Lightning strikes


After the terrible case of Nicholas Salvador who was found to be insane when he beheaded north London grandmother, Palmira Silva, the Daily Mail columnist and Doctor, Max Pemberton wrote an article pleading that everyone with a mental illness should not be immediately demonised as a potential axe-murderer.

It was a thoughtful piece but unfortunately totally compromised by his claim that:

“the risk of being killed by someone with mental illness is about the same as the risk of being killed by lightning”

Similar versions have also been circulated for a while now by journalists and some mental health charities, such as Rethink and Time to Change, but without ever giving sources, or offering any evidence to back it up.

So what’s the real story?

In July 2013 the National Confidential Inquiry into Suicide and Homicide by people with Mental Illness at the University of Manchester published official figures for the numbers of homicide victims of mental health patients or people with mental illness at the time of the offence.

NCISH 2013 p 129They found that in England and Wales between 2001 and 2010, 1035 people had been killed by psychiatric patients and people with mental illness.

Now the Office of National Statistics also collects data on deaths in England and Wales, which, somewhat surprisingly, actually includes those killed by lightning strikes. (For the technically minded it’s ICD-10 classification X33).

They found the total number of fatalities by lightning for the same period was in fact, just eight.

So Max’s claim is factually inaccurate by only around a factor of 130.



There are alternative versions of the quote, usually made by the charities (or sloppy journalists copying their press releases), claiming:

“You’re more likely to be struck by lightning than killed by someone with schizophrenia”

which is just as untrue.

The National Confidential Inquiry also publishes figures on the numbers of people with Schizophrenia who have been convicted of a homicide.

As I’ve written elsewhere the NCISH figures don’t quite give the full picture, as

  • they are counting offenders not victims (some offenders kill more than one person),
  • they only count people who have been convicted (so don’t include people who take their own lives after committing a homicide and so aren’t convicted – around 189 cases between 2001-10), and
  • they don’t include mentally ill people who were only in touch with primary care services or none at all.

But even with all these limitations, the NCISH figures show that between 2001 and 2010 at least 356 people with schizophrenia were convicted of a homicide in England and Wales.

Now the most accurate data on lightning strikes in the UK comes from Oxford Brookes University’s Tornado and Storm Research Organisation, which has a unique database of all recorded incidents since 1988.

They found 147 lightning strikes between 2001 and 2010 in England and Wales, affecting 227 people.

As we have seen eight of them were killed.



The oldest version of the quote I’ve found comes from an article by the psychiatrist George Szmukler from 2000, when he claimed (again without offering any evidence) that

“Especially frightening to the public is the prospect of being killed by a stranger with psychosis. In fact the risk of this is around the same as that of being killed by lightning”

Now the vast majority of mental health homicides involve people killing their partners, family members, neighbours or acquaintances, and so the numbers of those killing strangers is small and hardly representative of the true scale of the problem.

But even here NCISH has collected data (again just for convicted offenders, so the usual limitations apply) and they have found 86 cases where a mentally ill person or psychiatric patient has killed a stranger.

If we compare that to the eight killed by lightning, we see his claim again doesn’t stand up.



So to sum up in England and Wales between 2001 and 2010 we have:

  • 1035 victims of mental health homicides,
  • a minimum of 356 people with Schizophrenia involved in unlawful killings
  • 227 people hit by lightning
  • 86 mental health homicides involving a stranger
  • 8 deaths from lightning strikes

so in any variation at all this claim is totally factually incorrect and not supported by the evidence.

Now this isn’t just a statistical exercise.

Why this matters is because it seems to me to be part of a worrying trend to seek to minimise, deny and dispute the true scale and impact of these tragedies.

It’s to seek to explain them away, to try and suggest that the numbers are so small that they are hardly worth bothering with, and there is, in fact, nothing to worry about, and certainly nothing to learn or change in current mental health care or practice.

With over a thousand deaths these killings are certainly not rare.

Many of the victims are particularly vulnerable – young children, the elderly, and quite often mental health patients themselves.

I think what Max and the charities have failed to realise is that generally the public are well enough informed to understand that not everyone with a mental health problem is a potential killer.

But at the same time they understand, particularly after hundreds of these cases and inquiries, that some people with serious mental illnesses are often not getting the effective care and treatment they need, which leaves them in positions where bad things can happen.

And whilst people keep on trying to deny and minimise the problem, I reckon nothing will change and more innocent people will continue to lose their lives.




Thanks to Prof Derek Elsom for his help with figures from the TORRO Database and his forthcoming article Derek Elsom and Jonathan Webb. Lightning Injuries and Fatalities in the United Kingdom 2014 and a summary of personal-injury lightning incidents from 1988 to 2014 – Tornado and Storm Research Organisation – International Journal of Meteorology May Jun 2015 Vol 40 

The number of mental health homicide victims is from 2013 NCISH annual report Table 12 page 129

The ONS figures for lightning strikes is usually in Table 5.19 of the annual mortality statistics. The 2013 table is here

Figures for convicted homicide offenders with Schizophrenia from NCISH 2013, For England Fig 32, p 51; For Wales see NCISH annual report 2011, Fig 37 p 49 and NCISH 2012 p 59

Numbers of Murder Suicide from NCISH 2013 p 130

Numbers of MH Homicides of Strangers, NCISH 2013 – England Fig 27 p 45; Wales p 120 also NCISH 2012 p 59

One Reply to “Lightning strikes”

  1. Thank you for this excellent research, and demolishing these absurd and dangerous claims circulating in the press. They are like knife crime, where for the last five years everyone has been claiming they are dropping, yet it was quite clear from the stats they were increasing. Finally this has been recognised, but not before funding to tackle the problem has been cut.

Leave a Reply to Peter Sinclair Cancel reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.