Southern Health leadership condemned in report on patient deaths

The review of deaths of people with learning disability or mental illness who had been in contact with the trust was released this afternoon. It was conducted by audit firm Mazars for NHS England, in response to the death of 18 year old Connor Sparrowhawk, who had autism and epilepsy, at the trust’s short term assessment and treatment unit in Oxfordshire.

A draft version of the report’s key findings was leaked last week.

The full report says:

• “There has been a lack of leadership, focus and sufficient time spent on reporting and investigating unexpected deaths of mental health and learning disability service users at all levels of the trust including at the trust board.”
• A “lack of strategic focus relating to mortality” has contributed to deaths having “little prominence at board level”.
• “Too few deaths were investigated in learning disability and older people’s mental health services”. However, it is not possible to say how many deaths should have been investigated that were not, because national guidance on reporting deaths in these services is open to interpretation by trusts.
• “The systems in place provide no evidence that the trust has fully reported or investigated unexpected deaths or taken remedial action where appropriate.” The “inconsistent and incomplete nature of the data supplied” also supports this conclusion.
• “A lack of board challenge to the systems and processes around the investigation of deaths”, including how decisions are made on whether to investigate deaths.
• Investigations that were carried out were of poor quality overall, making it difficult to learn from these deaths. “There is little evidence that there was any effective effort to improve the quality of the reporting until very recently”, and reports were not rigorously challenged.
• No effective action was taken by the Southern Health board on the standard of investigations, despite repeated warnings from coroners.
• Although commissioners had raised concerns over Southern Health’s reports, they were also criticised for “insufficiently strong enforcement or attention paid” to delays in reports being published, and the quality of reports.
• An “ad hoc and inadequate approach” to involving families and carers in investigations. Two thirds of investigations relating to mental health did not involve families. Family involvement relating to learning disability was “negligible”.
• A failure to identify or report a drop in the number of deaths being reported as incidents, despite a there being an overall rise in overall patient safety incident reporting. “This is likely to mean that… decisions on risk and safety are made which may be based on incomplete or inaccurate data.”
• None of the initial assessments of learning disability deaths had action plans.
• The trust did not have a systematic approach to producing action plans, reviews and changing services in response to deaths.
• The trust did not follow its own procedures for serious incident reporting, which it adopted voluntarily from the now defunct National Patient Safety Agency.
• Although Southern Health does not have a higher than average mortality rate, it appears to report fewer deaths of mental health service users than other trusts in the region, and compared to the national average. “It is likely that the low level of reporting is a function of the trust reporting practice.”

Controversial data on the numbers of deaths investigated was included in the final report. These featured prominently in the section that was leaked in draft form last week, and were disputed by the trust.

The final report notes that there is no statutory requirement for trusts to report or investigate deaths, except for reporting to the Care Quality Commission. Within the scope of the report, Mazars found that the trust had met its responsibilities under regulation.

The data was disputed in part because Southern Health may not have been the main health provider caring for patients when they died, and so might not have been the most appropriate organisation to investigate the deaths.

However, a new key finding in the final Mazars report, which was not present in the draft, says: “If the trust determine that another part of the system should undertake [an initial assessment of the death] it has a responsibility to ensure the incident is reported to the commissioners.”