The Relationship between the Availability of Commercial Gambling Opportunities and Problem Gambling
The research evidence suggests that most people who gamble do so in such a way as to enhance their enjoyment of life, and without causing themselves significant harm. A minority of the population gambles in a manner that causes substantial harm to themselves and their families, and such individuals find it unusually difficult to stop or to control their gambling. In a minority of these cases, the harm caused by severe pathological gambling is as devastating as that caused by alcoholism and other addictions.
It is a matter of judgment to decide to what extent restricting freedom of choice for the majority of people who gamble harmlessly is justified in the hope of protecting a minority who might otherwise gamble excessively. However, prohibition and restriction of opportunities to gamble clearly do not eliminate the incidence of problem gambling. It is not known whether or to what extent, if any, they reduce it.
It is plausible to think that some people who have a psychological, genetic, or neurobiological propensity to become problem gamblers are more likely to be elicited by certain kinds of legal and regulatory frameworks than others. Evidence and casual observation suggest that the risk of developing a gambling problem increase if gambling opportunities:
- are continuous
• offer frequent prizes
• offer what are perceived to be high prizes
• allow large sums to be staked
• allow credit to be used
• are located in venues where people are likely to gamble on impulse
• are introduced without an accompanying public education campaign which makes people aware of the dangers of gambling and how to avoid them.
No form of gambling is risk-free, but one can deduce that casino games are riskier than betting on sporting and other events, and betting on events is riskier than buying weekly lottery tickets. Casino games are safer if they are offered at a single venue where people must make a conscious decision in advance to visit rather than simply encountering an opportunity to gamble while undertaking other activities, i.e. gaming machines in a bar. They are less safe in venues that are close to where people live and work. They are less safe in venues—such as supermarkets and bars—where people go for other purposes and may be more readily tempted to gamble on impulse.
The introduction or expansion of casinos does not necessarily lead to an increase in problem gambling. A study by Volberg on the introduction of casinos and gaming machines in Montana, North Dakota, Oregon and Washington State compared problem gambling rates before and after the introduction of casinos or gaming machines. (Volberg, R. A. 2001. Gambling and Problem Gambling in North Dakota: A Replication Study, 1992 to 2000. Report to the North Dakota Office of the Governor. Bismarck, ND: Office of the Governor) She found that in Montana and North Dakota the incidence of problem and pathological gambling—as measured by the South Oaks Gambling Screen—increased substantially. In Montana, which had the largest increase, problem gambling grew from 2.2% of the adult population to 3.2%, and pathological gambling from 0,7% to 1.6%. However, in Oregon, numbers for problem and pathological gamblers declined from 3.3% to 2.3% for problem gamblers and from 1.4% to 0.9% for pathological gamblers. The critical variable, according to Volberg, was whether the introduction of casinos was accompanied by the provision of services for problem gamblers, including programs to enhance public awareness about gambling and its dangers.
Volberg’s findings—that the introduction of casinos or gaming machines does not necessarily lead to an increase in problem gambling—is replicated in the study which she undertook with Abbott into the incidence of problem gambling in New Zealand. They found that, before and after the introduction of casinos—where problem gambling services were extensive— problem gambling numbers decreased. (Abbott, M. W., & Volberg, R. A. (1996), “The New Zealand National Survey of problem and pathological gambling,” Journal of Gambling Studies, 12, 143–160.) In summary, the evidence linking expanded gambling opportunities to problem gambling is scientifically inconclusive, though there is some ad hoc evidence (i.e. number of Gamblers Anonymous meetings, hotline calls, etc.) suggesting that increases in availability leads to increases in the level of problem gambling. This is clearly an area where greater scientific research is needed to clarify the underlying cause-effect relationships.
Availability of Services for Problem Gamblers
Researchers for this report received little information about the availability of problem gambling services as a result of our inquiries either from government or from service providers. The information that follows has been extracted from published sources including chapter three of a report prepared for the Gambling Industry Charitable Trust in the United Kingdom. (Collins, P. et al (2003), Towards a Strategy for Addressing Problem Gambling in the UK: A Report to the Gambling Industry Charitable Trust, The Responsibility in Gambling Trust)
There are four main categories of service or program which jurisdictions tend to put into place on either a statutory or a voluntary basis in order to mitigate harm caused by problem gambling: treatment; public awareness and prevention; training; and research.
It is only in the UK’s Gaming Act 2005 that provision is made to ensure that all these activities are developed and funded through the establishment of an industry body—the Responsibility in Gambling Trust— (RIGT, formerly the Gambling Industry Charitable Trust) and through the requirement that licensees demonstrate social responsibility.
The two other European jurisdictions which have significant responsible gambling programs authorized by the government are Holland and Sweden where the casinos industry as well as the lottery are owned by the state. Otherwise, from our findings, EU jurisdictions treat problem gambling in the context of the mental health services as are provided. Also, some jurisdictions have self-help groups for compulsive gamblers based on the program of Gamblers Anonymous or Alcoholics Anonymous.
In various Member States, there are established services available for problem gamblers. For example, in the UK, The Netherlands, France and Sweden, there are help-lines available for problem gamblers, as well as dedicated out-patient treatment services. In the UK, these are mainly offered through the service provider GamCare, which also operates the helpline. In The Netherlands, they are offered through the Jellinek Foundation which specializes in all addictions. In France, the helping service SOS Joueurs offers a variety of services over the internet and other information for problem gamblers. In Sweden, such services are offered through local organizations financed by local communities. Furthermore, research and education programs are offered in Sweden through the National Institute of Health and the Spel Institutet. Gordon House in the UK, as far as we were able to determine, is the only charitable organization which offers in-patient treatment for compulsive gamblers in the EU. Increasingly, help for problem gamblers in all these jurisdictions is being made available on the internet.
The following information is offered to the extent that it may be of interest:
- Dutch mental health services in total treated 3,941 problem gamblers in 2001 compared with 25,510 people with alcohol problems and 36,658 people with drug problems
• GamCare’s helpline recorded 14,915 calls in 2004, of whom by far the largest number learnt of the service through the Yellow Pages
• GamCare treated 239 new clients on a face-to-face basis in 2004.
Methodological Problems in Addressing Problem Gambling
The following is offered as an example of the difficulties inherent with non-scientific considerations that affect the quality of published research in the area of problem gambling. One can argue that the process of peer-reviewed refereeing reduces the incidence of compromise in the research process.
The report produced by NORC (National Opinion Research Center) was commissioned by, and submitted to, the (U.S.) National Gambling Impact Study Commission (NGISC) in April 1999. (Gerstein, D.R., Volberg, R.A., Toce, M.T., Harwood, H., Johnson, R.A., Buie, T., Christiansen, E. et al. (1999), Gambling impact and behavior study: Report to the National Gambling Impact Study Commission. Chicago: National Opinion Research Center, the University of Chicago.) The major purpose of this study was to conduct a comprehensive national survey to update and expand upon a prior survey conducted in 1975 (published in 1976) by the University of Michigan Survey Research Center, for the Commission on the Review of the National Policy toward Gambling. (Kallick, M., Suits, D., Dielman, T. & Hybels, J. (1976), Survey of American Gambling Attitudes and Behavior: Final Report to the Commission on the Review of the National Policy Toward Gambling, Ann Arbor, Michigan: University of Michigan Press.)
To achieve the goals of the NGISC, the sample size for the study was based on a number of factors: number of households with adult females and adult males, population of the two strata – lottery and non-lottery states, expected number of pathological and problem gamblers among males and females, distance to major gambling facilities, and the expected number of completed interviews (NORC 1999, Appendix B). It is important to note that one of the criteria used by NORC was the expected number of pathological and problem gamblers.
It should also be noted that the NORC report states that: “…as the data collection progressed, we determined that we would achieve our sampling objectives without the safety margin; thus we never released these cases” (NORC 1999, Appendix B). The NORC sample started out with 11,500 RDD (random digit dial numbers) of households stratified by lottery and non-lottery states. Of the initial 11,500 RDD households, 80%—or 9,200 numbers—were initially assigned for data collection, holding an additional 20%—or 2,300—in reserve to provide a safety margin. The completion rate of 2,417 for the 4,358 working residential numbers was 56%. The assumption was that there was no bias in the large group of non-respondents (respondents), an assumption which could not be tested.
This is a problem common to all primary data surveys and not just to the NORC survey. As was pointed out in the NORC report, the sample size for the problem and pathological groups (30 problem gambler and 21 pathological gambler respondents) is too small for generalizable analysis. This result seems somewhat unexpected (even though NORC stated that it was expected) since the sample was designed using—as one of the criteria—the expected number of problem and pathological gamblers. It should also be pointed out that even though the sample size was small it was, by design, unbiased (under certain qualifying assumptions concerning the 44% non-respondents).
Under pressure from the National Gambling Impact Study Commission, NORC increased the sample size of problem and pathological gamblers, and supplemented their sample with an intercept sample of gambling facility patrons, stratified by lottery and non-lottery states and at various gaming facilities in the United States. But this was not a random sample (equally likely to be selected) of households, such as the RDD sample. Intercept samples are not truly random and do not have well-defined statistical properties; it is therefore not possible to draw valid general conclusions for the population from their results. In addition, the interviews were conducted face-to-face, unlike the relatively more impersonal RDD telephone interviews.
Nonetheless, NORC combined the intercept sample, conducted to “find” more problem and pathological gamblers, with the RDD sample and used the combined sample for analysis. This procedure is fatally flawed since the patron survey was not a truly random sample as was the RDD sample. All of the care taken to assure an unbiased random sample using the RDD method was negated when the two samples were combined. Thus, it was not possible to analyze the respondents using the combined using statistically methods. Unfortunately, even though they acknowledged the problems with the merged sample, the NORC team still proceeded to analyze the results of the survey with respect to pathological and problem gambling. However, the results of their analysis with respect to problem and pathological gambling are not valid in terms of their ability to generalize to the target population, and therefore should not be used in research or policy analyses.
© European Union