Combating mortality rates

Northampton General Hospital Trust was one of a number of hospitals that made its way into the 2011 Dr Foster Hospital Guide for having a high mortality rate. Although the trust had been aware of an issue with its hospital standardised mortality ratio (HSMR) for some years and focused on a number of projects and initiatives, none had led to a full understanding of whether there were significant overall deficiencies in clinical care or whether information capture was responsible.

Identifying the problem

In individual areas, detailed case reviews had revealed clinical issues that had been successfully understood and resolved or coding issues that had been corrected. However, for the largest group contributing to the HSMR overall - unselected emergency admissions in older patients - there was a lack of understanding about whether or not there were specific areas that required attention from a clinical care viewpoint. In addition, the majority of the consultant body had not engaged with the issues likely to underpin coding and information flows.

In order to ensure the trust would be able to support focused improvement work a detailed casenote review study of over 200 consecutive deaths was undertaken on the premise that, regardless of the actual HSMR figure, a substantial percentage of adverse events and inpatients deaths are preventable.

Our casenote review process was set up to identify these preventable deaths. We also wanted to examine the areas requiring most improvement on the basis of analysing themes relating to failures of care at the trust. Even if the deaths were not preventable, the premise was that there would be useful learning points in terms of targeting improvement activity. The study was focused not on identifying error but on identifying suboptimal care. This focus differentiates this study from many others looking at preventable deaths.

The review format engaged a wide range of consultants who volunteered after a “call to action” e-mail asking for help to understand the issues relating to deaths in “our hospital”. They agreed to undertake a systematic and very detailed approach to casenote analysis using a series of structured questions; these would take reviewers through the notes such that they were enabled to form judgements.

These judgements were made in relation to overall care quality (excellent, good, adequate, room for improvement) and the preventability of the death (not preventable, possibly preventable, probably preventable). Reviewers were asked to look for evidence of “failure to rescue”, “failure to plan” or “failure to communicate”, and to comment on general issues of concern or notable care. They were then asked: “Is this the care you would wish for your own family?” and: “Could this admission have been avoided?”

The power of analysis

Using casenote review, the study has provided the trust with objective information regarding patterns of harm and mortality, which has been used to inform improvement work.

Our judgement was that around 6 per cent of deaths may have been preventable (in line with international studies quoting percentages of possible preventable death in hospitals).

This should be set in the context of the subjectivity of this assessment but has usefully framed the discussions and plans currently in construction.

We found this study’s power lay in the detailed examination of the records required, the wide engagement from medical staff and the questions that focused the reviewers on areas likely to be important for hospital care.

The question that forced reviewers to consider whether this was the care they would want for their own family was particularly useful in identifying issues. The meeting to discuss findings was also an important forum and added considerable value. The trust-specific questions that linked the review process to other focused work in the trust were also helpful.

Inevitably the study had some limitations but as a tool for improvement work and a method of engaging consultant staff in important care quality issues it has proved invaluable. It has also helped us to engage with our clinical commissioners to look at addressing the avoidable admission issue.

System-level learning has resulted such that an ambitious programme of work has been set out as part of the trust’s patient safety strategy and the redesign of emergency care project to improve standards using a defined set of key projects with investment in clinical leadership.

The HSMR has fallen significantly over the last 12 months due to work done in several areas. As this review process proved such a useful way of focusing improvement activity and engaging clinical staff in significant issues, the trust plans to repeat it regularly.

Dr Sonia Swart is medical director at Northampton General Hospital Trust.