Casino Gambling and Symptoms of Gambling Addiction: It May Be Worse Than You Think

The American Gaming Association maintains that “only about 1 percent of the population [is] classified as pathological gamblers and another 2 percent experience problems gambling.”  Yet, given the estimated cut-off points for “at-risk” gambling and the reported number of visits and losses of casino gamblers, we might expect higher rates of gambling addiction.

First of all, the AGA does not accurately report the available data.  They ignore one national survey with the highest estimates of gambling addiction, and misrepresent percentages of “problem gamblers” understood as those reporting any symptoms of gambling addiction.[1]  Four different studies conducted between 1998 and 2003 give a more accurate range of estimates.  Those estimates of individuals with 1 or more symptoms of gambling addiction range from 2.9% to 11.5% for lifetime rates, and 2.8%-5.5% for past year rates [2]

But more importantly, estimates of the national lifetime prevalence of “problem gambling,” or “pathological gambling,” whether 3% or 11%, do not tell us much about casino gambling.  True, of all the people who have gambled at least once in their lives, most gamble without symptoms of gambling addiction; but most people are not gambling regularly at casinos or racetracks.  Most people gamble infrequently, maybe joining an office pool or playing cards with friends.  Pointing to this fact is a diversion from the real issue: what are the rates of gambling addiction among those who go to casinos?[3]

They are startlingly high.

A 2010 survey of patrons at a California casino found that nearly 70% of respondents reported at least one symptom of gambling addiction.[4]  While the sample in that survey was not based on a random sample, a 1996 survey that was based on a random sample of residents of Connecticut also found less than 1/3 of casino patrons were problem free, with 72% of respondents reporting at least one symptom of gambling addiction and 15% reporting more than five.[5]  The lowest estimate of non-problem gamblers, also based on a non-random sample, is from a 1999 survey of patrons at 5 different casinos in Nevada and Atlantic City.  That study found that 68.4% of casino patrons reported no symptoms of gambling addiction.[6]

Whichever estimate is closer to the truth for casinos in general – the estimates that 2/3 of casino patrons reported at least one symptom of gambling addiction, or the estimate that only 1/3 of casino patrons reported any symptoms of gambling addiction – the idea that casinos offer harmless entertainment is hardly supported by the evidence.[7]  The study of a California casino’s patrons concluded:

“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.”[8]

FOOTNOTES

[1] The misrepresentation occurs because some studies will report those with 1-4 symptoms of gambling addiction as “problem gamblers,” while for others the cut-point is 3 symptoms, labeling those with 1-2 symptoms as “at-risk.”  So, in order to maintain something close to a 3% estimate, the AGA must report percentages of “problem gamblers,” but not “at-risk” gamblers.  Including all gamblers reporting problems, no matter the reported category, changes the estimates of gambling addiction.

The “at-risk” label itself is misleading since, as a team of Australian researchers point out, “[w]hen the label ‘at-risk’ gambling has been used, it typically  describes individuals who have existing gambling problems but at a level that falls short of clinical criteria.”  (Rodgers, Bryan; Caldwell, Tanya; Butterworth, Peter;  Measuring Gambling Participation; Addiction; 2009, July, Vol. 104 Issue 7; p. 1066).  In addition, these reserachers not, defining “at-risk” gambling as those with one or two symptoms “may be useful for developing secondary (or indicated) prevention, but it is not well suited to primary prevention.” (Ibid).

[2]  Those studies are:

National Opinion Research Center (NORC). (1999). Gambling Impact and Behavior Study, Report to the National Gambling Impact Study Commission. Chicago, IL

Welte, John; Barnes, Grace; Wieczorek, William; Tidwell, Marie-Cecile; Parker, John  (2001); Alcohol and Gambling Pathology among U.S. Adults: Prevalence, Demographic Patterns and Cormorbidity;  Journal of Studies on Alcohol; Sep2001, Vol. 62 Issue 5, p706

Howard J. Shaffer, Matthew N. Hall; Updating and Refining Prevalence Estimates of  Disordered Gambling Behaviour in the United States and Canada Canadian Journal of Public Health; Volume 92, No. 3; May-June 2001.

R. C. Kessler, I. Hwang1, R. LaBrie, M. Petukhova1, N. A. Sampson, K. C. Winters; and H. J. Shaffer; DSM-IV pathological gambling in the National Comorbidity Survey Replication; Psychological Medicine (2008), 38, 1351–1360.

Use of different survey methods to assess symptoms of Pathological Gambling as well as a lack of specificity on a time frame for the estimate (whether “lifetime,” those reporting symptoms ever in their life, “pasty year,” either reporting symptoms in the past six months, or reporting in the pasty year) can also cause confusion.  Three studies cited above that were based on a national survey each used a different survey instrument (i.e. the set of questions used to assess levels of gambling problems).  The fourth study, Schaffer et al., was based on a “meta-analysis,” a research method that uses results from a number of different surveys, not all with the same survey instrument. 

So, more specifically:

NORC (1999) used a random digit dial national survey conducted in 1999, using their newly developed NORC DSM–IV Screen For Gambling Problems found 10% lifetime, and 2.8% past-year reporting at least one gambling problem, with non-gamblers at 14.4% lifetime, and 36.7% past-year.

Welte and Barnes, (2002) based on a random digit national telephone survey conducted between August 1999 and October 2000 found 5.5% current, 11.5 lifetime had at least one symptom of gambling addcition when using the most often used survey instrument, the South Oaks Gambling Survey.  They also used the Diagnostic Interview Schedule and found 3.5% current, and 4.8% lifetime had at least one symptom of gambling addiciton.

Schaffer et al. (2001), a meta-analysis of studies conducted in the United States and Canada between 1975 and 1999 found 6.1% lifetime, and 4.0% past-year reporting any symptoms of gambling addiction.  An earlier meta-analysis with fewer studies by the same research team found 5.45% lifetime, and 3.94 past-year rates of any symptoms of gambling addiction.  A secondary analysis of this previous meta-analysis using studies only form the United States found 5.4 lifetime, and 2.9 past-year rates of any symptoms of gambling addiction.

Kessler et al. (2008) used data from the US National Comorbidity Survey Replication conducted between 2001 and 2002.  The measures of gambling addiction were part of the Composite International Diagnostic Interview.  This study found 3.9% lifetime reporting any symptom of gambling, and .06% past-year reporting 5 or more symptoms.

Another study, not reported above, by Petry 2005 (Petry NM, Stinson FS, Grant BF (2005) Comorbidity of DSM-IV pathological gambling and other psychiatric disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions; Journal of Clinical Psychiatry. 2005 May; 66(5):564-74).  found a lifetime pathological gambling rate, reporting 5 or more symptoms, of 0.42.  That is the lowest estimate of any national survey.  Unfortunately we could not get access to the entire study to present a full report on the findings.

[3] There are only a few “patron surveys” available.  All the studies that have conducted patron surveys point out the need for more patron surveys.  Some are not based on a random sample (meaning those responding were not chosen as randomly as possible, thus limiting our ability to make accurate generalizations about gamblers at a casino or casinos).  Yet they should, nevertheless, raise alarm bells.  The numbers of people suffering from gambling addiction are shockingly higher than the national survey numbers.

[4] Timothy W. Fong; Michael D. Campos; Mary-Lynn Brecht; Alice Davis; Adrienne Marco; Viviane Pecanha; Richard J. Rosenthal; Problem and Pathological Gambling in a Sample of Casino Patrons; Journal of Gambling Studies (2011) 27:35-47.

Respondents were scored using the NORC DSM–IV Screen For Gambling Problems (NODS).  They found, lifetime, 30.3% of respondents were non-problem gamblers, 29.2% were at-risk gamblers (1-2 symptoms); 10.7% were problem gamblers (3-4 symptoms), and 29.8% were pathological gamblers (5 or more symptoms).

[5]  WEFA Group; A Study Concerning the Effects of Legalized Gambling on the Citizens of the State of Connecticut; Prepared for State of Connecticut: Department of Revenue Services Division of Special Revenue; June 1997.

[6]  Although the NORC study did not have a purely random sample, efforts were made to simulate a random, representative sample.

[7]  For some guidance on estimates of Philadelphia area casino patrons we can turn to a study based on a national survey in Brazil that found spending more than 20% of household income for gambling was “highly predictive” of Pathological Gambling. (Hermano Tavares; Elizabeth Carneiro; Marcos Sanches; Ilana Pinsky; Raul Caetano; Marcos Zaleski; Ronaldo Laranjeira; Gambling in Brazil: Lifetime prevalences and socio-demographic correlates in Psychiatry Research, 2010, Vol. 180, No. 1, pp. 35-41).

A gambler who visits 3 times a week, losing 25$ a week loses $3,600 a year.  David Jonas of Parx casino suggest most of his customers visit this often or more. Using the 20% cut-off as a predictor, at those levels of gambling participation, anyone with a household income under $50,000 is likely suffering from Pathological Gambling.  In 2011 64% of Philadelphians had a household income of less than $50,000. (United States Census Bureau, American Community Survey).

[8]  Fong et al. 2011; p. 45