2012-Prevention of problem gambling.pdf

1

2012

PREVENTION OF PROBLEM GAMBLING: A Comprehensive Review of the Evidence and Identified Best Practices

Robert J. Williams Professor, Faculty of Health Sciences, and Coordinator, Alberta Gambling Research Institute University of Lethbridge, Lethbridge, Alberta, Canada

Beverly L. West Research Associate, Faculty of Health Sciences University of Lethbridge, Lethbridge, Alberta, Canada

Robert I. Simpson Past CEO, Ontario Problem Gambling Research Centre Guelph, Ontario, Canada

REPORT PREPARED FOR THE ONTARIO PROBLEM GAMBLING RESEARCH CENTRE AND THE ONTARIO MINISTRY OF HEALTH AND LONG TERM CARE

October 1 2012

2 History of this document Short version appears in a book chapter: Aug 2007

Dec 2007

Nov 2008

Dec 2008

Current document

Williams, R.J., West, B.L., & Simpson, R.I. (2007a). Prevention of problem gambling. In G. Smith, D. Hodgins, and R.J. Williams (Eds.), Research and Measurement Issues in Gambling Studies. pp. 399-435. San Diego, CA: Elsevier. http://hdl.handle.net/10133/414 Long version prepared as a report for the Ontario Problem Gambling Research Centre: Williams, R.J., West, B.L., & Simpson, R.I. (2007b). Prevention of Problem Gambling: A Comprehensive Review of the Evidence. Report prepared for the Ontario Problem Gambling Research Centre. Guelph, Ontario, Canada. Dec 1, 2007. http://hdl.handle.net/10133/414 Separate report prepared on Promising Practices in Prevention: Williams, R.J., & Simpson, R.I. (2008). Promising Practices in the Prevention of Problem Gambling. Final Report prepared for the Ontario Problem Gambling Research Centre and the Ontario Ministry of Health and Long Term Care. November 14, 2008. http://hdl.handle.net/10133/751 Minor literature update to the Dec 2007 report: Williams, R.J., West, B.L., & Simpson, R.I. (2008). Prevention of Problem/Pathological Gambling: A Comprehensive Review of the Evidence. Report prepared for the Ontario Problem Gambling Research Centre. Guelph, Ontario, Canada. December 1, 2008. http://hdl.handle.net/10133/414 Substantive literature update and integration with the Promising Practices in the Prevention of Problem Gambling document.

Citation Williams, R.J., West, B.L., & Simpson, R.I. (2012). Prevention of Problem Gambling: A Comprehensive Review of the Evidence, and Identified Best Practices. Report prepared for the Ontario Problem Gambling Research Centre and the Ontario Ministry of Health and Long Term Care. October 1, 2012. http://hdl.handle.net/10133/3121

Acknowledgements There are many external reviewers who have provided helpful feedback on the current and earlier versions of this document. However, in particular, the authors acknowledge the very useful feedback provided by Dr. Tim Stockwell (Director, Centre for Addictions Research of British Columbia, University of Victoria).

Contact Information Dr. Robert Williams Professor, Faculty of Health Sciences & Coordinator, Alberta Gambling Research Institute University of Lethbridge Lethbridge, Alberta, Canada; T1K 3M4 robert.williams@uleth.ca

3 INTRODUCTION………………………………………………………………………………………………..…………………….……

5

ETIOLOGICAL FRAMEWORK FOR PROBLEM GAMBLING………………………………….……………………….…..

7

Indirect Biological Risk Factors…………………………………………………………………………..… Indirect Environmental Risk Factors…………………………………………………………………….. Direct Risk Factors……………………………………………………………………………………………….. Problem and Pathological Gambling……………………………………………………………………..

7 9 9 10

METHODOLOGICAL APPROACH……………………………………………………………………………………………………

13

EDUCATIONAL INITIATIVES……………………………………………..……………………………………………………………

15

Childhood Interventions ……………………..………..…………………………………………………,…. Information/Awareness Campaigns………………………………………................................ Responsible Gambling Information Centres…………………………………………………………. Statistical Instruction…………………….……………………………………………………………………… School-Based Prevention Programs……………………………………....................................

15 15 20 21 22

POLICY INITIATIVES………………………………………………………….…………………………………………………………..

25

RESTRICTIONS ON THE GENERAL AVAILABILITY OF GAMBLING……………………………………….. Restricting the Number of Gambling Venues……………………………………………………….. Restricting More Harmful Types of Gambling….…………………………………………………… Continuous Forms of Gambling………….…………………………………………………….. Internet Gambling……………………………….…………………………………………………… Restricting the Number of Gambling Formats……………..….…………………………………… Restricting Gambling to Dedicated Gambling Venues ………….................................. Restricting the Location of Gambling Venues……………….......................................... Limiting Gambling Venue Hours of Operation………………….…..………………………………

25 26 29 29 35 37 37 40 41

RESTRICTIONS ON WHO CAN GAMBLE………………………………………….………………………………… Prohibition of Youth Gambling…………………………………………………………………………….. Increasing the Legal Age for Gambling …………………………………………………… Restricting Venue Entry to Non-Residents……………………………………………………………. Restricting Venue Entry to Higher Socioeconomic Classes….................................... Casino Self-Exclusion…………………………………………………………………………………………….

43 43 44 45 46 46

RESTRICTIONS OR ALTERATIONS ON HOW GAMBLING IS PROVIDED.................................. Modifying Electronic Gambling Machine (EGM) Parameters………………………………… Win/Loss Patterns…………………………………………………………………………………….. Game Play Speed……………………………………………………………………………………… Near Misses…………………………………………………………….……………………………….. Number of Play Lines………………………………………………………………………………… Bill Acceptors………………………………………………………….…………………………………

52 52 52 53 53 54 54

4

Stake and Prize Size…………………………………………………………………………………. Payback Percentage……………………………………………….………………………………… Interactive Features………………………………………………………………………………….. Pop-Up Messages…………………………………………………………………………………….. On-Screen Clocks……………………………………………………………………………………… Mandatory Cash-Out………………………………………………………………………………… Privacy……………………………………………………………………………………………………… Money versus Credits……………………………………………………………………………….. Lights and Sounds…………………………………………………………………………………….. Seating……………………………………………………………………………………………………… Summary of EGM Parameter Modifications………………………………………………….. Player Pre-Commitment………………………………………………………………………………………. Eliminating Reward/Loyalty Cards or Changing Their Parameters.……………………….. Operator Imposed Maximum Loss Limits……………………………………………………………… Proactive Intervention with ‘At-Risk’ Gamblers……………………………………………………. Problem Gambling Awareness Training to Employees………………………………. Automated or Mandated Intervention……………………………………………………… Restricting Access to Money………………………………………………….……………………………. Restrictions on Concurrent use of Alcohol and Tobacco……..................................... Restricting Advertising………………………………………..…………........................................ Gambling Venue Design ………………………………………………………………………………………. Increasing the Cost of Gambling…………………………………………………………………………… Government Provision of Gambling..……………………..…………………………………………….

55 56 57 58 59 59 59 60 60 60 61 61 65 66 67 67 68 70 72 74 76 77 78

SUMMARY OF THE EFFECTIVENESS OF EDUCATIONAL AND POLICY INITIATIVES………………………….

81

BEST PRACTICES TO PREVENT PROBLEM GAMBLING…………………………………………………………………… 1. Strive for Optimal Design and Evaluation of New Initiatives………………………….…………… 2. Recognize that Effective Prevention requires Decreased Revenue and some Inconvenience to Non-Problem Gamblers……………………………………………………………….… 3. Employ a Wide Array of Educational and Policy Initiatives…………………………………………. 4. Coordinate these Multiple Educational and Policy Initiatives..…………………………………… 5. Decrease the General Availability of Gambling………………………………………………………….. 6. Eliminate, Reduce, and/or Constrain Higher-Risk Forms of Gambling………………………… 7. Eliminate Reward Cards or use them to Foster Responsible Gambling………………………. 8. Restrict who is Eligible to Gamble……………………………………………………………………………… 9. Restrict the use of Tobacco and Alcohol while Gambling…………………………………………… 10. Restrict Access to Money while Gambling…………………………………………………………………. 11. Impart Knowledge, Attitudes, and Skills to Gamblers ..……………………………………………… 12. Keep Prevention Initiatives in Place for a Sustained Period..………………………………………

84 84 86 87 89 90 92 93 94 95 96 97 99

REFERENCES………………………………………………………………………………………………………………………………… 100

5

INTRODUCTION The past 30 years has seen a dramatic worldwide expansion of legalized gambling. The overall social and economic costs and benefits of this expansion are mixed (Williams, Rehm, & Stevens, 2011). However, what is clear is that one of the primary negative impacts of widely available gambling opportunities is the development of disordered gambling in a minority of people. Various terms have been use to describe disordered gambling, including ‘compulsive gambling’, ‘addictive gambling’, ‘problem gambling’, and ‘pathological gambling’. The term used in the present document is ‘problem gambling’. The definition of problem gambling put forward by Neal, Delfabbro, & O’Neil (2005) captures the essential elements common to almost all definitions: “Problem Gambling is characterized by difficulties in limiting money and/or time spent on gambling which leads to adverse consequences for the gambler, others, or for the community.” Essentially, a problem gambler is someone with a pattern of excessive gambling; impaired control over their gambling behaviour; significant negative consequences deriving from this impaired control; and persistence in excessive gambling despite these negative consequences. Problem gambling is assumed to have varying degrees of severity, ranging from mild, moderate to severe.

The term ‘pathological gambling’ is synonymous with severe problem gambling.1 Depending on the year and the jurisdiction, the past year prevalence of problem gambling ranges from 0.5% to 7.6% of the adult population (Williams, Volberg, & Stevens, 2012). Problem gambling is associated with a range of negative consequences for the individual, his/her family, and for society in general. Financial difficulties tend to be the most common type of problem. A significant percentage of problem gamblers will eventually file for bankruptcy (Petry, 2005; Williams, Rehm, & Stevens, 2011). Mental health problems in the form of guilt, depression and anxiety are also common. Moreover, problem gamblers have a significantly elevated risk for suicide attempts and suicide (National Research Council, 1999; Williams, Rehm & Stevens, 2011) (Las Vegas has had the highest per capita rate of suicide in North America for many years). A small percentage of problem gamblers will develop stress-related physical health problems either in addition to or instead of mental health problems (i.e., high blood pressure, gastric disorders). Some problem gamblers experience difficulties at work or school because of their gambling. Poorer grades or reduced work productivity is not uncommon, and school failure or job loss sometimes occurs (National Research Council, 1999; Petry, 2005).

Criminal activity to finance gambling (typically fraud or embezzlement) occurs in a minority of problem gamblers (National Research Council, 1999; Williams, Rehm & Stevens, 2011). Upward of 33% of prison inmates have histories of problem gambling (Williams, Royston, & Hagen, 2005). Problem gamblers typically have conflicts with family and/or friends over their gambling. These conflicts often lead to relationship problems (with spouse, children, and friends), sometimes lead to divorce, sometimes lead to domestic violence, and sometimes lead to child abuse and neglect. These problems, in 1

‘Pathological gambling’ is a less common term that is used mostly in the United States and among medical professionals (primarily due to the fact that ‘pathological gambling’ is the term used in the Diagnostic and Statistical Manual of Mental Disorders (DSM) produced by the American Psychiatric Association). ‘Problem gambling’ is the preferred term because it has fewer etiological connotations (i.e., ‘pathological’ means ‘disease-like’) and because pathological gambling in DSM is dichotomous, whereas evidence indicates that disordered gambling lies on a continuum. Most people now use the term ‘pathological gambling’ to denote ‘severe problem gambling’.

6

turn, may produce depression, anxiety, and poorer mental health in the spouse and children. There is also an inter-generational impact, as children who have a parent who is a problem gambler are at elevated risk for developing problem gambling themselves (Kalischuk et al., 2006; Shaw et al., 2007). The past 15 years has seen a considerable amount of interest and effort being put into developing strategies to prevent problem gambling. Unfortunately, the development, implementation, and evaluation of most of these initiatives have been a haphazard process. Most have been put in place because they ‘seemed like good ideas’ and/or were being used in other jurisdictions, rather than having demonstrated scientific efficacy or being derived from a good understanding of effective practices in prevention. The primary purpose of the present document is to help change this state of affairs. More specifically, by:  



Proposing an etiological framework for understanding how problem gambling develops based on the available evidence and drawing from established models of addictive behaviour. Comprehensively evaluating the effectiveness of the various initiatives that have been used around the world to prevent problem gambling based on their demonstrated efficacy and/or their similarity to initiatives that are empirically effective in preventing other addictive behaviour. Based on this etiological framework and this critical review of the research, identifying current ‘best practices’ for the prevention of problem gambling.

7

ETIOLOGICAL FRAMEWORK FOR PROBLEM GAMBLING In order to effectively prevent something, it is necessary to understand what causes it. The present section describes an etiological framework for understanding the causes of problem and pathological gambling based on the best available evidence. This framework is still somewhat tentative, as the exact etiology of problem gambling hinges on having longitudinal research that clearly identifies which variables are etiologically involved, their temporal sequence, and their causal connections. While several longitudinal studies of gambling exist (Slutske, 2007), all have limitations either because they do not include all the relevant variables, have small sample sizes and/or insufficient numbers of ‘at risk’ gamblers, too circumscribed an age range, short time spans, a limited number of collection periods, or poor retention rates. However, there are some major longitudinal studies of gambling with improved methodological design nearing completion Canada, Sweden, and Australia that will provide much stronger empirical evidence about the etiology of problem gambling.

That being said, sufficient evidence is available to propose a tentative etiological framework based on the results of the longitudinal studies that do exist, the large amount of cross-sectional research that has identified correlates of problem gambling, twin studies, the retrospective reports of problem gamblers, and the etiological models that have been developed for addictive behaviour more generally (Griffiths, 2005a; Griffiths & Delfabbro, 2001; Kumpfer, Trunnell, & Whiteside, 1990; Marlatt et al., 1988; Shaffer et al., 2004; Sharpe, 2002; Wallace, 1993). Further to this latter point, the biopsychosocial approach is central to all etiological models of addiction, including the present framework. Essentially, this approach posits that there are a large number of biological, psychological, experiential, and social factors which interact in complex ways to both contribute to and protect individuals from developing addictive behaviour. In keeping with this biopsychosocial orientation, and incorporating the best current evidence about the likely etiology of problem gambling, the following is a description of the risk factors thought to contribute to problem gambling, and their plausible temporal sequence.

Indirect Biological Risk Factors Indirect risk factors are those that create vulnerabilities and tendencies which predispose people to gambling problems when later combined with more ‘direct’ factors. Genetic inheritance creates biological propensities that increase or decrease a person’s likelihood of engaging in gambling and/or developing problem gambling. Twin studies indicate that 40-50% of the propensity for developing or not developing problem gambling (heritability) can be predicted by genetic factors (Eisen et al., 1998; Lobo & Kennedy, 2006, 2009; Shah et al., 2005; Slutske et al., 2010). Although this figure may seem high to some people, it is very consistent with the heritability estimates of substance dependence, which ranges from 30% to 70% depending on

8

the substance (Agrawal & Lynskey, 2008; Goldman et al., 2005) as well as the heritability estimates of the major psychiatric disorders, which range from 30 – 85% (Shih et al., 2004). Although there may be some gambling-specific genetic vulnerability, what people are often inheriting is a shared genetic vulnerability for several conditions, which explains the high degree of problem gambling comorbidity with a) substance use and abuse (Crockford & el-Guebaly, 1998; el-Guebaly et al., 2006; Petry, 2007; Petry, Stinson, & Grant, 2005; Vitaro et al., 2001; Zimmerman, Chelminski, & Young, 2006); b) delinquency and/or antisocial personality (Crockford & el-Guebaly, 1998; Petry, Stinson, & Grant, 2005; Vitaro et al., 2001); and c) mood disorders (Crockford & elGuebaly, 1998; el-Guebaly et al., 2006; McCormick et al., 1984; Mood Disorders Society of Canada, 2004; Suck Won et al., 2006; Toneatto & Nguyen, 2007; Turner et al., 2008; Zimmerman, Chelminski, & Young, 2006) (see Black et al., 2006; Comings, 2006; Goodman, 2008; Goudriaan et al., 2004; Grant et al., 2006; Ibanez et al., 2002; Lobo et al., 2010; Potenza et al., 2005; Slutske et al., 2000; Slutske et al., 2010).

This genetic inheritance appears to express itself neurologically by means of differential functioning of the ventral striatum/mesolimbic pathway (“reward” pathway) (Buchel, 2006; Goodman, 2008; Goudriaan et al., 2004) as well as the ventromedial and dorsolateral regions of the prefrontal cortex (involved in executive functions and inhibition, among other things) (Chambers & Potenza, 2003; Dannon et al., 2011; Goudriaan et al., 2004; Grant, Brewer, & Potenza, 2006; van Holst et al., 2010). One of the behavioural manifestations of this neurobiology is a general decreased sensitivity to reward (Grant, Brewer & Potenza, 2006; Oberg, Christie, & Tata, 2011; Reuter et al., 2005). This, in turn, likely explains the elevated concordance of attention deficit disorder (Breyer et al., 2009; Carlton et al., 1987) as well as increased propensity for risk taking and sensation seeking (Gibbs Van Brunschot, 2009; Parke, Griffiths, & Irwing, 2004; Powell et al., 1999; Toneatto & Nguyen, 2007).

Another behavioural manifestation of this neurobiology is a stronger preference for immediate over delayed rewards (Chambers & Potenza, 2003; Goudriaan et al., 2004; Oberg, Christie & Tata, 2011; Parke, Griffiths, & Irwing, 2004; Petry & Madden, 2009; Shead et al., 2008; van Holst et al., 2010), which explains why problem gamblers are also often identified as having higher levels of impulsivity (Buchel, 2006; Chambers & Potenza, 2003; Goodman, 2008; Lawrence et al., 2009; Nower & Blaszczynski, 2006; Skitch & Hodgins, 2004; Steel & Blaszczynski, 1998; Toneatto & Nguyen, 2007; Turner, Jain, Spence, & Zangeneh, 2008; van Holst et al., 2010) Having genetic, neurobiological and/or behavioural characteristics that are opposite to those listed above provides some protection from the development of problem gambling.

9

Indirect Environmental Risk Factors The above biological propensities are then shaped by environmental experience (primarily social) to either further increase or decrease risk. Indirect environmental risk factors include: having an abusive or neglectful upbringing, parental involvement or modelling of gambling, peer group involvement in gambling, lower income, less education, societal acceptance of gambling, gambling opportunities being readily available, the presence of significant ongoing stressors and/or poor support systems, and gambling being commercially provided in an unsafe manner 2. Having the opposite experiences or environmental attributes provides some protection from engaging in gambling in the first place and/or developing problem gambling (Alegria et al., 2009; Cunningham-Williams et al., 2005; Grant & Kim, 2002; Gupta & Derevensky, 1998; Kausch, Rugle, & Rowland, 2006; Langhinrichsen-Rohling et al., 2004; Lester, 1994; Loo, Raylu, & Oei, 2008; National Gambling Impact Study Commission, 1999; Petry, 2005; Petry & Steinberg, 2005; Productivity Commission, 1999, 2010; Raylu & Oei, 2004; Welte, Wieczorek, Barnes et al., 2004).

Direct Risk Factors Ongoing involvement in gambling is influenced by all of the above indirect risk factors as well as several more direct risk factors which have a more immediate influence on the likelihood of becoming a problem gambler. Many direct risk factors constitute the ‘psychological’ realm in the biopsychosocial model.3 One of these concerns whether the person has erroneous cognitions or lack of knowledge about gambling. These erroneous cognitions (‘gambling fallacies’) include the failure to understand that most gambling outcomes are random and independent of past outcomes, that almost all forms of gambling have negative expected returns (i.e., the odds are biased against the gambler and it is not possible to make money over the long term), and that for most of these games there are no systems or behavioural strategies that can influence these random outcomes or the negative mathematical expectation (Chan & Ohtsuka, 2009; Fortune & Goodie, 2012; Gaboury & Ladouceur, 1989; Joukhador, Blaszczynski, & MacCallum, 2004; Joukhador, MacCallum, & Blaszczynski, 2003; Ladouceur, Sylvain, Boutin, et al., 2001; Ladouceur & Walker, 1996; Miller & Currie, 2008; Toneatto et al., 1997).

A second proximate risk factor concerns whether gambling serves a psychological need for the individual such as reliably providing temporary escape from other concerns or providing a very pleasurable (and needed) form of excitement. Alternatively, it may be that being identified as a ‘gambler’, and/or having money is a particularly important goal or measure of success to the

2

3

The characteristics of this factor will be discussed in detail in the next section.

Implicit in the present framework is the belief that a person’s psychology is primarily derivative of one’s biological endowment and life experience.

10

person (Blaszczynski & Nower, 2002; Nixon & Solowoniuk, 2009; Nixon, Solowoniuk, & McGowan, 2006; Volberg, Reitzes, & Boles, 1997).4 A third direct risk factor concerns the reward magnitude and frequency of the person’s early gambles. Having an ‘early big win’ appears to be an important determinant of gambling continuation (Turner, Zangeneh, & Littman-Sharp, 2006; Weatherly, Sauter, & King, 2004)5, as is engaging in ‘continuous’ forms of gambling that can provide a high frequency of reinforcement in a short period of time (i.e., electronic gambling machines, casino table games, ‘continuous’ lotteries; Dowling, Smith, & Thomas, 2005). Once gambling is regularly engaged in, operant and classical conditioning begin to increase the frequency and strength of the behaviour and the physiological processes underlying it, making it progressively more difficult to wilfully resist (Petry, 2005; Skinner, 1953). At a psychological level, the person begins becoming preoccupied with thoughts of gambling and planning opportunities to gamble. At a behavioural level, the person starts engaging in risky gambling practices such as playing more often and longer than intended, spending above planned spending limits, and chasing losses. Preoccupation and risky gambling practices constitute the most direct risk factors for problem gambling.6

Problem and Pathological Gambling In light of the negative consequences that begin to occur for the gambler, the number and strength of the direct and indirect risk factors determines whether gambling continues or abates. Gamblers who have developed impaired control are persistently unable to stay within time and financial limits, often chase their losses, and find themselves unable to cut back or quit gambling. Impaired control is usually associated with a range of negative consequences. As described in the Introduction, financial, psychological and social problems are the most common consequences, but health impairment, problems at school or work, and/or criminal/legal problems also sometimes occur. Significant negative consequences combined with impaired control constitute

4

Some of these psychological needs may derive from an abusive or neglectful upbringing, poor coping skills, lack of social supports, presence of severe stressors, etc. 5

People who report an ‘early big win’, also tend to report ‘early big losses’. Thus, it may be that ‘early big wins’ are simply reflective of heavy involvement in gambling (which is the true risk factor). 6

Combinations of these indirect and direct risk factors likely account for the association of problem gambling with other well-known variables. For example, the higher rate of gambling and problem gambling in young people (Williams, Volberg, & Stevens, 2012) is likely due to their greater propensity for risk-taking combined with their lack of knowledge and experience with gambling. The higher rate of gambling and problem gambling in males may be due to their higher rates of risk-taking and impulsivity in general (Cunningham-Williams et al., 2005). The higher rates of gambling and problem gambling found in people with Antisocial Personality (Blaszczynski & McConaghy, 1994; Meyer & Fabian, 1992; Slutske et al., 2001) is perhaps due to their higher rates of impulsivity, more materialistic orientation, and their adverse upbringing. The higher rate of gambling and problem gambling in people with Attention Deficit Disorder (ADD) is likely due to ADD’s strong association with impulsivity and risk-taking.

11

‘problem gambling’, with severe forms of impaired control and negative consequences being known as ‘pathological gambling’. In many people, the same risk factors that lead to problem gambling independently lead to problems in other areas (i.e., other addictive behaviour, mental health problems, interpersonal problems, poor health practices, school/work problems, antisocial behaviour) (Petry, 2007; Petry, Stinson, & Grant, 2005; Rush, Bassani, Urbanoski, & Castel, 2008; Westphal & Johnson, 2007). Thus, most problem gamblers will also have associated comorbidities. Within this constellation of problems, each reinforces each other’s existence, hampering recovery from any. Even with recovery, most problem gamblers will have an ongoing propensity for re-engaging in the behaviour, and a significant portion will experience at least a few relapses in their lifetime. Figure 1 illustrates the etiological framework described in the foregoing text. The particular pattern of risk factors that leads to problem gambling varies considerably between people, as does the age at which problem gambling develops. For example, in some people the combination of risk factors could be impulsivity + depression + parental gambling; in others it could be societal acceptance of gambling + gambling being readily available + erroneous cognitions or, alternatively, propensity for addiction + low income.

Having said this, the pattern of risk factors within an individual is not totally random. Rather, evidence points to at least two main subtypes or routes to eventual addiction: the impulsive/antisocial pattern (often in males) versus the emotionally vulnerable pattern (often in females) (e.g., Blaszczynski & Nower, 2002; Windle & Scheidt, 2004). This etiological framework has several important implications for prevention of problem gambling:  The effectiveness of any prevention initiative depends on its ability to modify one or more of these risk factors, particularly the direct risk factors.  Because of the large number of risk factors as well as the biological basis of some of them, the risk of problem gambling in a population can be reduced, but is unlikely to ever be eliminated.  Because many of the risk factors for problem gambling are the same risk factors for the development of substance abuse, psychopathology, and other problems, generic prevention initiatives targeting a wide range of problems (especially in youth) are likely both efficient and essential components for the prevention of problem gambling.  Because of the large number of risk factors, the effectiveness of any single prevention strategy will be limited.

Indeed, effective prevention will almost certainly require a multifaceted, coordinated, and sustained array of educational and policy initiatives provided across a wide range of age groups.

12

Figure 1. Etiological Framework for Problem Gambling.

Biological Risk Factors

Environmental Risk Factors Abusive/Neglectful upbringing

Self-controlled

↔ Impulsive

Risk aversive

↔ Risk seeking

Parental gambling

↔ Parental nongambling

Peer group gambling

↔ Peer group nongambling

Societal acceptance of gambling No predisposition to addiction

↔ Vulnerable to addiction Gambling readily available

No predisposition to mental health problems

Vulnerable to mental health

Low income

↔ problems

Low educational level

↔ Gambling not readily available

↔ High income ↔ Higher educational level ↔ Safe provision of gambling

High stress & low support

↔ Low stress & high support

Gambling serving a psychological need Rewarding early gambling experience & subsequent conditioning

Comorbid Mental Health Problems

Comorbid Interpersonal Problems

Societal nonacceptance of

↔ gambling

Unsafe provision of gambling

Erroneous cognitions or lack of knowledge about gambling

Comorbid Antisocial Behaviour

↔ Nurturing/Disciplined upbringing

Comorbid Addictive Behaviour

Problem Gambling

Comorbid Poor Health Practices Comorbid Work/School Problems

13

METHODOLOGICAL APPROACH The traditional way of categorizing prevention efforts is by the type of people toward whom efforts are directed. Primary Prevention is an effort to prevent individuals in the general populace from becoming problem gamblers. Secondary Prevention is an effort to prevent the development of problem gambling in individuals with risk factors for the condition. Tertiary Prevention is an effort to stop and potentially reverse the problems occurring in existing problem gamblers and is analogous to ‘treatment’. However, the approach used in the present paper differs in two respects. First, the focus is primarily on initiatives intended to prevent the development or onset of problem gambling (i.e., primary and secondary prevention) rather than a more general focus on initiatives intended to reduce the harm associated with gambling. (Inherent in this strategy is the assumption that intervention which effectively prevents problem gambling will also have beneficial effects on reducing the harm caused to non-problem gamblers.) Second, initiatives are grouped according to how they are intended to work. Educational Initiatives are intended to change internal knowledge, attitudes, beliefs, and skills so as to prevent problem gambling.

Policy Initiatives are intended to prevent problem gambling through the alteration of external environmental controls on the availability and provision of gambling. The purpose of the following section is to comprehensively review what is known about the nature and effectiveness of existing educational and policy initiatives to prevent problem gambling. A multi-stage search strategy was used to identify relevant articles. It started with the use of the keywords “prevention”, “gambling”, “problem gambling”, “pathological gambling” in the following electronic databases, restricting the search to articles published in English:  AGRI DSPACE Repository @ the University of Calgary – http://dspace.ucalgary.ca  Australian Gaming Council’s eLibrary - http://www.austgamingcouncil.org.au/  Academic Search Complete  Cochrane Database of Systematic Reviews  DARE (EBM Reviews)  E-Library – Responsible Gambling Council (Ontario)  Gambling Research Australia’s Gambling Research Database http://www.gamblingresearch.org.au/  Gambling Research Database (GambLIB)- http://www.gamblib.org/  Google Scholar  Medline  Problem Gambling Library (New Zealand) - http://www.pgfnz.co.nz/library  PsycINFO  Science Direct

14

The reference list in each of these articles was then consulted for additional articles. Finally, the authors a) used an ongoing Ingenta alerting service to identify all published articles with the word ‘gambling’ in the title; b) scanned the literature and news stories in the weekly gambling news services provided by Newscan (Responsible Gambling Council, Ontario), the Global Gaming Business Magazine, and the newsletter of the Global Gambling Guidance Group.

15

EDUCATIONAL INITIATIVES Childhood Interventions It is well established that negative early childhood experience significantly influences the development of problematic behaviour later in life. Accordingly, it is not surprising that interventions to strengthen families and create effective parenting practices are generally one of the most powerful ways to reduce adolescent problem behaviours, and to also serve to reduce problems at later ages (Foxcroft, Ireland, Lowe, & Breen, 2011; Kumpfer & Alvarado, 2003; Petrie, Bunn & Byrne, 2007). This approach is likely to apply equally well to the prevention of problem gambling among adolescents (and their subsequent adult years), although the approach has yet to be empirically tested. Nevertheless, there is good evidence that family-based programs are effective for the primary prevention of other addictive behaviour such as alcohol and drug use in young people (Foxcroft et al., 2011; Gates, McCambridge, Smith, & Foxcroft, 2006; Toumbourou, Williams, Waters, & Patton, 2005). It would be useful if future family/parenting interventions were to include the incidence of gambling and problem gambling among the outcome measures.

For similar reasons, it is to be expected that exposure to well socialized peer groups, supportive teachers, and good schools would have the same beneficial effect on prevention of problem gambling as it does on the prevention of other problematic behaviour (Durlak, 1997; Durlak & Wells, 1997; Nation et al., 2003; Toumbourou et al., 2005; Weissberg & Gullotta, 1997).

Information/Awareness Campaigns When most people think of problem gambling prevention they think of information campaigns targeted specifically at gambling. These are known variously as ‘information/awareness campaigns’, ‘mass media campaigns’ or ‘social marketing’. Such initiatives are directed at the general public and usually contain information consisting of one or more of the following elements (e.g., Alberta Alcohol and Drug Abuse Commission, 2001; Jackson, Thomas, Thomason, & Ho, 2002):  Encouragement to ‘know your limits’ or ‘gamble responsibly’.  Warnings about the potential addictive nature of gambling.  Identification of the signs/symptoms of problem gambling.  Information about where people can go for help or more information on problem gambling (i.e., treatment agencies; 24 hour telephone help-lines) (‘tertiary prevention’).  Provision of the true mathematical odds of various gambling activities.  Efforts to dispel common gambling fallacies and erroneous cognitions.  Provision of guidelines and suggestions for problem-free gambling.

16

These initiatives are usually developed and delivered by governmental health or social service agencies, schools, or commercial gambling providers. The information itself is provided:  On the gambling product (e.g., odds printed on the back of lottery tickets, ‘responsible gambling messages’ on electronic gambling machines, or EGMs).  On posters and pamphlets at gambling venues and elsewhere throughout the community.  In the form of ‘public service announcements’ on radio, television, and newspapers.  By means of presentations, plays (Bell, 2004), or videos (most often presented in educational settings).  On Interactive CDs.  On interactive terminals in the gambling venue (e.g., ‘Player Awareness Terminal’)  On government, social agency and/or gambling provider websites. Information/awareness campaigns are relatively inexpensive ways of delivering preventive health messages to a large portion of the population, and are intended to counter the often considerable commercial efforts to promote the product. The mass media are a particularly good vehicle for reaching young people, who are estimated to spend at least as many hours watching television as they spend in formal education. Reaching young people is particularly important, as they may be more susceptible to the considerable investment in commercial advertising promoting the product (Atkin, 1995; Strasburger, 1995).

Although awareness campaigns to prevent problem gambling are relatively common across many jurisdictions, there is limited research on their impact. The existing evidence suggests that improvements in knowledge and awareness are reliably produced in people who attend to these messages. For example, a brochure on pathological gambling was found to effectively convey useful new information to members of the general public in Quebec who were shown it (Ladouceur, Vezina, Jacques, & Ferland, 2000). Awareness initiatives appear to have a very limited impact if people are not explicitly asked to attend to the information or have no intrinsic interest in it. For example, Indiana implemented a statewide awareness campaign to promote responsible gambling using radio announcements, billboards, brochures, newspaper advertisements, posters, pens, and t-shirts, press conferences, and ‘problem gambling town hall meetings’. Najavits, Grymala, & George (2003) found that only 8% of the general public recalled seeing or hearing any advertising. However, of that 8%, 72% reported that the advertising had increased their knowledge of problem gambling. (One percent of the total sample reported taking action, such as calling the help line, as a result of seeing/hearing the message). A similar result was obtained in Ontario, Canada.

Turner, Wiebe, Falkowski-Ham, Kelly, & Skinner (2005) found that 66% of the Ontario public was unaware of any initiatives to reduce problem gambling. This is notable considering that Ontario is cited as spending proportionally more on problem gambling prevention, treatment, and research than any other jurisdiction in the world (Sadinsky, 2005). However, people who played slots and instant lotteries were significantly more likely to be report being aware of initiatives to reduce problem gambling, suggesting that gamblers may be noticing the presence of the help line number on tickets and slot machines. Also in Ontario, a media campaign to

17

dispel myths about how slot machines worked was successful in significantly reducing these fallacies among a random sample of 900 Ontario gamblers in February 2005 before the campaign started, compared to a random sample of 900 gamblers in April 2007 after the campaign was completed (B. Palantzas (Ontario Lottery and Gaming Corporation [OLG]), personal communication, June 6, 2007). In 1995 the Victoria Department of Human Services in Australia initiated a state-wide problem gambling awareness program consisting of a 5-week multi-language radio, newspaper and billboard advertisement phase in the first year, a 14-week television advertisement phase in 1996, and a 30-week radio and television advertisement phase in 1997 and 1998. Jackson, Thomas, Thomason and Ho (2002) evaluated the program, concluding that it produced an increased number of callers to the gambling helpline, and an increase in the number of new clients entering treatment. In addition, there was also increased collaboration between help services and the staff at gambling venues, many of whom attended training/information sessions about problem gambling. In 2001, the Victoria government initiated a similar informational campaign which reportedly resulted in a 70% increase in calls to the helpline and a 118% increase in clients presenting themselves to treatment (Victoria Department of Human Services, 2002).

It is important to note that providing support to distressed gamblers or recruiting problem gamblers into treatment is a much less satisfactory as a ‘prevention’ outcome than inoculating the general public from developing gambling problems in the first place. There is no direct evidence on the effectiveness of awareness campaigns as a primary prevention tool for problem gambling. The general public’s lack of awareness of these initiatives is not very encouraging. There have been several evaluations of short duration school-based presentations. The Addictions Foundation of Manitoba evaluated their 45 – 60 minute gambling education and awareness presentation (“It’s Your Lucky Day”) among 894 grade 7 and 8 students (Lemaire, de Lima, & Patton, 2004). One month after receiving the presentation, students in the Intervention group showed improved knowledge of gambling and problem gambling and decreased gambling fallacies relative to students in the Control group. The International Centre for Youth Gambling Problems and High-Risk Behaviors (2004) in Montreal, Quebec undertook an evaluation of their interactive CDs for the prevention of problem gambling (“Hooked City” for grade 7 – 12 students and “The Amazing Chateau” for grades 4 – 6).

Several months after being exposed to these interactive CDs, students had significantly improved knowledge about gambling, more awareness of the signs of problem gambling, and fewer gambling fallacies. However, there was no significant change in gambling behaviour, although a trend in this direction was found. In Ontario, a pre-post evaluation of a 60 minute program developed by the Centre for Addiction and Mental Health (CAMH) found the program significantly improved students’ understanding of random chance, but did not result in any change in gambling behaviour, coping strategies, or attitudes towards gambling (Turner, MacDonald, Bartoshuk, Zangeneh, 2008). Taylor & Hillyard (2009) evaluated a 45 minute program delivered to 8,455 students in primary school, junior high and high school in the Midwestern United States.

18

Immediately after receiving the program students showed significant improvement in their knowledge about gambling and the negative effects it could have. Behavioural change was not measured. Ferland, Ladouceur, & Vitaro (2002) evaluated the relative effectiveness of a 20 minute video (“Lucky”) compared to a 40 minute presentation, and a combined 40 minute presentation plus video among 424 grade 7 and 8 students in Quebec. One week later, all three groups had significantly improved knowledge about gambling and decreased gambling fallacies compared to a Control group, with the combined presentation plus video group having the greatest improvement. Ladouceur, Ferland, & Vitaro (2004) evaluated the same video (translated into English) in a group of 506 grade 7 and 8 students from Quebec and New Brunswick. Here again, after one week, students in the Intervention group had improved gambling knowledge and a decrease in gambling fallacies compared to students in the Control group. The same video was also evaluated against a presentation plus video session by Lavoie & Ladouceur (2004) in a group of 273 Quebec grade 5 and 6 students.

Immediately after seeing the video, students in both Intervention groups demonstrated significant improvements in gambling knowledge and a decrease in gambling fallacies compared to students in the Control group (with no greater advantage to the presentation plus video condition). Ladouceur, Ferland, Vitaro, & Pelletier (2005) evaluated the effectiveness of a different 20 minute video (focusing on the adverse consequences of excessive gambling) among 586 Quebec grade 11 and 12 students. One month later, students demonstrated improved knowledge of both gambling and problem gambling relative to a Control group. Ladouceur, Ferland, & Fournier (2003) evaluated the effectiveness of a one hour interactive presentation among 153 Quebec grade 5 and 6 students. Students who received the presentation demonstrated improved knowledge and decreased gambling fallacies relative to students in the Control group, with the effect being stronger when the session was administered by a specialist in this area, as compared to the students’ regular teacher. A similar one hour interactive session was evaluated in 345 Quebec grade 7 and 8 students by Ladouceur, Ferland, Roy, Pelletier, Bussières, & Auclair (2004). Again, improved knowledge and decreased gambling fallacies were achieved in the Intervention group relative to the Control group. There have been a couple of efforts to educate EGM gamblers.

Wohl, Christie, Matheson, & Anisman (2010) evaluated the effectiveness of a short video that provided education on how slot machines work, the value of setting financial limits, and strategies to avoid problems. Nonproblem gamblers (n = 242) were randomly assigned to view either the educational video or a control video. A total of 130 of these individuals (46% attrition) completed an evaluation questionnaire 1 day later and again 30 days later. After one day, those who watched the educational video were significantly more likely to stay within their pre-set slot spending limits, to have fewer erroneous cognitions, to endorse strategies to avoid problematic play, and to indicate an intent to use these strategies. However, there was no difference in the actual amount of money spent on slots between the experimental and control conditions. At the 30 day evaluation the decrease in erroneous cognitions was the only effect that was maintained.

19

A more substantial initiative7 with a more positive outcome was conducted by Doiron & Nicki (2007). These researchers delivered a two session program to 20 VLT gamblers who scored in the ‘At-Risk’ range on the Canadian Problem Gambling Index. Session 1 consisted of a 20 minute video providing information on gambling and problem gambling and a homework manual with a review of the presentation and exercises to practice cognitive restructuring and problem-solving. In Session 2 the homework was reviewed, further discussion occurred around the role of problem solving and faulty thinking, and a plan for the future was discussed. The program was evaluated after the second session and one month later compared to a wait-list control group of 20 people (100% retention of participants at one month). Large and significant decreases in erroneous gambling beliefs and actual gambling behaviour (# VLT sessions; nonVLT gambling sessions; VLT expenditure) occurred in the experimental group both at posttraining and one month follow-up. There is a considerable amount of literature on the utility of public education and awareness campaigns for other health behaviours that contain lessons for the prevention of problem gambling (Byrne, Dickson, Derevensky, Gupta, & Lussier, 2005).

In general, research has found that sustained information/awareness initiatives have significant potential to improve people’s knowledge and/or change their attitudes at a community-wide level (Carson et al., 2011; Duperrex, Roberts, & Bunn, 2002; Grilli, Ramsay, & Minozzi, 2004; Sowden, 2002). Indeed, population surveys have long been known to show that mass media are the leading source of information about important health issues, such as weight control, HIV/AIDS, drug abuse, asthma, family planning and mammography (Chapman & Lupton, 1994). While changes in knowledge and attitudes have been fairly reliably obtained, the ability of awareness campaigns to produce actual changes in behaviour is less common (Bala, Strzeszynski, & Cahill, 2008; Duperrex et al., 2002; Grilli et al., 2004; Slater et al., 2005; Sowden, 2002; Stacy, Bentler, & Flay, 1994). Furthermore, the knowledge and attitudinal impact of prevention messages often decays with time, requiring that they be repeated regularly (Duperrex et al., 2002; Vidanapathirana, Abramson, Forbes, & Fairley, 2004). Although less common, instances of clear behavioural change have sometimes been found. For example, after extensive media reports on dietary studies relating cholesterol-rich foods with heart disease, consumption of beef, eggs and fatty milk products in North America declined.

Similarly, reports on the risks of excessive sodium consumption were associated with increased use of salt-free food products. A decline in the use of birth control pills and IUDs between 1970 and 1975 correlated closely with publicity about their possible adverse effects (Jones, Beniger, & Westoff, 1980). Vidanapathirana et al. (2004) found that mass-media interventions have immediate effects in promoting HIV testing. Similar positive effects have been found for cancer screening (Bish, Sutton & Golombok, 2000), and the use of sunscreen. In the gambling context, evidence of behavioural change is seen the increase in help line calls or treatment-seeking as reported by Victoria Department of Human Services (2002) and Jackson et al. (2002) (and, 7

This initiative is better described as a training program rather than an educational session or awareness campaign.

20

anecdotally, by agencies in other jurisdictions when promoting these services). The commonalities to these situations are that the information is personally relevant, the consequences of not changing behaviour are potentially severe, the benefits of adopting the behaviour are quite beneficial, the person perceives that the behavioural change required is relatively easy to achieve, and the person receives plenty of cues for action. These necessary elements for behavioural change are elucidated in the ‘Health Belief Model’ (Becker, 1974; Janz, Champion, & Strecher, 2002), which has been shown to predict health behaviour in a wide variety of situations (Harrison, Mullen, & Green, 1992). The other situation where awareness campaigns (and other individual educational initiatives) have greater potential to contribute to behavioural change is when they are part of more comprehensive and coordinated community-wide programs involving several other educational and policy initiatives (see Best Practice 4 under Best Practices to Prevent Problem Gambling).

Responsible Gambling Information Centres ‘Responsible Gambling Information Centres’ (RGICs) located within gambling venues are a fairly new initiative. The Crown Casino in Melbourne, Victoria, Australia has operated the Crown Customer Support Centre since May 2002. In Canada, the first RGIC opened in 2003 at the McPhillips Street Casino, Winnipeg, Manitoba. Many other countries have since implemented RGICs (most beginning operations in 2005 and 2006). The primary purpose of the RGIC is to provide, on patron request, information and education about the risks of gambling (e.g. odds of winning and losing; demonstrations/tutorials about slot machine workings/random number generation). A second purpose is to identify, support, and refer RGIC visitors who are experiencing problems with gambling. Immediate crisis intervention and counselling may be provided by on-site staff, but ongoing counselling services are not usually included in RGIC mandates. RGIC’s are sometime staffed by casino employees and sometimes by employees of addiction agencies. In all jurisdictions, operational funding for RGICs has been provided either directly or indirectly by governments. Utilization rates for RGICs appear to be fairly low by patron utilization standards, although high by treatment provider standards.

Approximately 61,400 people visited one of Ontario’s Responsible Gaming Centres (RGC) from 2005 to 2009 (P. Pellazarri, personal communication, July 13, 2009). However, during this same time period there were roughly 200 million visitors to Ontario casinos. Approximately 8,000 customers are reported to have accessed Manitoba’s RGICs between 2003 and 2006. Of these, 75% were for information only, 10% for support and referral, and 15% for other reasons (Mehmel, 2006). By comparison, approximately 10,000 people visit Manitoba casinos every day. Significantly higher rates of contact have been reported in Alberta, where there were 100,000 customer contacts from 2006 to 2009 in their 15 provincial casinos (K.Verlik, personal communication, October 2009). The Alberta Gaming and Liquor Commission (AGLC) attributes this higher rate to a) RGIC staff being allowed on the gambling floor, and b) RGIC staff being employed by AGLC rather than Alberta Health.

21

A formal evaluation of the effectiveness of 2 of Ontario’s RGICs found that 70% of casino patrons were aware of the RGIC’s existence, and more than half of casino patrons were aware of its goals (Osborne Group, 2008). Among casino staff, 95% were aware of the RGIC and the large majority were aware of its goals. However, only 14% had ever referred a patron to the RGIC. The majority of the RGIC visitor population received information about problem gambling and/or responsible gambling, and the majority of these people rated this information as useful. Nine percent of the RGIC visitor population received a referral to a problem gambling treatment agency, but relatively few followed through on this referral (Osborne Group, 2008). An evaluation of visitors to an RGIC in Quebec (with a 28% participation rate in the evaluation survey) found that most respondents were seniors, occasional slot machine gamblers, and in control of their gambling activities. Satisfaction with the services of the RGIC was high. Finally, visits to the RGIC appeared to modify misconceptions about randomness, but did not have any immediate or long-term impact on gambling behaviour (Boutin, Tremblay, & Ladouceur, 2009).

Statistical Instruction As noted earlier, the advantage of awareness campaigns is their potential to reach large numbers of people. They have two main disadvantages. First, the exposure of any individual to the message is uncertain. Second, the duration of the message is quite short for people exposed to it. Thus, it is useful to examine the effectiveness of more sustained and directed educational efforts to prevent problem gambling. It is a fairly commonsensical idea that if gamblers truly understood the negative mathematical expectation of gambling, they would be inoculated from over-involvement. There are two general areas of research that support the contention that improved knowledge of gambling probabilities should positively impact gambling behaviour. The first is research demonstrating a positive impact of educating problem gamblers in treatment on the nature of randomness, and other errors of thinking underlying gambling fallacies (e.g., Ladouceur, Sylvain, & Boutin, 2000; Ladouceur, Sylvain, Letarte, Giroux, & Jacques, 1998; Sylvain, Ladouceur, & Boisvert, 1997). The second is research that shows statistically-trained college students to be less susceptible to certain specific fallacies (Benassi & Knoth, 1993), and to have better general reasoning skills for everyday problems (Fong, Krantz, & Nisbett, 1993; Kosonen & Winne, 1995).

However, the literature specific to the impact of improved statistical knowledge on gambling behaviour is mixed, at best. Lichtenstein, Slovic and Zink (1969) found that explaining the concept of ‘expected value’ (EV) and making it explicit in a set of gambling tasks resulted in only one-third of subjects moving to maximize EV (the desired outcome), with one-quarter moving toward lower EVs. Schoemaker (1979) found that university students who received statistical training made superior choices in a gambling task compared to untrained students. However, Gibson, Sanbonmatsu & Posavac (1997) found that students explicitly asked to evaluate the probability of a certain sports team winning tended to overestimate the team’s actual chances, and subsequently gambled more relative to students not asked to evaluate any specific team.

22

Similarly, Hertwig, Barron, Weber, & Erev (2004) found that students educated about the exact probabilities for events gambled on rare events more than was statistically optimal. In contrast, students who were not explicitly provided with the actual odds demonstrated a more optimal gambling strategy. Steenbergh, Whelan, Meyers, May, & Floyd (2004) found that university students who were given an explicit warning about erroneous gambling beliefs and the negative mathematical expectation of gambling gained superior knowledge about these things, but were just as likely to gamble on a roulette game compared to students not given these messages. In contrast, Floyd, Whelan, & Meyers (2006) found that viewing messages on irrational gambling beliefs caused university students to adopt less risky gambling behaviour in a computerized roulette game with imaginary money. Williams & Connolly (2006) provided 198 Introductory Statistics university students lessons on probability theory using examples from gambling as well as several laboratories providing hands-on experience with casino games. Six months after the intervention, students receiving the intervention demonstrated superior ability to calculate gambling odds as well as resistance to gambling fallacies. However, this improved knowledge and skill was not associated with any decreases or changes in actual gambling behaviour.

Similarly, Lambos & Delfabbro (2007) and Delfabbro, Lahn & Grabosky (2006) found that pathological gamblers had significantly higher rates of cognitive biases, but did not differ from nonpathological gamblers in knowledge of gambling odds or numerical ability. Lastly, Pelletier & Ladouceur (2007) found that people with knowledge of mathematics had an equal number of erroneous perceptions and behaviours during gambling compared to people with much less knowledge about mathematics. In many ways, it may be that teaching people about gambling odds is analogous to telling smokers about the harmful effects of smoking, or alcoholics about the harmful effects of drinking. Individuals involved in these behaviours are usually already aware of these facts. Knowing something and having this knowledge alter your behaviour are often two different things. While knowledge is, in most instances, a necessary antecedent to changing or preventing harmful behaviour, it is seldom sufficient on its own (e.g., Stacy, Bentler, & Flay, 1994; Williams & Gloster, 1999). Accordingly, prevention frameworks that rely heavily on providing gamblers with ‘informed choices’ (e.g., Blaszczynski, Ladouceur, & Shaffer, 2004; Blaszczynski, Ladouceur, Nower, & Shaffer, 2005; Blaszczynski et al., 2011) are likely to have limited preventive effect.

School-Based Prevention Programs Comprehensive and substantive elementary and high-school based prevention programs for problem gambling are relatively uncommon, but they do exist in some jurisdictions. Such programs typically have a broad scope of topics that include teaching statistical knowledge about gambling, providing information on the potentially addictive nature of gambling, explaining gambling fallacies, building esteem, and developing peer resistance skills. Examples of these types of programs are “Don’t Bet On It” in South Australia for ages 6 to 9; “Gambling: Minimising Health Risks” in Queensland for grade 5 students; “Facing the Odds” in Louisiana for grades 5 to 8; “All Bets are Off” in Michigan for grades 7 and 8; “Kids Don’t Gamble…Wanna

23

Bet” in Minnesota and Illinois for grades 3 to 8; “Youth Making Choices” for high school students in Ontario; “Count me Out” in Quebec for ages 8-17; the “Problem Gambling Prevention Program” in Florida for middle and high school students; and “Gambling: A Stacked Deck” in Alberta for grades 9-12. There has been very little published evaluation of these programs. Gaboury & Ladouceur (1993) evaluated a 3-session program (75 minutes per session) in 289 juniors and seniors from 5 Quebec high schools. The program included an overview of gambling, discussion of legal issues, how the gambling industry manipulates the chances of winning, gambling fallacies, the development of pathological gambling, and coping skills. Six months later, students in the Intervention group had improved knowledge about gambling relative to the Control group. However, the improvement in coping skills seen immediately after training was not maintained at 6 month follow-up, nor were there any significant change in students’ actual gambling behaviour or attitudes toward gambling at either post-test or follow-up. Ferland, Ladouceur & Vitaro (2005) evaluated a somewhat different 3-session program (60 minutes per session) with 1193 Quebec students in grades 8, 9 and 10.