There is one gambling-related suicide in the UK every day. There are up to 496 gambling-related suicides a year. Ten per cent of all the suicides in England are caused by gambling. 

These statistics, and other iterations of them, have become mantras for the anti-gambling lobby since January 2023 when the Office for Health Improvement and Disparities (OHID) published a report claiming that there are “between 117 and 496 suicides associated with problem gambling” in England. Activists naturally focused on the larger of these two numbers and started putting it on billboards. The monetised value of years of life supposedly lost to suicide make up most of the “up to” £1.77 billion that gambling is said to cost “wider society” each year.

It turns out that these figures are based on nothing. They are a will o’ the wisp. A mirage. They exist only on a laptop in Whitehall. They are worthless.

How can we possibly know how many suicides are linked to problem gambling, let alone how many are solely caused by it? Gambling is only mentioned on one coroner’s report a year, on average, which is presumably an under-estimate. In the absence of better evidence, OHID’s predecessor Public Health England turned to a study from Sweden which looked at 2,099 hospital patients who were diagnosed with pathological gambling between 2005 and 2016. Sixty-seven of them died, including 21 who took their own life. The authors noted that the suicide rate among this cohort of pathological gamblers was fifteen times higher than the suicide rate of the general Swedish population. 

Upon this sliver of evidence, everything else rested. In 2021, Public Health England simply estimated how many problem gamblers were in England and then multiplied the number of expected suicides by fifteen. This produced a figure of 409 suicides a year which anti-gambling activists then put on T-shirts. 

Public Health England was closed down soon afterwards and replaced by OHID. Last January, OHID used the same methodology but produced two different estimates, one based on how many people are thought to have “gambling disorder” (previously known as pathological gambling) and the other based on how many suffer from the less severe condition of “problem gambling”. The figures were 117 and 496 respectively.

You don’t need to be intimately acquainted with basic statistics to see the problem here. People who are being given medical or psychiatric treatment in hospitals are inherently different to people who are not. If you are admitted to hospital, there is already something wrong with you. If you are admitted to hospital and asked to take a survey to diagnose gambling disorder (or any other psychological problem) then you are very likely to be at the higher end of the risk spectrum.

Sure enough, there was a lot wrong with the 2,099 people in the Swedish study. Between 2005 and 2016, 65 per cent of them suffered from “injury, poisoning, and other consequences of external causes”. 60 per cent had an anxiety disorder. 51 per cent suffered from depression. 41 per cent had a substance-use disorder. 29 per cent had an alcohol-use disorder. 19 per cent had a personality disorder. 19 per cent intentionally self-harmed. 12 per cent were bipolar. 9 per cent had schizophrenia. In the context of all this human misery, a suicide rate of one per cent does not seem too surprising and it is absurd to assume that all the suicides were the result of problem gambling. For many of these unfortunate people, gambling may have been the least of their worries.

The authors of the study freely admitted that these hospital patients were unlikely to be representative of the average problem gambler:

It is therefore likely that results may be skewed toward a population of individuals with more severe forms of GD [gambling disorder]. It is likely that this once again implies that this study sample might contain patients with higher mental health comorbidity, as well as individuals with more severe forms of GD, since these individuals are more likely to receive specialized psychiatry care.

Public Health England and OHID ignored all this and extrapolated the suicide rate among pathological gamblers with multiple co-morbidities in Swedish hospitals across the estimated number of problem gamblers in the general population in England. No attempt was made to adjust for the many other risk factors for suicide that these people obviously had. 

Last year, however, one of the two authors of the Swedish study did exactly that. Using the same dataset in a new study for her PhD thesis, Anna Karlsson found that “gambling disorder did not appear to be a significant risk factor for the increase in suicide and general mortality when controlling for previously known risk factors”. She concluded that her research “could not determine whether GD [gambling disorder] is an independent risk factor for suicide”.

This does not mean that there is no link between gambling disorder and suicide. History and common sense tell us that people who get into severe financial difficulties are more likely to take their own lives and it is obvious that problem gambling is one way to suffer financial distress, albeit the only one that is now treated as a “public health” issue. What it does mean is that gambling disorder, on its own, was not a big enough risk factor for suicide to show up among the people studied by the Swedish authors using standard statistical practice. If you extrapolated the properly adjusted figures from the Swedish study across the English population, the number of gambling-related suicides would be zero.

It is hard to believe that OHID was not aware that it had made an error that a literal schoolboy could have spotted. Public Health England was closed down because it was incompetent and was too easily distracted by lifestyle issues when it should have been focusing on public health. It was more of an in-house lobby group than a serious scientific agency. It seems that closing it down and re-opening it under a new name with the same staff was not enough to make the leopard change its spots.