Once again we are forced to read that a brutal killing was caused by a “cumulative failure” of state institutions. This time it is not a parole board but a mental health trust and tribunal that have been exposed as a danger to the public which foots their bills. It is clear that John Barrett, a paranoid schizophrenic, should never have been free to kill Denis Finnegan in Richmond Park in 2004. The question is why no one has ever been disciplined for this lapse and why the secure unit involved has been allowed to blunder on.

The independent inquiry by South West London Strategic Health Authority into the care and treatment of Barrett is devastating. His conditional discharge by a tribunal came barely a year after he had almost killed a man, in the absence of his medical officer and in opposition to the wishes of the Home Secretary. His psychiatrist’s decision to give him an hour’s leave from the secure unit at Springfield Hospital was “fundamentally flawed” given his history of violence. And “too much confidence was placed in clinical judgments unsupported by evidence and rigorous analysis”.

This murder came barely a year after another schizophrenic from the same hospital beat a nurse to death, and the year before two more of its dangerous patients went on the run. The St George’s Mental Health Trust, which is responsible for the hospital, claims to have improved its pro- cedures. The inquiry categorically doubts whether the trust has the capacity to remedy its deficiencies. This clear difference in opinion demands investigation by the Home Office.

There are wider lessons here, too. Murders by the mentally ill are not uncommon: there have been more than 300 inquiries into such murders since 1994, when Jonathon Zito’s widow campaigned for these inquiries to be made automatic. Yet the mental health charity SANE believes that one in three of these homicides was preventable. Its analysis of 69 inquiries into homicides by the mentally ill found that, in half the cases, mental health professionals simply ignored warnings from family and friends. This was certainly true in the Barrett case, where his partner had become alarmed.

The Barrett inquiry describes a culture that values “engaging” with dangerous patients over protecting the public — despite clear Home Office guidance to the contrary. It describes an almost wilful lack of communication and supervision by professionals, which campaigners say is common in units where eminent psychiatrists may spend half their time on teaching and research.

The Barrett inquiry states that a lack of resources and legislation is not the problem. But it seems likely that government targets do contribute to the reluctance of mental health professionals to admit patients to hospital, and to keep them there. Crisis resolution teams are dedicated to keeping people out of hospital, which is expensive and indeed is often disconcerting for the patient. But when success in psychiatric care is defined as the number of days not spent as an in- patient, risks will inevitably be taken. Barrett should not have been discharged in 2003, and should not have been readmitted in 2004 only as a voluntary patient. Tragedies happen when tribunals and professionals act on partial information that suits their own sometimes too optimistic view of a complex world.

Explore Health News

• Swine flu

Times Recommends

• How I lost the gift of joy for a while
• Why Ann Widdecombe has got it wrong
• Alas, it’s the end of the road for petrolheads