A number of large scale epidemiological studies have been conducted internationally and in the US to determine the prevalence of problem/pathological gambling in the general population; however, less work has examined the prevalence of pathological gambling among gambling venue patrons. The current study, using data from a sample of individuals surveyed in a casino setting, sought to determine the frequency of gambling problems in a sample of gambling venue patrons. Further, the current study sought to examine alcohol and tobacco use, health status, and quality of life differences by gambling problem status in this sample.
Based on prior work, our primary hypotheses were as follows: (a) rates of problem and pathological gambling would be higher among casino patrons relative to rates for these disorders found in general population samples; (b) individuals with gambling problems would report higher quantities of tobacco and alcohol use; (c) individuals with gambling problems would report lower ratings for health and quality of life.
Prevalence of Gambling Problems
The results of our analyses supported our first hypothesis related to increased rates of gambling problems among casino patrons relative to the general public. Based on the NODS score classification described above, 30.3% of respondents were non-problem gamblers, 10.7% were problem gamblers, 29.2% were at-risk gamblers, and 29.8% were pathological gamblers. The NODS questions assessed lifetime gambling-related problems, therefore, these categories reflect lifetime rates of gambling disorders and not necessarily frequency of current gambling disorders.
The prevalence rates obtained in our sample were higher than those obtained in a nationwide casino sample (Gerstein et al. 1999
) and in a study of casino patrons in the UK (Fisher 2000
). Gerstein and associates found that nationwide, 17.9% of casino patrons were at-risk gamblers, 5.3% were problem gamblers, and 7.9% were pathological gamblers. Fisher (2000
) and Gerstein et al. (1999
) present weighted prevalence estimates that controlled for the likelihood of being sampled. We did not weight our prevalence estimates because we had no measure of the frequency of casino attendance. Had we weighted our data we may have obtained lower rates of pathological gambling than those we observed using un-weighted data.
Conversely, the rate of pathological gamblers found in the current study was lower than that obtained in a study of problem/pathological gambling done in Brazil (43.5%) (Oliveira and Silva 2000
). Oliveira and Silva (2000
) used the SOGS scores to classify pathological gamblers and found a higher rate of probable pathological gamblers than we found in our data. In the current study, we employed the NODS, which has been shown to more strictly define gambling problems (Hodgins 2004
); further, the SOGS has been shown to have good agreement with DSM-IV criteria for pathological gambling, but may overestimate gambling problems in non-clinical, general population samples (Stinchfield 2002
The obtained high rate of pathological gambling in our study may be due to the fact that we had a primarily male sample (78.7% male), and that our sample was primarily non-Caucasian (69.1%). Male gender and non-Caucasian ethnicity have been associated with higher rates of problem gambling (Fisher 2000
; Gerstein et al. 1999
; Volberg 1996
). Although in our sample we found that problem gamblers were more likely to be male, just under 16% of women surveyed in this study were pathological gamblers.
Relative to non-problem gamblers, a greater number of pathological gamblers showed heavy current gambling involvement as indicated by gambling 10 or more days in the last 30 and gambling greater than $400.00 in the last month. The majority of pathological gamblers (61.8%) reported that they gambled to make money.
In the current sample, pathological gamblers were more likely to report ‘action’ oriented game play (e.g., cards, betting on animals, betting on sports), with the exception that they also reported increased frequency of lottery play. This may be due to the nature of the casino from which data were collected. The casino offers only card games, therefore, it likely attracts those individuals with a primary interest in more action oriented games. Our sample may have a bias towards action oriented gamblers for this reason.
Alcohol and Tobacco Use
Our hypotheses regarding alcohol and tobacco use were partially supported. Somewhat consistent with other studies, we found that pathological gamblers smoked more cigarettes per day than non-pathological gamblers (Cunningham-Williams et al. 1998
; Grant and Potenza 2005
; Petry and Oncken 2002
; Potenza et al. 2004
). We also found that smokers had higher NODS scores relative to non-smokers, which is consistent with the report of more severe gambling pathology among pathological gamblers who smoke (Petry and Oncken 2002
). Unlike past work, we did not find a relationship between alcohol use and pathological gambling (Cunningham-Williams et al. 1998
; Welte et al. 2001
; Welte et al. 2004
). The failure to find an association between alcohol use and pathological gambling may be a result of our alcohol use measurement method. Had we employed alcohol abuse or dependence diagnoses, we may have detected a relationship between pathological gambling and alcohol abuse/dependence.
Health and Quality of Life
We hypothesized that individuals with gambling problems would report worse health than non-problem gambling casino patrons and that individuals with gambling problems would report reduced quality of life relative to non-problem gambling casino patrons. In the case of self-reported health, our hypothesis was not supported. Although pathological gamblers reported lower self-rated health than non-pathological gamblers, the difference was not statistically significant beyond the trend level. But, for self-reported quality of life, our hypothesis was supported. Pathological gamblers reported lower quality of life than non-pathological gamblers. Mean self-reported quality of life scores for the at-risk and problem gambling groups were higher than those reported by the pathological gambling group, but lower than those reported by the non-problem gambling group.
Evidence from our data suggest that individuals with gambling problems were aware of family histories of gambling problems and may have insight into their own gambling problems. Significantly more individuals with gambling problems reported that someone in their family had or has a gambling problem. Approximately 84% of individuals who reported 3 or more gambling-related problems as assessed by the NODS indicated that they felt that they either had or currently have a gambling problem.
Despite the fact that a high percentage of individuals who reported 3 or more gambling-related problems also indicated that they felt they may have a gambling problem, it is unlikely that these individuals have sought or will seek treatment. In the US lifetime treatment seeking among individuals with pathological gambling (PG) disorder is low (Kessler et al. 2008
; Slutske 2006
) as compared with psychiatric disorders such as substance-related disorders and major depression (Kessler et al. 1998
). Slutske (2006
) compared rates of treatment seeking among individuals with PG surveyed as part of two national studies: the Gambling Impact and Behavior Study (GIBS) and the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). In the GIBS data only 7.1% of the lifetime pathological gamblers reported seeking professional treatment or participation in self-help groups. In the NESARC data only 9.9% of the lifetime pathological gamblers sought professional treatment or had attended at least one Gamblers Anonymous meeting. Kessler et al. (2008
), in an analysis of data from the National Comorbidity Survey Replication (NCS-R), found that of study participants meeting lifetime criteria for PG, none reported seeking treatment specifically for gambling problems; however, 49% of those with a lifetime diagnosis of PG reported treatment for substance disorders or emotional problems at any point in their life. Comparatively, for addictive disorders and major depression, roughly 50 and 70%, respectively, make treatment contact with a physician, mental health professional, or other professional over their lifetime (Kessler et al. 1998
Among reasons such as embarrassment regarding gambling behavior, denial of a gambling problem, social stigma, and concerns about the effectiveness of treatment, the simple lack of available services has been cited as a barrier to PG treatment utilization (Rockloff and Schofield 2004
In order to increase awareness of gambling treatment availability, effectiveness, and utilization, casino-based interventions for gambling problems may be necessary. Such interventions could include posting information about available treatment services, formalized screening, identification, and referral procedures for individuals with gambling problems in casino settings, and/or gambling problem screening kiosks at key locations within a casino. The incorporation of curricula on the recognition of the signs and symptoms of problem gambling into standard training practices for casino staff, coupled with a formal procedure to refer interested individuals to gambling treatment services could also be implemented. Our examination of alcohol and tobacco use, health status, and quality of life suggest that interventions for smoking cessation and improving quality of life are needs identified among casino patrons with gambling problems.
Study Strengths and Limitations
The current research may be characterized as having a number of strengths. First, the data come from a ‘real-world’ casino setting rather than in a laboratory setting, or a college population. Second, we employed a three-day round-the-clock sampling method that increased the likelihood that individuals with varying gambling patterns would be sampled. Third, we employed a purpose-built measure, the NODS, in order to assess for pathological gambling. The NODS was designed specifically to operationalize pathological gambling criteria for community based studies. Finally, all data were collected anonymously, which may have increased the likelihood that respondents would provide accurate and reliable information regarding gambling and associated behaviors.
Findings from the current study must be considered in light of a number of limitations. First, we present data gathered primarily as a convenience sample from a single Los Angeles County casino. Systematic bias may have been introduced in our sample as a result of specific aspects of the casino from which data were sampled, by lack of selection criteria for inclusion in the study, or participant characteristics which may be related to choosing to participate in research surveys. A second limitation was the fact that our data are entirely self-report and may be subject to recall bias, social desirability bias, and other distortions. Finally, the study was limited by the fact that we did not use DSM-based measures for substance disorders in our survey and used single-item quality of life and health status measures. More refined measurement techniques would have allowed for more detailed analysis of group differences within these domains.
Summary and Conclusions
In the current study of casino patrons, we found higher rates of pathological gambling relative to some previous work (Fisher 2000
; Gerstein et al. 1999
), but lower rates than work conducted in Brazilian gambling venues (Oliveira and Silva 2000
). Gambling related variables were significantly different in that pathological gamblers were more likely to report a family history of gambling problems, to report more frequent gambling with larger sums of money, to report engaging in sports betting and skilled gambling, and to report gambling to make money. A high percentage of pathological gamblers acknowledged having a gambling problem. Individuals with gambling problems reported more smoking, but not more drinking. Pathological gamblers reported lower quality of life, which may be due in part to a trend for lower self-reported health.
The larger implications of our findings are that, given the potentially high rate of gambling problems among casino patrons, there is a need for formal prevention and intervention measures in casino settings. More data from casino samples may be helpful for state and county policy makers in that they provide information upon which ‘evidence-based’ legislation regarding gambling regulations and controls could be based. Legislators may serve the public interest and prevent untoward consequences resulting from gambling problems by supporting the implementation of primary, secondary, and tertiary prevention efforts in casino settings. The current study––as well as the studies mentioned in our review of the literature––have identified a high frequency of gambling-related problems among casino patrons. Such problems are associated with poor health, substance use, reduced quality of life, psychopathology, and family/social problems. Carefully designed prevention and treatment programs that account for the significant comorbidity present among individuals with problem/pathological gambling disorders are needed. These programs, at minimum, should cooperate with gambling venues by providing training in recognition of gambling problems among patrons, providing casinos with information to provide treatment referrals for patrons with gambling problems.