A news article this weekend gives rise to a common set of questions around what we can broadly call “risk assessment” that you often hear debated in police stations, usually just after officers have been despatched to take a report of a missing patient. On Saturday afternoon, a convicted murderer Daniel ROSENTHAL absented himself from hospital whilst he was on an unsupervised walk around the gardens of Tatchbury Mount Hospital, in Hampshire. He was found on Sunday morning in Southampton and returned to the hospital.
Hampshire Police were immediately notified of Mr ROSENTHAL’s absence and commenced a “High Risk” missing person’s enquiry, including Press Releases suggesting that the public should not approach the man, but ring 999, etc.. Without knowing the details, we could see from the way in which the media was released that there was a significant concern.
Many members of the public commenting on social media and I’m sure many others were asking, “If he was such a risk that no-one should approach him, why was he unsupervised and fully at liberty to leave?”
Most missing patients are not murderers! – in fact, most are not people who have been involved in the criminal justice system at all. But police officers see this sort of thing frequently: patients granted leave of absence from hospital who fail to return and the hospital’s estimation of risk invites the question “so why grant leave?!”
Most police and NHS organisations for the purposes of AWOL patients agree on a low, medium or high categorisation of patients. High risk usually means life could be at risk, whether this is a risk of suicide or homicide or both; medium risks usually cover patients who will become unwell if they don’t take medication, if they take drugs or alcohol or there is potential for them to self-harm. Low risk patients are usually not thought likely to present a risk of harm to themselves or others but may be vulnerable for a range of reasons around medication or their current cognitive or emotional state, etc., etc..
Firstly, can we agree? – most patients detained under the MHA, even including some who committed (serious) offences en route to their detention in hospital, are not going to spend the rest of their natural lives detained in hospital. If we do agree about this, then we have to accept that there will be decisions about treatment are care which will involve taking risks around the granting of leave and whether or not leave will be supervised or unsupervised, whether it will be tightly controlled, subject to conditions or more loosely managed. If we accept that decisions will be taken about the granting of some leave, do we accept that leave will gradually be subject to fewer restrictions and potentially granted for longer periods? Can we agree, especially as we are talking about patients, not prisoners, that this is a necessary part of recovery and rehabilitation.
THEORY AND PRACTICE
Of course I know that most people get this, at least in theory. In practice we have heard examples of leave being granted in circumstances where risks were insufficiently understood and where questions have been rightly asked:
I always think of the decision in the case of John BARRETT who was given a restricted hospital order in 2002 after committing several assaults in London. Having been conditionally discharged from hospital in 2003 – after a due process of recovery which involved the granting of leave – he again became unwell and was re-admitted. He was granted leave from hospital by a doctor on 01st September 2004, without that doctor actually having assessed or even seen him and without understanding his risk background. He murdered Denis FINNEGAN in Richmond Park on 02nd September.
But this is not an argument against ever granting leave for patients who may fail to return or who have risk histories. It is an argument for a public understanding of the benefits to us all of patients recovering sufficiently to live independent, valuable lives outside the confines of hospitals. Bear in mind that the re-offending rate for patients who were subject to restricted hospital orders is over 5%, compared to a prison re-offending rate of over 50%. So it is an argument for better risk assessment, better training and better scrutiny. I know, I know … we’ve heard this all before because when reviews are published into the treatment and care of people like John BARRETT, we read with shock and amazement that certain basic things weren’t done. Who grants leave to a patient previously deemed to pose “a serious risk of harm to the public” without at least examining them?! Yes, we wearily and hopeless read promises that “lessons have been learned” in order that “such things could never happen again.”
The one thing I notice when hearing officers talk about MHA leave which has broken down: it is sometimes similar in tone to the conversations we hear about criminals who were granted bail with conditions and broke them. But patients are not prisoners: leave is usually about recovery and rehabilitation, even if the route to admission was via the criminal courts. Where leave is granted to a patient and it fails, does this mean leave should never been tried again? The reality is that it should be, although with due regard to understanding the reasons why it broke down originally and in the face of the risks being understood and mitigated. But if we agree that our model of mental health care is rightly about a community model of care, then we are going to have to understand and accept that MHA leave will fail. What we can be robust about, is understanding that when it does fail, we need to learn the lessons as to why and build that into future decision-making; and we need to understand that if it does fail, the recovery of AWOL patients is not (just) a matter for the police!
Even accepting that Mr ROSENTHAL was a convicted criminal, convicted over thirty years ago, he had assumed quasi-patient status by the prison service transferring him under s47/49 of the Mental Health Act to hospital. He was no longer detained in a high or even medium secure unit: Tatchbury Mount Hospital is a low-secure unit so it may be inferred that his care plan was aiming for discharge and that the Responsible Clinician was satisfied that security risks were of an order to grant unsupervised leave. Does this mean if leave failed, by him absconding from the grounds that he is high risk? – yes.
Risk assessment is (able to be conceived as being) based upon the probability versus the impact of threats: by any standards, a convicted murderer who has absconded is going to be high risk, irrespective of whether he was acutely unwell and experiencing psychosis having recently been admitted; or whether he was sufficiently stable to be granted unsupervised leave and is looking toward being conditionally discharged from hospital.
Most murderers do not live out their lives in prison or hospital and so we need to grasp the reality that professionals employed on our behalf who have to take these tough decisions may from time to time get it wrong; but they may make decisions which are perfectly balanced and fair which still lead to a failure in the purpose and outcome.
Winner of the Mind Digital Media Award.