Fall 1997

A recent report evaluating Minnesota state-funded compulsive gambling treatment programs concludes that people who participate in treatment reduce their compulsive gambling behavior and its negative consequences, and that the state's investment in funding compulsive gambling treatment is having positive results. These programs were developed in response to concerns about growth in compulsive gambling activity.

The report, Evaluation of the Minnesota State-Funded Compulsive Gambling Treatment Programs, was conducted by Abt Associates of Cambridge, Massachusetts, an internationally recognized evaluation firm. The study was commissioned by the Minnesota Department of Human Services' (DHS) mental health division, which oversees the state gambling treatment program, to determine:

• Does treatment for problem gamblers work¼
• Does one type of treatment work better than others¼

The evaluation was based on data gathered from February 1992 to January 1997 from approximately 1,800 individuals in six state-funded compulsive gambling treatment programs building on survey work done previously by University of Minnesota researchers. Major conclusions of the study, based on results of a multiple regression analysis, include:

• Treatment reduces compulsive gambling behavior and appears to be delivered at a reasonable cost by the programs surveyed. (The average treatment program lasts six months and costs about $3,000.)
• No treatment model was found to be more effective than any other.
• One-third of problem gamblers who received treatment for compulsive gambling also received treatment for either chemical dependency or mental health problems.
• Gamblers who wagered larger amounts (i.e., $1,000-$10,000) in a day were more likely to complete treatment than those who did not.
• Those with more than a high school education were more likely to enter and complete treatment than those with less education.
• People with incomes below $10,000 a year were four times more likely to incur gambling debt than those with higher incomes. Low-income people had higher debt-to-income ratios, meaning they were least likely to be able to repay their debt and might gamble even more to resolve their debt problems. The study included people who completed treatment, left treatment prematurely, and declined treatment. Gamblers were evaluated six and twelve months after they left treatment. Seventy percent of those admitted completed treatment and:
• Had a greater reduction in their gambling behavior than those who were partially treated or received no treatment.
• Were less likely to have legal and personal problems than those untreated.
• Would have relapsed had treatment not been provided.

Two additional ways to describe the findings are based on changes in weekly gambling and scores from the South Oaks Gambling Screen (SOGS).

The South Oaks Gambling Screen is a 20 item questionnaire for pathological gambling developed by Henry Lesieur, Ph.D. and Sheila Blume, MD. A score of five or more affirmative responses is an indication of probable pathological gambling.

Changes in Weekly Gambling The regression analysis provides a basis for answering the question: By how much does treatment reduce compulsive gambling? As illustrated in FIGURE 1, at baseline prior to treatment, the average gambler assessed by the six treatment programs had better than an 0.80 probability of gambling on at least a weekly basis. Following treatment, the estimated probability fell to about 0.11 at six and twelve months. If the average gambler had not been treated, the estimated probability of gambling at least weekly would have been about 0.36 to 0.54, depending on when the measurement was taken and whether the gambler was partially treated or untreated. Although far from precise, these estimates nevertheless suggest a significant positive treatment effect.

Changes in SOGS Scores Another way to examine the effectiveness of treatment is by using the SOGS score. As illustrated in FIGURE 2, at baseline, the average gambler assessed by the six treatment programs had a SOGS score of 11.3. Following treatment, the estimated SOGS declined to about two. But the estimated SOGS score was higher for those who were not treated-between five and six depending on whether the gambler was partially treated or untreated and depending on when the measurement was taken. Another way of interpreting this finding is that, on average, compulsive gamblers who completed treatment had estimated SOGS scores that were below the threshold of five or higher for probable pathological gambling, while those who did not receive or did not complete treatment had estimated scores that were above the threshold for pathological gambling.

The report also describes characteristics of compulsive gamblers, the types of gambling in which they participate and problems they experience as a result of their gambling. This information will be provided in future issues of Beyond the Odds.

"The results of this study show that treatment is helping people to stop or significantly reduce their compulsive gambling," said Sharon Autio, director of the DHS mental health division. "The results will help us look to new directions in treatment and to target those services to groups who most need it."