For pity’s sake, it is now time to cut the crap

Fake Dictionary, Dictionary definition of the word Evidence.

For pity’s sake, it is now time to cut the crap

The loudest voices we hear are those who advocate conflict, divisiveness”
John C. Danforth, former US Ambassador to the United Nations

Have you ever noticed that the most vocal commentators are often the most ill-informed? It is something that Shakespeare observed when he wrote that “the empty vessel makes the loudest sound”; and so it appears to be the world over with the public policy debate on gambling.

Claptrap is much in vogue these days whenever the subject of gambling comes up in Britain’s Parliament or the press. Only last month, the House Of Lords was told by the Conservative peer Lord Chadlington that there are 430,000 people in this country with “a serious gambling problem”; that only 2% of them receive any treatment; that a further two million are in danger of addiction; that 55,000 11-to-14-year-olds are already addicted to gambling; and that two people commit suicide every working day in relation to gambling.

In his short speech (a contribution to the Lords debate on the NHS Long Term Plan), Lord Chadlington raised a number of important issues that all right-minded people (within and without the gambling industry) would agree on: a large number of people are likely to be experiencing harms in relation to gambling; provision of treatment services tends to be inadequate; effective prevention strategies are critical; we should be particularly concerned about the well-being of children; and some of the harms related to gambling can be extremely (life threateningly) severe.

Lord Chadlington’s assessment of the problems is spot-on – it’s just his facts that are faulty. The intention here is not to pick on the noble Lord, whose contribution to the debate on gambling – and in particular his probing questions to Government – has generally been intelligent and constructive. However, his recent comments are part of an increasingly common ‘alternative narrative’ that warrants attention.

Let us start with claims in relation to gambling addiction. The truth is that we simply don’t know how many people in Britain are clinically addicted to gambling. We can make a rough stab (using prevalence surveys) of the number likely to meet the diagnostic criteria for problem gambling – but this is not the same as addiction. The Problem Gambling Severity Index screen (which indicates the presence of two million ‘at-risk’ gamblers) and the DSM-IV derived screen (which indicates the 55,000 child problem gamblers) are simply not designed to diagnose addiction.

In 2015, Professor Mark Griffiths wrote this about his experience in co-authoring the 2010 British Gambling Prevalence Survey: “We reported that 0.9% of individuals in the sample were problem gamblers (i.e., they scored three or more on the DSM-IV criteria). What we did not say (and never have said) was that 0.9% of British adults (approximately 500,000 people) are “addicted” to gambling”. Professor Griffiths noted that while “all gambling addicts are problem gamblers… not all problem gamblers are gambling addicts”; and observed with dismay that “many stories in the British media when they talk about problem gambling will claim ‘half a million adults in Great Britain are gambling addicts’ (or words to that effect)” without any scientific basis for the claim.

Speaking at a conference earlier this year, Griffiths estimated that the true level of gambling addiction in Great Britain is likely to be around 0.1% of the population. While this is only a fraction of the problem gambling rate, it still represents a worryingly high number of people likely to be experiencing severe harm.

The equivalence of problem gambling with gambling addiction is not simply misleading; it may also have negative consequences if it serves to undermine a sense of autonomy amongst problem gamblers. For example if an individual believes that he is “addicted” it may affect his ability to regain control (by suggesting that he is pathologically unable to do so). It may also reinforce unhelpful stigmas, deter help-seeking and lead to poor choices in treatment.

The two-million ‘at-risk’ population is predominantly composed of people with very low scores on the PGSI. We estimate (based on ratios disclosed in the 2010 British Gambling Prevalence Survey) that around three-quarters of this group are likely to be classified as ‘low-risk’ gamblers. A low-risk diagnosis involves (according to the designers of the PGSI) “few or no identified negative consequences” from gambling; and evidence from other jurisdictions suggests that the vast majority of low-risk gamblers do not progress to a problem gambling diagnosis.

The identification of 55,000 11-to-14-year-olds is also incorrect. The figure (from the Gambling Commission’s ‘Young People and Gambling’ report) cites this figure in relation to 11-to-16-year-olds – not 11-to-14-year-olds (indeed the inclusion of 16-year-olds significantly skewed the rate in the most recent survey). While we are on the subject of gambling by children, oft-cited claims that problem gambling rates among 11-to-16-year-olds “quadrupled” between 2016 and 2018 are also wide of the mark. The Gambling Commission’s report states on at least four different occasions that the 2016 survey data are not comparable with the 2018 data due to significant differences in methodology (most notably the exclusion of 16-year-olds in the surveys prior to 2017).

Then there is the assertion that 430,000 people have a “serious gambling problem”. The central estimate from the most recent household survey (the 2016 Health Survey) cites a figure of 340,000 problem gamblers. The 430,000 figure comes from the 2015 Health Survey – the dramatic apparent reduction between surveys illustrates the sensitivities involved in such research projects. It may be appropriate to take the mean of the two surveys (385,000) but simply choosing the larger of the two figures on offer (regardless of recency) is dishonest. There is also something unsettling in the idea that 340,000 is somehow not considered a large enough number of people and must therefore be inflated.

The 430,000 and 340,000 figures are estimates of those diagnosed with a gambling problem. There is no clinical diagnosis for “a serious gambling problem”. We might consider that the DSM-IV classification of ‘pathological gambling’ is a proxy for a “serious gambling problem” – but if so then we must also accept that the number in this category is a little under half the number for problem gambling. We may legitimately conclude that all problem gambling is serious but we ought not to invent clinical classifications.

Then there is the claim that only 2% of these problem gamblers receive any treatment. Again, the only thing we really know about the scale of treatment-seeking is that we don’t really know very much at all. The 2% figure is based upon the assumption that only counselling services funded by GambleAware may be considered to constitute treatment. It therefore excludes the valuable work of Gamblers Anonymous (among others) as well as support provided by the national gambling helpline and netline.

We should also question the implicit suggestion that everyone with a qualifying score under the PGSI or DSM-IV requires direct treatment services of the kind funded by GambleAware. Based upon clinical evidence, this seems unlikely to be the case.

Finally there is the very sensitive question of suicide. It is a matter of some concern that we have no reliable data on gambling-related suicide data in Great Britain (and many other jurisdictions). However, the idea that we might derive an estimate from a decade-old study of 17 deaths in Hong Kong appears questionable. We would never dream of estimating the level of AIDS infection by transposing ratios from Botswana or guessing at the prevalence of respiratory diseases by reference to Mexico City – so why do people think it is acceptable to calculate the number of gambling suicides using a small scale study from Hong Kong? At the time of the study, the problem gambling rate in Hong Kong was 5.5% (around seven or eight times the level in Great Britain); and around one-third of the victims in the study had received menaces from loan sharks in the days leading up to their deaths. Are we really suggesting this is a valid comparison?

Perhaps though all of this is mere pedantry. If we can accept the general observation that too many people are experiencing harm with too little done to address it, does it matter whether the facts and statistics are correct? Maybe it is ok to exaggerate if it draws attention to a problem that would otherwise be neglected. This is an entirely defensible position for gambling concern or lobby groups; but not for those in positions of regulatory, political or moral authority. We rely on these people to be unimpeachable.

One obvious negative consequence of mis-stating the scope, depth and nature of gambling harms is that it is likely to lead to inefficient or ineffective solutions. For example, if we genuinely believe that there are two million people in danger of addiction, we may decide to devote substantial resources to this group (for example through costly population-wide programmes) – and this may in turn lead to underfunding of front-line treatment services for those in serious difficulties.

If we make things up, it becomes very hard to work out whether things are getting better or worse. If for example, we discover (through robust coroner reporting) that there are say 325 gambling-related suicides a year in Britain, could we really conclude that the rate had halved by comparing the figure with Hong Kong derived estimates? Indeed, it is a curious quirk of using the Hong Kong methodology that it implies a significant decline in gambling-related suicides over the last decade (which is not perhaps the point its proponents intend).

Finally, the spreading of misinformation creates distrust, frustrates collaboration and gets in the way of effective problem solving.

So what is to be done about this? To be fair to the British Government, the Department for Culture, Media and Sport is generally very sober and scientific when responding to questions on gambling harm; but understandably may not be inclined to expend too much energy on correcting misperceptions in the press and Parliament. Similarly, the Gambling Commission may consider that it is either not its job or that matters are too highly charged to wade in; although the regulator has in the past criticised the industry for downplaying issues of harm so there is a question of even-handedness.

It is too easy for operators to bemoan their fate; but in order to change the narrative, they must dump their own spin and face up to issues of harm in a far more honest way than in the past. As the excellent Matt Zarb-Cousin of the Campaign for Fairer Gambling recently observed, “there is a long way to go in ensuring representations to government by corporations, vested interests [of whatever stamp] or their advocates are entirely truthful”.

In recent years, there has been a perceptible shift in this regard – at least amongst more progressive companies. Regulators, ministers and politicians have the opportunity to encourage greater candour and maturity from licensees if they themselves are prepared to take responsibility for how facts are presented and the policy debate conducted.

Across a large number of jurisdictions, operators are increasingly finding themselves on the wrong side of the discussion about what sorts of gambling ought to be permitted and under what circumstances. Indeed, in certain areas licensees are finding themselves deliberately excluded from these debates.

Blinkered approaches to issues of consumer harm have damaged trust in the industry and allowed critics to put forward ever more strident (and perhaps moralistic) views through the distortion of ‘facts’. For licensees, getting a grip on the evidence – and presenting balanced assessments rather than cherry-picking – is critical to restoring integrity to policy discourse on both gambling and gambling harm.