Tracey Crouch MP resigned as Sports Minister on 1 November 2018 citing the decision to implement the reduction in maximum FOBT stakes in October 2019 as the reason for so doing. She was applauded as being principled and courageous for her action. In her resignation letter she claimed “two people will tragically take their lives every day due to gambling-related problems”.  This appears a shocking statistic and has already been repeated several times since then by various media. But is it accurate?

An indication that the statistic was slightly suspect came in an interview she gave on BBC Radio 5 Live on Sunday 4 November. Crouch repeated the claim and justified it with the curious phrase that it is “the statistic that is out there”, rather than citing the source for the statistic.

There are certainly plenty of statistics “out there” but that does not make them accurate or correct. GBGC has attempted to verify the claim.

One statistic that is out there comes from the UK Office for National Statistics (ONS). In the 15 years between 2001 and 2016 the ONS reported 21 suicides “where the death certificate mentioned gamble or gambling”. 

The ONS states the limitations of this data: “the data is not considered completely reliable because the coroner will not always record detailed information regarding the deceased’s history.” 

But, even allowing for under-reporting, this is still of an order of magnitude lower that the statistics quoted by Crouch.

The source of Crouch’s statistic appears to be the website Gambling With Lives, which says: 

Research in the UK and Hong Kong found that 4-11% of suicides were related to gambling … equivalent to 250-650 deaths per year in the UK

The range of 4% to 11% is based upon two papers:
1. Wong, P.W.C. et al. (2010) A psychological autopsy study of pathological gamblers who died by suicide. Journal of Affective Disorders 120, 213-216
2. Appleby, L. et al. (2017) Suicide by children and young people. National Confidential Inquiry into Suicide and Homicide by People with Mental Illness (NCISH). Manchester: University of Manchester

Reading the original research highlights several weaknesses in extrapolating the data to determine the number of deaths claimed by the website.

Small sample size

Wong et al. considered a total sample of 150 in Hong Kong, of which 17 (11%) met the criteria for the diagnosis of pathological gambling.

Appleby et al. considered a sample size of 106 people aged 20-24 over a two-year time period, of whom four (4) (4%) were found to have gambling debt.

Hong Kong vs UK

In this complex area it is too simplistic to apply findings from the Hong Kong research directly to the UK. Hong Kong is a very different gambling market – the propensity to gambling is higher, the gambling spend per capita is higher, and the role of gambling in society is very different.

Based on the research cited, therefore, it is difficult to conclude an 11% rate in the UK.


The issue of problem gambling and its associated harm is an emotive one and a distressing one for the families of those harmed by their gambling.

It is, therefore, very disappointing that the former minister with responsibility for gambling policy should publicly quote such loose statistics about the harm caused by gambling. 

The papers’ authors are aware of the methodological difficulties in researching this complex area, with Appleby et al. stating that the study’s design means they “cannot be certain of risk factors and we cannot establish cause and effect.”

If effective gambling policy and responsible gambling measures are to be developed to help those with gambling problems and to prevent any gambling-related deaths, the debate must be based on verifiable evidence and research rather than what makes the best headlines.