“We have a lot of Incidents”: Why the Investigation Process for Mental Health Deaths Must Change

My sister Sophie O’Neill

The phone call

“Unfortunately I’ve got some bad news about Sophie,” my mum said over the phone. I knew my sister had been in a mental health facility in London for nearly two weeks. She had a diagnosis of schizophrenia but was generally well and lived an independent life. I prepared myself for the worst. Perhaps there was some bad news about her health, or she needed to stay in the hospital longer than expected.

“She’s passed away,” my mum said.

In just seconds, those three words sucked all the reason and logic out of the universe. Sophie couldn’t be dead. Just a couple of months before, we had spent the day together in London. We had visited the Sherlock Holmes Museum, eaten overpriced avocado on toast, and saw the Food: Bigger than the plate exhibition at the Victoria and Albert Museum.

Sophie had been full of life and a talented artist, having studied at some of the most prestigious art colleges in the UK. Though unable to be in paid employment for much of her adult life, she was constantly producing artwork and working towards either a course (completing an MA in Public Sculpture) or her own projects. She was an active member in a local arts organisation for people with mental health challenges, acting as both a trustee and volunteer. I was 35. She was 37.

“Was it suicide?” I asked. It was the only explanation I could think of. Sophie was physically healthy, she didn’t smoke, played ping-pong, and did yoga.

“Suicide…no.” My mum sounded surprised. She said the hospital had told her there would be a full internal and external investigation. The cause of death was not known.

Sophie and me together in London, a couple of months before her death


I held onto the idea of a full internal and external investigation throughout the following month. In that time, I flew back and forth between China, where I had been living, and the UK, three times. I spent Christmas with my UK family; the empty space at the table far more gaping than the lack of presents under the tree. But throughout it all we could, at least, focus on finding out the truth of what had happened.

In late January, we visited the City and Hackney Centre for Mental Health in London. We found out from the clinical director that no investigation into Sophie’s death had started after nearly five weeks. No witness statements had been taken, no one had been interviewed. Her death had merely been recorded in the notes, as if it was a normal incident.

Sitting in that stark white room, filled with grief and disbelief from my brother, my mum, and me, and the appeasing expressions and evasive remarks from the clinical director, a thought crossed my mind: Is this a normal incident?

“How many people die unexplained in your hospital?” we asked.

“We have a lot of incidents,” was the answer.

What is an incident?

The word “incident” was repeatedly used as a euphemism for Sophie’s unexplained death. When the NHS investigation into Sophie’s death eventually started (it still hasn’t finished), it was called an SI Review or Serious Incident Review.

If the clinical director wouldn’t tell us how many deaths happened in the hospital, perhaps the media would give us some context. Later, it didn’t take us long to find the following:

Guardian: NHS to look into deaths of more than 100,000 mental health patients a year

Guardian: Hundreds of mental health patients died after NHS care failures

BBC: NHS trust ‘failed to investigate hundreds of deaths’

Telegraph: Hundreds of mental health deaths may not have been investigated

Telegraph: NHS ‘failed to investigate 1000 deaths’

Community Care: NHS trust investigated ‘too few’ deaths of mental health and learning disability patients

From the moment we started trying to get answers about Sophie’s death, I had the feeling that it was as if she had died in prison. After all, had she died suddenly and unexpectedly outside of state detention, it would have been the job of the police to check that her death was not suspicious, and investigate it if it was.

The mental health centre where Sophie lost her life

How are deaths in mental health facilities investigated?

When an unexpected death happens in a mental health facility, the process is largely as follows (this is what should happen, but media reports confirm internal investigations do not always take place):

  1. The death is recorded in the notes by the hospital staff.
  2. The death is reported to the coroner.
  3. Post-mortem results are collected and sent to the coroner.
  4. NHS conducts internal “Serious Incident Review”. Results of Serious Incident Review are sent to the coroner.
  5. Coroner conducts inquest.

Why internal investigations are a serious flaw in the process

In case you consider it acceptable that NHS trusts are allowed to investigate themselves over unexplained deaths, consider the following hypothetical examples. If these situations actually happened, would you consider them reasonable, and would you trust the results?

  • Donald Trump independently investigates whether there are any issues with his tax records, and posts the results on Twitter.
  • Apple is accused of poor labour conditions in its supply chains. It issues a statement reassuring customers that it has asked all the factories it uses to investigate themselves.
  • A private detective is employed to investigate a murder he himself stands accused of.

I believe that most people would be skeptical of the results of the investigations above. Why then, are we expected to trust the results of NHS investigations into NHS failings?

On more than one occasion, a senior member of the SI Review panel said their goal was not to “apportion blame.”

Let’s go back to our hypothetical examples. Would we trust Donald Trump if he said he would look into his tax records, but would not be looking to apportion blame, or Apple if they said they would investigate labour abuses, but would not apportion blame, or the private detective if he said he would investigate the murder, but would not apportion blame.

Only an effective and objective inquiry into the truth can truly be called an investigation. If a death was related to neglect, malpractice, or even abuse, how can an investigation seeking not to apportion blame find the truth about that neglect, malpractice, or abuse?

We were also told that the SI Review was not looking to find the cause of death (that is the job of the coroner) but to find if anything was wrong with the care.

Let’s put those two statements about the Serious Incident Review together. The SI Review:

  • Is not looking to apportion blame
  • Is investigating if anything is wrong with the care

Surely, an investigation that is trying to find out if there was anything wrong with the care, but is not looking to apportion blame, will not find that anything was significantly wrong with the care. Even if something was significantly wrong, the investigation would not find it, because it would be impossible to find something seriously wrong with the care without apportioning blame.

Even prisons don’t investigate themselves

A robust mechanism for pre-inquest investigation matters because the results of the investigation are what is presented as evidence, along with the post mortem, at the coroner’s inquest. Only objective and accurate evidence can facilitate an objective and accurate verdict.

While putting together this petition, I discovered that there are actually independent bodies who investigate deaths in prison and police custody. The Prisons and Probation Ombudsman is responsible for investigating prison deaths before the inquest. No such mechanism exists for deaths in mental health facilities.

Previously, I had been thinking “It’s almost as if she died in prison.” After reading this report I started thinking “It’s worse than if she died in prison.”

Why are mental health patients treated in this way? Mental health facilities should be safe places where patients are respected and can receive effective treatment.

Mental health facilities cannot pretend to offer effective treatment if they cannot offer safety. And the NHS cannot claim to offer a safe environment if unexpected deaths are not independently investigated. Please share this story or sign this petition to change the system so all mental health deaths are properly investigated.

I write about mental health issues. Sign my petition for fair investigations into NHS mental health deaths here: https://petition.parliament.uk/petitions/333684