In October the Iowa Department of Public Health released a statewide report on gambling behavior and gambling disorders (Gonnerman & Lutz, 2011). Using the NODS instrument[1]the study found the rate of pathological gambling (PG), the most severe form of the disorder, in the past 12 months to be 0.3 percent. Using the PGSI instrument, researchers found that past-year rates of PG to be 0.6 percent. These rates are similar to the prevalence rate found in large-scale national surveys, such as the National Comorbidity Survey Replication (NCS-R) that reported a past-year rate for pathological gambling of 0.3 percent (Kessler et al., 2008). The Iowa report used two screening measures to collect data: the National Opinion Research Center’s DSM Screen (NODS) (Gerstein et al., 1999) and the Problem Gambling Severity Index (PGSI), a validated brief measure based on the Canadian Problem Gambling Severity Index (CPGI) (Ferris & Wynne, 2001).
In the Iowa report, the researchers presented rates of problem gambling, which is a less severe form of the disorder that includes people who experience some gambling-related problems but not enough to be considered pathological gamblers (the PGSI refers to this group as “Moderate Risk”). The researchers found that 0.2 percent (NODS) and 2.6 percent (PGSI) of adult Iowans can be categorized as problem gamblers.
The difference between the NODS and PGSI rates may be due to the differing emphasis of the measures. The NODS was modeled on the American Psychiatric Association’s (APA) criteria for PG while the PGSI was designed to take more social and environmental factors into consideration. The PGSI’s authors have previously hypothesized that this emphasis has probably caused the PGSI to report higher rates of moderate and low risk gambling than other measures (Ferris & Wynne, 2001). However, both of the rates reported in the Iowa study are similar to rates found in other national studies (e.g., Kessler et al., 2008). It is also important to remember that people in both the subclinical group (those who experience some problematic symptoms but not enough for a problem or pathological classification) and problem gambling group often move to groups with both more and fewer symptoms (Shaffer & Hall, 2002). That is, today’s subclinical gambler could be tomorrow’s non-gambler.
The Iowa study also presented a variable that these researchers called “Any Problem Gambling Symptom.” This variable included everyone in the sample who had experienced any symptom on the NODS or the PGSI screens during the past year, as well as those who answered affirmatively to an additional question that the researchers added for people to self-identify as having gambling problems. The study found that 13.1 percent of the sample fit into this “Any Problem Gambling Symptom” group, though it was not clear from the study why the researchers combined these three separate measures in this way and reported it as one all-inclusive percentage.
The Iowa report represents the first time that the NODS, the PGSI and the additional question have been combined in this way. For this reason there are no other studies to compare this rate to, nor is there any data on the characteristics of this participant group or information on what this group is likely to do (or not do) in the future. This is atypical in research studies because new variables are typically released with substantial theoretical and psychometric data to explain precisely what it is measuring and how it could be useful to the field.
There are other potential problems with combining these measurements in this way. First, there has been very limited research on people with subclinical gambling problems and it is not clear that having one symptom is meaningful in population studies. This is particularly true for lifetime measures, but past-year measures can also be problematic. For example, if a person reports that they once drank a lot of alcohol and it took them a few days to fully recover, then they have met one criterion for alcohol dependence, according to the APA’sDiagnostic and Statistical Manual of Mental Disorders(American Psychiatric Association, 1994). However, this doesn’t mean that they should be considered “at-risk” for alcohol dependence or that they will develop alcohol dependence in the future (most people respond to negative consequences by changing their behavior). One place where it is important to know if a person has had even one symptom of a gambling (or alcohol) disorder is in a clinical setting. This information may be beneficial to a treatment provider as they try to understand the individual and conduct further assessment. However, the usefulness of this data in a population study is not well understood and needs to be the focus of significant research before it is used as a valid and useful variable.
Another concern is the tendency of the public, and sometimes the media, to misunderstand the implications of such a finding by believing that a larger percentage of people are “at-risk” for developing a gambling disorder. Media reports that are not clear about subclinical gambling problems may add to this confusion. Is this report suggesting that 13.1 percent of Iowans will become pathological gamblers? The researchers did not draw this conclusion, and new research, referenced above, shows that individuals with some gambling problems, but not PG, are as likely to move back to health as they are to more problems. In other words, being “at-risk” is not necessarily a slippery slope to PG. While this study found rates of PG and problem gambling similar to the rest of the U.S., more research is needed on people that display at least one symptom of gambling disorders but do not meet diagnostic criteria for problem or pathological gambling.
References
American Psychiatric Association. (1994).DSM-IV: Diagnostic and Statistical Manual of Mental Disorders(Vol. Fourth). Washington, DC: American Psychiatric Association.
Ferris, J., & Wynne, H. (2001). The canadian problem gambling index: user’s manual.Toronto (ON): Canadian Centre on Substance Abuse.
Gerstein, D., Murphy, S., Toce, M., Volberg, R. A., Harwood, H., & Tucker, A. (1999).Gambling Impact and Behavior Study. Report to the National Gambling Impact Study Commission, April 1, 1999.(p. 104). Chicago, IL: National Opinion Research Center. Retrieved from http://ezp-prod1.hul.harvard.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=nts&AN=PB2002-104073%2fXAB&site=ehost-live&scope=site
Gonnerman, M. E., & Lutz, G. M. (2011).Gambling Attitudes and Behaviors: A 2011 Survey of Adult Iowans(p. 125). Cedar Falls. IA: Center for Social and Behavioral Research, University of Northern Iowa.
Kessler, R. C., Hwang, I., LaBrie, R., Petukhova, M., Sampson, N. A., Winters, K. C., & Shaffer, H. J. (2008). DSM-IV pathological gambling in the National Comorbidity Survey Replication.Psychol Med,38(9), 1351-60. doi:S0033291708002900 [pii] 10.1017/S0033291708002900
Shaffer, Howard J., & Hall, M. N. (2002). The natural history of gambling and drinking problems among casino employees.Journal of Social Psychology,142(4), 405-424.
[1]Discussed with other measures, below.
NCRG staffResearch UpdateIowaprevalence rates