March 2022 - Losing the information war

The next meme is that of the cost of gambling harms.

 

The economic and social cost of harms associated with gambling in England, 2019 to 2020 published by Public Health England (PHE) should go down in history as one of the most flagrant attempts to attack the gambling industry under the cover of so called academic research this decade. It is little wonder that PHE would be shut down a few months later.

 

The report estimated the annual economic burden of harmful gambling to be about £1.27 billion (in 2019 to 2020 prices). This included £647.0 million (51%) of direct costs to government and £619.2 million (49%) of wider societal costs associated with suicide.[i]

 

In the Lords debate, the Lord Bethell mentions this figure as an unacceptable cost and Baroness Bakewell mistook problem gambling with gambling when paraphrasing the report: ‘the six harms caused by gambling. These are: financial, relationship breakdown, mental and physical health, employment and education, criminal and anti-social behaviour, and cultural’. [ii] We can be certain that as the war over Britain’s gambling progresses these memes, like the others examined above, will be repeated ad infinitum and, if the gambling industry’s representatives continue their current strategy, go almost completely unchallenged.

 

Anti-gambling activist academics have used the ploy of undertaking cost benefit analysis of gambling in the past. In America in the 1990s and early 2000s, when land-based casinos were opening up in many states, a number of overtly anti-gambling academics published research papers that argued that the social costs of gambling far outweighed the benefits.[iii] The usual methodology was to create a list of all the social woes possible that could be linked to gambling, however tenuously, and then allocate an enormous cost to them through the most generous estimation possible, regardless of evidence to the contrary. This would then, if benefits were even considered, be pitted against a desultory figure of the benefits of gambling, usually just reduced to employment and tax revenue which is then heavily caveated by statements about how better jobs and more taxes could be gained from alternative enterprises than gambling.

 

The late Professor Bill Eadington, the father of gambling policy studies stated:

 

many of the studies conducted to date have resulted in reporting a single number or a range of numbers—i.e. “. . . the cost to society is $15,000 per problem gambler per annum,” or “. . . the range of costs to society is from $9,000 to $50,000 per pathological gambler per annum,”—on the basis that such numbers should—indeed must—be measured and reported, regardless of their lack of context, difficulties in conceptualization and measurement, or the ease with which such reporting can be misinterpreted or misunderstood.

 

One problem that sometimes occurs is when researchers approach their analysis with preconceived biases, hoping for a higher (or lower) number because of their beliefs with regard to the seriousness (or lack of seriousness) of the issue.[iv]

 

Dr Douglas Walker, one of the most sane academics dealing with gambling, points out in his paper from 2008 that one of the biggest issues with measuring the social costs of gambling is co-morbidity.[v] This is the concept that problem gamblers have pre-existing problems other than gambling. Walker states:

 

Examples include alcohol and drug problems, and mood and anxiety disorders. If pathological gambling is simply a symptom of some more basic disorder, then, it is the more basic disorder rather than gambling itself that is the underlying cause of the adverse consequences and social costs of the pathological gambling.[vi]

 

Walkers argument is that given we know that a high proportion of problem gamblers have other conditions, how do researchers allocate costs? and why do activist academics allocate 100% of the social costs due to a co-morbidity, such as alcoholism, to problem gambling? This is very pertinent to the PHE report.

 

Walker goes on to pose other challenges with calculating social costs. He identifies the use of surveys as problematic due to the fact that problem gamblers are very unreliable when it comes to correctly remembering how much they’ve spent on gambling, when and where they got the money from.[vii] This is important as such surveys are used when calculating social costs. Walker then points out that many studies make the mistake of extrapolating from the experience of the most serious problem gamblers (those who can be classified as pathological) to the general population of problem gamblers, which is obviously inappropriate and appears in the PHE report, although it provides the rather fatuous division between at-risk and problem gamblers, the term at-risk implying an imminent progression to problem gambling when evidence suggests this is just not the case .[viii]

 

Walkers final critiques are of government spending and the counter-factual argument.[ix] He argues that the point of trying to identify the social cost of problem gambling is to argue this is the additional cost to the nation of gambling due to its social externality, problem gambling. On top of the issue of co-morbidity, this implies that that the government services researchers have allocated to the cost of problem gambling would not be there if there were no problem gamblers. Would the number of police, hospitals and prisons be in anyway different if the number of problem gamblers was zero?

 

Up until very recently the cost of all treatment for problem gambling was paid for by the gambling industry itself. Now that the NHS has opened some clinics, this is just a tiny fraction of a state cost compared to what the NHS spends on alcoholics and drug addicts, which we all pay for out of our taxes.

 

And what does the social cost of problem gambling compare with? The cost of no problem gamblers? How would that be achieved given that evidence from countries such as the now defunct GDR (East Germany) shows that there was still problem gambling. This shows up the integral flaw in cost benefit analysis, especially one like the PHE report, is that what it ignores is the counterfactual. If there is no legal gambling, you have illegal gambling, this means massive amounts of crime and problem gambling as there aren’t the regulations and enforcement to prevent it. We have ample evidence of this from looking at Britain from 1850 -1960. Every cost benefit analysis should be conducted with a comparison with what the social externalities would look like with a massive black market.

 

A cursory critique of the PHE report is as follows:

 

Financial Harms

• The financial data used came from the 2021 Lloyds bank report that didn’t include cash expenditure (e.g. practically all retail gambling), gambling purchases in non-gambling venues such as all of the National Lottery’s venues or include any winnings.[x] The report acknowledges that: we cannot tell if at-risk or problem gamblers were included in the sample, an important limitation in this costing exercise for our purposes.[xi]

• They reference a survey from the Annual Statistics from the National Gambling Treatment Service which shows that gamblers reported spending a median of £1,000 and a mean of £2,102 on gambling in the previous 30 days before assessment. What they don’t mention is such behaviour is undoubtedly what led the gamblers to seek treatment as this was their low point – as is generally accepted, addicts have to hit rock bottom before they seek help. They do caveat that ‘caution is needed when comparing such figures as the populations are not the same across different analyses’.[xii]

• PHE rightly conclude that ‘The limited evidence available meant that no costing analysis was possible for financial harms to the individual’.[xiii]

 

Homelessness

• There is an acceptance that there is little UK based research on homelessness and problem gambling. Considering the financial consequences of problem gambling it is not illogical to suggest that it may be a contributing factor to homelessness.

• Much of the analysis is based on a report that used a sample of just 72 homeless people to ‘establish if gambling is more commonly a cause or a consequence of homelessness’.[xiv] This report was based on an earlier report that had a larger sample (456) which asserted that that the rate of problem gambling was far higher in the homeless community than the general population.[xv] Both reports were based on samples from hostels in Westminster, London.

• The main report states that 16 participants (61.5%) disclosed experiencing gambling problems prior to becoming homeless but does not give this as a reason for the homelessness. Witnessing violent acts, long term unemployment, losing a job and trouble with the police are the most reported negative life events by those considered some risk of problem gambling. This ‘some risk’ category is anyone with a PGSI score >0 which includes a large number of people having no negative consequences from their gambling

• The homeless people sampled also had a high rate of comorbidity with substance abuse. Of the thirty participants that disclosed any drug use in the past month, the most common primary substances were cocaine/crack (36.7%), heroin (33.3%), and cannabis (20%).Twenty-six participants (36.1%) endorsed at least one DSM-IV alcohol item, and in that subgroup, the modal number of items endorsed was all nine items (30.8%).[xvi] All the report mentions is that problem gambling and substance abuse are not correlated even though a high proportion did both and that they had similar engagement patterns

• What is not apparent from the research paper is evidence for the researchers claim ‘Data from the current study reinforces the assertion that problem gambling is a significant issue within the homeless population, and is more commonly a cause than a consequence of homelessness’. All we have is the data that shows that 61.5% (16 people who had a PGSI of >0) have suffered gambling problems prior to being homeless. There is no data of when or to what extent they suffered these problems and as stated the loss of employment would appear to be a more convincing cause. Two of the researchers have publicly anti-gambling views and one has an academic career based purely on gambling and homelessness and this may be the reason for the methodological leap that supposedly proves their stated hypothesis.

• The PHE extrapolates the results of this 72 person survey up to the homeless population for the UK (estimated by Shelter to be 249,850 in 2021) and estimates the cost of excess statutory homeless applications associated with problem gamblers. It estimates that there are 21,438 excess statutory homeless applications associated with at-risk and problem gambling in England equating to a cost of £62.8 million (£2,929 per application).

• The PHE estimates are based on all at-risk and problem gamblers – this includes low-risk and medium risk gamblers – i.e. not problem gamblers

• The PHE estimate is based on the concept that problem gambling is the cause of homelessness, while we are certain it is a factor, the research used to argue its case doesn’t prove it

• The PHE estimate does not incorporate that other factors (co-morbidity) such as alcohol and drugs and unemployment may be contributing factors as well but makes no attempt at allocating a financial amount for this

• The PHE report admits its limitations in that it doesn’t include rough sleepers, that its London based sampling does not make it nationally representative and that it is based purely on a male sample.

• The PHE’s statement that ‘Nevertheless, this remains the best evidence available’ seems to be their justification for what is potentially methodologically flawed and/or ideologically driven research.[xvii]

 

Health Harms – Suicide

 

• This has been considered above – the number of suicides the PHE report uses is based on Swedish estimates and Sweden has double the suicide rate of the UK. Cultural attitudes to both gambling and suicide are very different in Sweden compared to the UK.

• The PHE report cites the Gambling With Lives figures which are dismissed above and is indicative of the PHE report’s ideological bias

• The PHE report cites the Wardle 2019 report.[xviii] This states that problem gamblers are more likely to think about suicide and attempt it then non-problem gamblers. This seems highly logical as these are people suffering a mental illness. The report also mentions that these people had higher rates of impairment, poor mental health and substance dependence, and exposure to a range of stressful experiences, including debt and homelessness and that the problem gamblers’ odds of having suicidal thoughts/attempts in the past year were halved once these other factors were taken into account.

• The Wardle report gives its own limitations as ‘The survey was conducted over a decade ago and the sample was relatively small, with just 172 people identified as at risk of problem gambling and 41 identified as problem gamblers. Both problem gambling and past-year suicide attempts/thoughts are relatively rare, and the analyses were underpowered. The survey is cross-sectional in design and the data cannot be used to establish temporal sequencing in problem gambling and suicidality, nor causality’ and concludes: While it is not possible to conclude from these analyses that problem gambling causes suicidality and self-harm, the results do show that problem gamblers are a vulnerable group warranting targeted support who are more likely than others to have suicidal thoughts and to harm themselves.[xix]

• The Swedish study used to provide the figure of 409 suicides is based on a study of 2,099 Swedish problem gamblers that were receiving medical treatment.[xx] While problem gamblers had 1.8 times increase in committing suicide (based on 21 suicides) compared to the population, the report’s conclusion was ‘Although common mental health comorbidities did not predict overall mortality, depression predicted suicide death.’

• It is very clear that there is no evidence of problem gambling being a direct cause of suicide. It is more likely that depression is the main cause and that problem gambling may be a symptom of that.

• The PHE report stretches the boundaries of reality when trying to calculate a cost for problem gambling related suicide. It comes to the cost of £619.2 million based firstly on the number of suicides it attributes to problem gambling, which, as said before, is a) of unproven causality and b) based on Swedish numbers that are mostly to do with depression. Then secondly, it uses a methodology used by the Department of Transport for the bizarrely named concept ‘willingness-to-pay to avoid fatality’.[xxi] This is a methodology for Cost Benefit Analysis in transport infrastructure schemes that estimates the costs of traffic accidents so that the costs of road safety measures can be put against them.  In the version the PHE uses, the number of healthy years these suicides would have lived if they had not committed suicide is estimated and then allocated a cost - £60,000 for each year, making each suicide worth £1.51 million and the total cost for suicide as £619.2 million

• The report acknowledges that Swedish numbers may be problematic but doesn’t consider the differing suicide rates or cultural differences

• Rather obvious in terms of critiquing this approach is that there is no direct evidence of problem gambling causing suicide, yet there is of depression. As stated numerous times, problem gambling can be a severe mental illness and severe mental illnesses can lead to suicide, so there is of course some linkage, but not what we would consider worthy of allocating the full value of a fatality as per the Department of Transport methodology. Those killed in road traffic accidents suffer the misfortune of an accident. A suicide is a fatality caused by a person, the reasons for which are multiple and incredibly complex.

• This is a prime example of scant evidence being shoe-horned into a methodology to come up with a big number

 

Health harms – depression

 

• Yet again the report uses scant research, a paper about depression in young adults in Canada.[xxii] It argues that ‘health care professionals should know that at-risk or PG [problem gambling] can lead to incident mental disorders and illegal drug use, as well as illegal drug use can lead to new onset at-risk or PG’.[xxiii] The research involved young adults (aged 18 to 20 years) who were surveyed at 4 time points across a 5-year span from 2007 to 2011. As stated before there is well established research that shows that problem gambling is linked to mental illness, in some cases it is the actual mental illness (psychopathic gambling), in the majority of cases it is a symptom of a mental illness (usually depression).

• The next study, also of young people (17-24) is at least based in Britain.[xxiv]It states that ‘In conclusion, although many young people gamble without any harm, a significant minority (mainly males) show problem gambling behaviours which are associated with poor mental health, involvement in crime, and potentially harmful use of drugs and alcohol’.[xxv]

• Both pieces of research acknowledge that young men, who are by far the majority, who are PG are likely to be of low IQ, financially disadvantaged, have parents who gamble, use alcohol and/or drugs and have a mental illness of some sort. To the lay person and not the activist academic this research does seem to reflect risk-taking problematic lifestyles of problematic young men and is not evidence of gambling causing depression. Although there is no doubt that for many with depression they gamble problematically and for a small number, their mental illness focussed on gambling obsessively.

• The PHE report takes the methodological leap of referencing two research papers specifically about young people and then applying their results to people of all ages with no mention that research into young people gambling is undertaken as this is the time most people are considered at risk of harm.

• The PHE report estimates that the number of people with depression associated with at-risk and problem gambling  is 212,511. This is a large number which is not surprising considering that at-risk gamblers includes 1.2 million people who are low risk (which is arguably no risk!) and the survey the numbers come from was done in 2018 when problem gambling rates were twice as high. But the advantage of such a large number, when multiplied with the average cost of everything to do with treating someone for depression, regardless of the fact that depression by its very nature is an illness caused by multiple factors, causing less or more harm to some and less and more cost to the NHS, is that you get a large amount supposedly attributable to gambling harms - £335.5 million.

• Even the PHE report admits the limitation that ‘the analysis does not differentiate the level of harm between at-risk and problem gamblers. We would expect to see varying levels of risk between types of gambler, given that low-risk gamblers, for example, are likely to experience fewer or no negative consequences from their gambling activity’.[xxvi]

• The PHE report again ignores the counterfactual, would all these people not have had depression if they didn’t gamble? We know that those that suffer from a mental illness that in part focusses on gambling are very unhappy people and would logically have some depression because of it, but this report is obviously not looking for nuance, but to make the statement, gambling must lead to problem gambling and this costs the country millions. Just the facts don’t stack up

 

Health harms – alcohol dependence

 

• It’s almost as if by this point in the report they just gave up any pretence of methodological integrity. They used the study of young Canadians used in the depression section above to calculate the association between gambling and alcohol dependence which it did consider.[xxvii] They even acknowledge that ‘Given the lack of available data, we assumed that the alcohol dependent prevalence rate is constant across all age groups’ which it obviously isn’t![xxviii] They then calculate the number of people with alcohol dependence associated with at-risk and problem gambling to be 28,312. They then estimate the number that receive alcohol treatment in England to be and multiply that by the cost of that treatment, £4.7 million.

• This would be logical if there was hard evidence that problem gambling caused alcoholism. What we do know is its linked, so yet again to allocate 100% of the cost is being unfair and just more evidence of big numbers being added to a total to make an ideological point. Again there is no counterfactual, would these problematic young men leading problematic lifestyles not have an alcohol problem if they didn’t have a gambling problem?

• Yet again the PHE report acknowledges this section has serious limitations. But rather than stating that there wasn’t enough research to make the assertions it does, they just go with the extremely low bar in order to come up with a number. They admit the Canadian research is looking at people from a) another country, and b) who are young and more prone to the issues being measured, and c) at-risk and problem gamblers who therefore run the whole range of PG severity, from practically nothing to pathological but with the majority at the practically nothing end. They also assume ‘that a gambler who is alcohol-dependent has the same behavioural characteristics as other alcohol-dependent individuals’.[xxix] Another case of comparing apples and pears and excusing it because there is no other research.

 

Health harms – illicit drug use

 

• As with all the work on health harms, the foundational research is scarce and when considering illicit drug use they use the research used in the depression section.[xxx] They take this research into young people who are at risk (e.g. a lot of people basically at no risk) or PG and make their assumptions, this time, purely based on 17-24 year olds. Obviously the extrapolating young people to the whole population had become too uncomfortable. They then forget that the Foundation research asked the young people about whether they took illicit drugs and only separated these from Cannabis. This meant that ecstasy, a low harm lifestyle drug with mass use amongst young people was put in with crack cocaine, heroin and crystal meth. In trying to look for a cost of illicit drug use they use the costs and the numbers for opiate and cocaine users only. This means smaller numbers, compensated by the larger numbers of the foundation research but higher costs as hard drugs cost more to treat. They estimate 712 young people receive community treatment for drug use associated with at-risk or PG at the cost of £2m.

• The issue with this is the same with all, scarce research, apples with pairs, no basis for allocating all the costs and the counterfactual, would these drug users have taken drugs if they didn’t have  a problem with gambling?

 

Health harms – quality of life

 

• There is not enough research to consider Quality of Life costs

 

Health harms – gambling treatment

 

• Unsurprisingly for an ideological document, when it comes to gambling treatment there is practically no mention of the millions spent by the gambling industry. All the report states is that the main treatment providers are ‘funded by GambleAware, a large charity provider where fundraising activities include requesting the gambling industry to donate 0.1% of their annual gross’.[xxxi] There is no comment that GambleAware was set up by the gambling industry for the very purpose of funding these treatment providers amongst other things.

• The PHE also states that based on problem gambling prevalence numbers and the number seeking treatment, ‘this implies that just 0.5% of individuals meeting these definitions accessed treatment in 2019 to 2020. This potentially reflects the lack of access to, and provision of, gambling treatment services in England’.[xxxii] It doesn’t dare consider that a) their numbers don’t incorporate the largest treatment provider Gamblers Anonymous which doesn’t provide statistics on its membership, or the large proportion of PG’s who self-treat or, b) the methodology for prevalence numbers are faulty and over inflate the number of PGs and at-risk

 

 

Employment and Education Harms

 

• There is just not enough research (if any) to calculate any costs to Education

• This section of the PHE report starts by accepting there is very little research to attempt a costing. It will rely mostly on a 2016 Institute for Public Policy Research (IPPR) report that attempted to estimate the cost of problem gambling.[xxxiii] The PHE report comments on the IPPR report that ‘Despite its limitations, this is the only estimate made to date in the English context and no better data has been identified to test a different method’.[xxxiv]

• It is of interest that the PHE report doesn’t comment on the fact that a report, which precedes it and attempts to achieve the same goal, estimates the cost of problem gambling in the UK in 2016 to be ‘between £260 million and £1.16 billion’ while the PHE report, just a few years later, estimates it as (no range) £1.266 Billion.

• One of the reasons for the discrepancy of £583.4 Million (from the mid-range of the IPPR) could be that the IPPR focuses on the numbers of actual problem gamblers and doesn’t add in the at-risk gamblers (PGSI Score >0) which includes millions of people who are not problematic with their gambling.

• Maybe the other reason that the IPPR report, which would seem like an obvious touch stone for this report is almost ignored,  is that they state:

First, it is not possible from the literature to establish the extent to which problem gambling has caused problem gamblers to access services at a rate above that of non-problem gamblers. In the absence of existing studies, and in order to help us understand the extent of causal relationships between problem gambling and the presence of supposed ‘impacts’, we devised a short survey that was completed by five UK-based academics with expertise on problem gambling. Respondents were asked to provide their own judgement on the extent to which problem gambling is a causal factor determining increased demand for some of the services we have explored above. They were able to rank the extent of causality according to a score of 0 to 5. The responses support the conclusion that it is not possible to establish causality. For each of the areas of interaction that were tested, there was no consensus on the extent of any causality derived from the presence of problem gambling….

 

Taking into account the lack of existing studies proving causality, and related absence of academic consensus, this study makes no attempt to claim that excess fiscal costs incurred by problem gamblers are directly caused by problem gambling. Instead, we can say only that they occur alongside, and are associated with, one another.[xxxv]

 

Employment and Education Harms – Unemployment

 

• Copying the approach of the IPPR, the PHE manages to come out with a cost of excess cost of unemployment benefits associated with problem gambling as £79.5 million for England. This compares to the IPPR’s estimate of £40 - £160 million for the UK but £30 - £80 million for England.

• Of interest is that in the IPPR report which the PHE uses as its foundation, the IPPR state: – an average of 0.05 JSA claims per working-age adult in the UK. – while in the PHE calculations they state: The number of claims made per working-age individual is 0.09.[xxxvi] This is an almost doubling of the amount in 3 years which seems problematic. The PHE report states that the difference in the IPPR and PHE estimates for unemployment costs can be explained partly by the uprating in unemployment costs and ‘During this period, we saw 40% more claims for the general adult population (in our baseline calculations)’, yet provides no explanation as to why.[xxxvii]

 

Criminal Activity

 

• The PHE estimate that the excess cost of imprisonment associated with problem gambling is £162.4 million based on there being 3,799 additional prisoners incarcerated due to problem gambling.

• The PHE report is based on one piece of research that we have been unable to access but from its abstract looks to identify the prevalence rate of problem gamblers in prison based on surveys in one male and one female prison.[xxxviii]

• The main issue with this research is that as part of it they asked prisoners to self-report on whether they thought their gambling was the cause of their crime and these numbers are a critical part of the calculations. In explaining the limitations of the research, the PHE accepts that there is a potential problem in prisoners self-reporting but only mentions it in the context of the prisoners not being honest about which crimes they were prosecuted for, it doesn’t even consider that a prisoner may want to suggest to a researcher on problem gambling being a cause of crime that their crime was caused by gambling in an attempt to in some way mitigate their sentence. The PHE also mentions that drug related crimes are over representative in the sample – but doesn’t make the suggestion that maybe this is the reason for incarceration rather than gambling?

• So yet again the glaring issues of no proven causality or the counterfactual – would these people be in prison if they didn’t gamble?

• A layperson would surely ask to examining court records and see if gambling was in anyway involved in the crime may be a better research strategy

 

The PHE report would be innocuous if it were just some academic paper in a Journal focussed on how impossible it was to make any viable assertions on the cost of gambling harms because of a lack of research, no proof of causality (but definite associativeness) and the difficulties of co-morbidity when it comes to allocating costs and the counter factual. Before it was disbanded for incompetence, Public Health England was an anti-gambling advocate and its ex-employees and associated researchers remain so. The PHE report is a political document and is being used politically as can be seen from it being referenced in our Lords debate of the 1st March 2022.

 

If we look at its costs estimate:

 

Type of Harm                                                          Cost

Statutory homelessness                                      £62.8M

Deaths from suicide                                             £619.2M

Depression                                                             £335.5M

Alcohol dependence                                            £4.7M

Illicit drug use                                                         £2M

Unemployment benefits                                       £79.5M

Imprisonment                                                         £162.5M

​

Total Estimated Excess Cost                                 £1,266.1M

Taken from Gambling-related harms evidence review: the economic and social cost of harms, p.6

 

As can be seen, the biggest cost, 48.9% of the total, is the supposed costs of suicide. This massive cost estimate completely ignores that there is no evidence of causality, that the professional advice is to never attribute suicide to just one cause due to the complexity of the issue and even ignoring this and attributing the suicide to gambling, the issue of co-morbidity and the counterfactual is still ignored. Then we have to consider that the number of suicides is purely based on Swedish estimates where the cultural attitude is different and the suicide rate is twice as high as in the UK. Then we need to consider the way the costs have been worked out. They have used a formula use for the Department of Transport to do a cost benefit analysis when building transport infrastructure. By estimating the cost of an accidental  road death they can estimate the benefits of road traffic safety features built in to traffic infrastructure which helps in deciding budgets. This is a million miles from the tragic self-caused death of someone with severe mental health issues. It is shocking that almost half the supposed cost of gambling harms is based on the most tenuous of assumptions that don’t stand up to even the most basic levels of scrutiny.

 

The second biggest cost, 26.49% of the total, is the supposed cost of depression. This cost is based on taking the cost of treatment for depression which implies severe depression and applying it to a large number calculated from scant Canadian research into young people extrapolated up to a UK population and taking all those with a PGSI score >0, which includes hundreds of thousands with only minor negative consequences from their gambling and then estimating that an enormous 212,511 people were in need of treatment for serious mental illness. In 2020/21 a record 634,649 people completed the NHS Talking Therapy programme in the UK.[xxxix] The PHE therefore estimates that an equivalent of a third of the people that received therapy for mental illness in the UK would be those receiving treatment due to gambling. This is obviously a preposterously large figure due to a methodologically flawed approach unworthy of publication and that’s before there is any discussion of the lack of evidence of causality or the counterfactual.

 

Probably the only conclusion that can be drawn from this report is that if its authors: Annalisa Belloni, Shaun Donaghy, Brian Ferguson, Jonathan Knight, Alexandra Melaugh, William Naughton and Ruth Puig-Peiro, turned their hand to painting and decorating they would probably provide you with an estimate of £27 million just to put a few coats of magnolia on the front hall.

 

The PHE report needs to be withdrawn and apologies made for its outlandish attempts at politicising gambling harms.

 

Another concerning conclusion, is there appears to be a small pool of academics researching problem gambling. Many of who are publicly anti-gambling activists. What we see is poor quality methodology or early stage research being dressed up as evidence fit for policy making. It would not be surprising if these academics peer reviewed each other and accepted low research standards in favour of ideological campaigning. This and the public health campaign against gambling will be the subject of another article.

 

[i] Public Health England:  Gambling-related harms evidence review: the economic and social cost of harms, 2021, p.5

[ii]https://hansard.parliament.uk/lords/2022-03-01/debates/6DD9F37C-0912-4B58-8DF8-FEFA60730097/GamblingAdvertising

[iii] See Grinols, E. L. (2004). Gambling in America: Costs and benefits. Cambridge University Press, Grinols, Earl Mustard, David, Measuring Industry Externalities: The Curious Case Of Casinos And Crime, March 2001, Thompson, William, Gazel, Ricardo, Rickman, Dan, Social and Legal Costs of Compulsive Gambling, Gaming Law Review, Volume 1, Number 1,1997

[iv] William R. Eadington, University of Nevada, Measuring Costs from Permitted Gaming: Concepts and Categories in Evaluating Gambling’s Consequences, Journal of Gambling Studies, Vol. 19 No. 2, Summer 2003, pp.193-94

[v] Walker, Douglas M., Challenges that confront researchers on estimating the social costs of gambling. Report American Gaming Association (2008).

[vi] Ibid p.2

[vii] Ibid p.4&5

[viii] Ibid p.6. See Reith, Gerda et al (2010). Situating Problem Gambling: The Social Context of 'Gambling Careers': Full Research Report ESRC End of Award Report, RES-164-25-0006. Swindon: ESRC

[ix] Walker, Douglas M., Challenges that confront researchers on estimating the social costs of gambling. Report American Gaming Association (2008).p.7&8

[x] Muggleton N, Parpart P, Newall P, Leake D, Gathergood J, Stewart N. The associationbetween gambling and financial, social and health outcomes in big financial data. Nat HumBehav. 2021.

[xi] Gambling-related harms evidence review: the economic and social cost of harms, p.23

[xii] Ibid p.23

[xiii] Ibid p.24

[xiv] Sharman S, Dreyer J, Clark L, Bowden-Jones H. Down and Out in London: Addictive Behaviors in Homelessness. J Behav Addict. 2016;5(2):318-24

[xv] Sharman S, Dreyer J, Aitken M, Clark L, Bowden-Jones H. Rates of problematic gambling in a British homeless sample: a preliminary study. J Gambl Stud. 2015;31(2):525-32.

[xvi] Down and Out in London: Addictive Behaviors in Homelessness in web form https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5387783/

[xvii] Gambling-related harms evidence review: the economic and social cost of harms, p.27

[xviii] Problem gambling and suicidal thoughts, suicide attempts and non-suicidal self-harm in England: evidence from the Adult Psychiatric Morbidity Survey 2007, Heather Wardle, Simon Dymond, Ann John, Sally McManus Prepared for GambleAware May 2019

[xix] Ibid p.6

[xx] Karlsson A, Håkansson A. Gambling disorder, increased mortality, suicidality, and associated comorbidity: A longitudinal nationwide register study. J Behav Addict. 2018 Dec 1;7(4):1091-1099.

[xxi] Gambling-related harms evidence review: the economic and social cost of harms, p.31

[xxii] Afifi TO, Nicholson R, Martins SS, Sareen J. A Longitudinal Study of the Temporal Relation Between Problem Gambling and Mental and Substance Use Disorders Among Young Adults. Can J Psychiatry. 2016;61(2):102-111.

[xxiii] Ibid p.109

[xxiv] A longitudinal study of gambling in late adolescence and early adulthood: Follow-up assessment at 24 years, Alan Emond, Mark D. Griffiths, Linda Hollén, Report for GambleAware, December 2019

[xxv] Ibid p.28

[xxvi] Gambling-related harms evidence review: the economic and social cost of harms, p.35

[xxvii] Afifi TO, Nicholson R, Martins SS, Sareen J. A Longitudinal Study of the Temporal Relation Between Problem Gambling and Mental and Substance Use Disorders Among Young Adults. Can J Psychiatry. 2016;61(2):102-111.

[xxviii] Gambling-related harms evidence review: the economic and social cost of harms, p.36

[xxix] Ibid p.38

[xxx] A longitudinal study of gambling in late adolescence and early adulthood: Follow-up assessment at 24 years, Alan Emond, Mark D. Griffiths, Linda Hollén, Report for GambleAware, December 2019

[xxxi]Gambling-related harms evidence review: the economic and social cost of harms, p.42

[xxxii] Ibid p.43

[xxxiii] Thorley C, Stirling, A., Huynh, E. Cards on the Table: Cost to the government associated with people who are problem gamblers in Britain. Institute for Public Policy Research; 2016.

[xxxiv] Gambling-related harms evidence review: the economic and social cost of harms, p.45

[xxxv] Thorley C, Stirling, A., Huynh, E. Cards on the Table: Cost to the government associated with people who are problem gamblers in Britain. Institute for Public Policy Research; 2016. pp.37-38

[xxxvi] Thorley C, Stirling, A., Huynh, E. Cards on the Table: Cost to the government associated with people who are problem gamblers in Britain. Institute for Public Policy Research; 2016. p.49. Gambling-related harms evidence review: the economic and social cost of harms, p.47.

[xxxvii] Gambling-related harms evidence review: the economic and social cost of harms, p.48

[xxxviii] May-Chahal, C, Wilson, A, Humphreys, L & Anderson, J 2012, 'Promoting an Evidence-Informed Approach to Addressing Problem Gambling in UK Prison Populations', The Howard Journal of Criminal Justice, vol. 51, no. 4, pp. 372–386

[xxxix] https://www.england.nhs.uk/2021/11/record-high-patient-numbers-completing-nhs-treatment-for-common-mental-illness/#:~:text=The%20NHS%20Digital%20figures%20show,up%20on%20the%20previous%20year