendation that “trusts [need to be able to] access timely and reliable information on mortality and other outcomes and to conduct objective and robust reviews of mortality rates and individual cases, rather than assuming data errors.”

They also include the recommendation that governance systems “that appear to be persuasive on paper actually work in practice, and that information presented to boards on performance (including complaints and incidents) is not so summarised that it fails to convey the experience of patients or to enable non-executives to challenge on issues relating to patient care.”

And they end by saying that commissioners such as primary care trusts must also “ensure that they have effective mechanisms to find out about the experience of patients and the quality of care in services they commission.”

“We know that the vast majority of patients receive a high standard of care from dedicated and professional staff,” Nicholson’s letter concludes. “However, events like this are a slight on all of us and we must – together – act quickly and decisively in response.”

Links:

The Healthcare Commission’s Investigation into Mid Staffordshire NHS Foundation Trust

The Department of Health’s response