Last week, GambleAware grabbed media headlines, publishing its annual YouGov treatment support survey that contradicted Gambling Commission % figures on disorder rates. Critical of GambleAware’s self-selected research, Regulus Partners warns that the charity has deflected from recognising real progress on treating and detecting gambling disorders.
Even by the standards of recent times, this was an utterly hopeless week for public policy discourse on disordered gambling. On Thursday, GambleAware released the result of its latest YouGov report on treatment needs in Great Britain. As in previous years, the charity led with the claim that national prevalence rates of ‘problem’ and ‘at risk’ gambling in Great Britain may be massively under-stated.
According to YouGov, 2.8% of the adult population in 2021 were estimated to be PGSI ‘problem gamblers’ – compared with 0.4% in the 2018 Health Survey for England and just 0.3% from the 2021 Gambling Commission’s 2021 Quarterly Telephone Survey. A further 9.9% were claimed to be ‘low risk’ or ‘moderate risk’ gamblers (compared with 3.5% in the HSE 2018 and 2.7% in the QTS).
The report rather predictably added fuel to the raging fire of media commentary on gambling but very few serious commentators (including those aligned with the anti-gambling lobby) are likely to take the results seriously. This is because YouGov uses non-probability, self-selection online panels which are generally considered inappropriate for estimating population prevalence. Pickering & Blaszczynski (2021) for example, observe that “the quality of data may be questionable due to careless responding and identity misrepresentation which can bias study results”; while Pierce et al. (2020), discuss results from convenience surveys on the effects of Covid policies warn that “the desire for quick information has driven the rapid propagation of online surveys using non-probability and convenience samples, some of which claim to be representative.
Understandably, many are receiving widespread media attention.” They add that, “these early insights might be valuable, but we caution against relying on them to drive policy and resource because they are prone to substantial bias: acting on misleading information could be worse than having no information at all.”
The Gambling Commission has also advised against the use of prevalence estimates obtained in the way used by GambleAware, writing: “There are a number of limitations with these types of non-probability online samples, including low response rates and bias, often a result of noncoverage of those without internet access and self-selection bias. There are also concerns around fraudulent and inattentive behaviour by panellists.”
Last year, Professors Patrick Sturgis and Jouni Kuha of the London School of Economics were asked by GambleAware to examine why the YouGov results differed so markedly from those obtained from NHS Health Surveys. They determined – by process of elimination – that selection bias from the use of online, non-probability panels was likely to be “a major source of error”; and that selection bias was “a particular risk for the online surveys because they either have low response rates, or use non-probability sampling”. The “core finding” from the LSE review was that “online surveys seem to substantially over-estimate gambling activity and gambling harm”. At the time, GambleAware appeared to concur, stating in its press release that “surveys conducted wholly or primarily online tend to overestimate the prevalence of gambling harms.”
Inspection of the wider GambleAware/YouGov survey supports concerns of bias. For example, the YouGov sample appeared to have identified substantially higher rates of hazardous drinking than are reported in Health Surveys; while a staggeringly large proportion (40%) of respondents were estimated to have a mental health disorder. Interestingly, YouGov’s results showed that psychiatric distress had a markedly lower prevalence among recreational gamblers than non-gamblers. Even low-risk gamblers appeared to enjoy better mental health than the non-gamblers – but these findings went curiously unremarked.
The survey also indicated that around 200,000 people in Britain had used a “specialist treatment service for gambling” in the prior year. Given that the National Gambling Treatment Service provided direct counselling for around 9,000 people and perhaps another 20,000 received support from the National Gambling Helpline, a question must be asked about where else people are assumed to be receiving this “specialist treatment”…if of course we take the results to be accurate.
Aside from the confusion occasioned by the GambleAware report, the episode highlights the hopelessness of discussions on gambling disorder and its treatment. GambleAware’s claim that only one-in-160 PGSI ‘problem gamblers’ are being helped by the National Gambling Treatment Service reveals an obsessively pessimistic outlook. In GambleAware’s view, even a ten-fold increase in treatment would still leave 93% of ‘problem gamblers’ unaided – a truly hopeless prognosis. In truth, however, things are a little more optimistic than depicted. As the eminent researcher, Dr Robert Williams of the University of Lethbridge points out in his 2018 work, ‘Gambling Fallacies of Gambling Researchers’, not all ‘problem gamblers’ require treatment (and the vast majority self-correct within a relatively short space of time).
The psychiatrist’s bible, the DSM-5 states that it is typically only those with moderate or severe gambling disorder who present for treatment. It may just be therefore that the population for whom face-to-face counselling (or even pharmacotherapy) is relevant is of the order of 50,000 people rather than 300,000 or 1.4 million. If true, this would suggest two things: 1) that greater resources are still required to make treatment capacity available and encourage those who need help to seek help; and 2) that a meaningful proportion of the treatment population is already receiving effective support (while those with less severe problems are accessing help in other ways). It is a far more hopeful (and likely more accurate) way of looking at the world for those involved in tackling gambling harm as well as for the disordered gamblers themselves. The message that help-seeking is relatively common (one-in-five) is more likely in our view to encourage treatment-seeking than the suggestion that it is non-normative (one-on-160).
Sadly, it is now de rigour for gambling disorder to be discussed in the bleakest of terms (and often discussed most volubly by those who know the least; able therefore to insulate simplistic moral views from the inconvenience of complexity). What is curious is that those who claim to care most about reducing harm from gambling are often the most likely to refute progress; to paint the picture in its darkest hues – and in so doing blot out the light of hope central to progress in any endeavour. There can surely be no doubt that British gambling consumers are better protected these days compared with five or ten years ago – and evidence from Health Surveys certainly supports the view that disorder is both a relatively rare condition and one that is likely declining in prevalence.
This is not to deny that the effects of disordered gambling can be devastating to individuals and those close to them – but it ought to be possible to recognise that the individuated harms of dysregulated gambling are severe (and that these harms may be occasioned by the negligence of operators) and yet at the same time to observe that they are in fact relatively rare. In any case, whether there are 50,000, 300,000 or 1.4 million people in need of specialist treatment does not in fact change priorities when only 9,000 are currently receiving it.
In the same way, the fact that gambling disorder can involve serious harms does not mean that discourse needs to be exclusively gloomy. Indeed, it may just be that positive-thinking and collaborative problem-solving might achieve far more than the current diet of sanctimony, hand-wringing and expedient mendacity that passes for public policy debate in Great Britain.
As the late Dr Hans Rosling noted in his best-selling work, ‘Factfulness’, “a fact-based world view is more comfortable. It creates less stress and hopelessness than the dramatic worldview simply because the dramatic one is so negative and terrifying.”