There are three major implications this study provides for the development of policy.
The first is the clear need for the development of treatment opportunities within the State for youth who are problem gamblers. With that said, it must be noted that the authors were only able to identify three adolescent gambling treatment programs in the US and Canada, thus suggesting a dearth of examples upon which to build a program. Along with this lack of collective experience comes an inability to accurately estimate the numbers of adolescents that might access treatment if it were available. For the Oregon adult prevalence study, Volberg (1997) utilized a projected penetration rate of 3% to determine estimated numbers of problem gamblers that should be expected to be seen in treatment. The numbers of adults accessing treatment in Oregon currently fall within that range (Moore, 1998) suggesting this estimated penetration rate, based primarily on penetration rates for alcohol and drug treatment as appropriate. Experience would suggest that penetration of adolescents to treatment is lower, (possibly around 2% comparing alcohol and drug treatment) than that for adults, suggesting a somewhat lower benchmark for gambling youth accessing treatment.
It should be noted that these estimates should be considered in light of the following caveats. Because these estimates are derived from survey and not clinical data there is no practical way to estimate how many of the level 3 gamblers in this study are subject to the exclusionary criteria suggested in the DSM IV. In the case of pathological gambling, a manic episode might better account for problematic gambling behavior in at least some of the youth (APA, 1994).
There is also no agreed upon clinical or theoretical basis to calculate false positive and negative classification of problem gamblers. "In order to determine a false assignment, scientists must invoke a standard against which we judge the classification system" (Shaffer, Hall and Vander Bilt, 1997, p. 70). As suggested in the introduction, there is currently no gold standard by which to judge the validity of survey estimates. The SOGS-RA has, however, been shown to be a reasonably valid and reliable instrument for assessing gambling behavior (Winters et al.,1993a) and as such is likely to balance false positive and negative classifications (Shaffer et al., 1997). Thus, although the SOGS-RA might not meet the high standards of accuracy required for a clinical screen, it certainly provides adequate population estimates of level 2 and level 3 gambling.
The second implication from this study is the need to develop prevention activities aimed at early intervention into problem gambling. The findings of this, and similar studies, suggest a relationship between the age of first gambling and the development of level 2 and level 3 problem gambling. Primary and secondary prevention may well be appropriate at the grade, middle, and high school levels. Primary prevention, for parents who gamble, may also hold some value in reducing future problems.
Finally, findings from this study, consistent with other studies, also suggests an association among high risk behaviors pointing towards a prevention message that is blended with existing prevention efforts for other high risk behaviors. Although this study suggests that Oregon's experience with under-aged gambling activities associated with Lottery and Casino gambling is quite similar to other states, policy makers may wish to explore if these reported rates of illegal gambling activity are acceptable.