The WAGER, Vol. 17(7) – You’re cut off! Can casinos identify problem gamblers?


“Hey buddy, I think you’ve had too much. Let me call you a cab.” We are all familiar with this bar scene. The bartender decides when a drinker should be cut off, based on how many drinks he has had and how he is acting. But, unless a gambler is suspected of cheating, we might not see similar behavior from a dealer in a casino. Why not? Unlike problem drinking or drugging, problem gambling does not produce specific, easily observable, and predictable behavior. This week, the WAGER reviews a study that tested whether casino employees could use their knowledge of regular players to identify whether these players had gambling problems (Delfabbro, Borgas, & King, 2012).


  • Researchers intercepted 303 (195 women, 108 men) gaming machine gamblers in seven south Australian gambling venues.
  • Gamblers completed a survey that included questions about their gambling behavior, as well as the Problem Gambling Severity Index (PGSI: Ferris & Wynne, 2001).
  • Venue staff, after researchers pointed out the gambler in question, completed surveys about each participating gambler, indicating how well they knew the gambler, how often they thought he or she gambled at the venue, and whether they thought the gambler had any problems with gambling.


  • Almost fifty percent (49.8%) of the gamblers who agreed to participate in the study played gambling machines weekly or more frequently, compared to less than 5% of the southern Australia general population.
  • The PGSI classified 31% of the sample as at moderate or higher risk for gambling problems, and 8.6% of the sample as actual problem gamblers.
  • Staff were familiar with three quarters of the participants in the study; those participants who were familiar to staff were more likely to report being frequent gamblers at the casino than other participants (58% compared to 20%).
  • As Table 1 shows, the staff identified 35 patrons as having at least some gambling problems. Most (30) of them reported being at some risk for gambling problems.
  • However, as Table 1 also shows, staff failed to identify many of the gamblers who reported problems.

Table 1. Staff and Participant Report of Gambling Problems (adapted from Delfabbro et al., 2012)

Staff Reported Status Gambler Self-Reported (PGSI) Status
  No Problems Low Risk Moderate Risk Problem Gambler
No Problems 77 55 49 14
Some Problems 2 8 10 7
Problem Gambler 3 2 2 1
Sensitivity (# of actual problem gamblers correctly identified by staff / # of actual problem gamblers):

  • For problem gambling (1/22): 5%
  • For some problems/moderate problems (20/83): 24%
Specificity (# of people with no gambling problems correctly identified by staff / # of people with no gambling problems):

  • For no problems – 77 / 82 = 94%
  • For no problems / low risk – 132/147 = 90%

Note. Table 1 N = 230 – only gamblers whom staff recognized.


  • The sample was not representative of gaming machine gamblers, because only a small proportion of those recruited agreed to participate.
  • Identifying problem gamblers for a research project is not the same as identifying them for intervention; staff might be unwilling to act on those assessments if such action would have consequences for the gambler or the venue.


This study suggests that the gamblers whom the staff identifies as having problems are very likely to be problem gamblers. However, staff identify very few of these individuals – the overwhelming majority go undetected. If these findings are replicated, interventions administered by trained venue staff might be more effective than originally thought. Though many problem gamblers would be missed, those targeted would likely be at risk for gambling problems. However, any intervention employed would need to be carefully designed and evaluated to minimize unintended consequences (see Op-Ed, March 21, 2007).

-Sarah Nelson

What do you think? Please use the comment link below to provide feedback on this article.


Delfabbro, P., Borgas, M., & King, D. (2012). Venue staff knowledge of their patrons’ gambling and problem gambling. Journal of Gambling Studies, 28, 155-169.

Ferris, J., & Wynne, H. (2001). The Canadian Problem Gambling Index: Final Report. Ottawa, ON, Canada: Canadian Centre on Substance Abuse.

2 thoughts on “The WAGER, Vol. 17(7) – You’re cut off! Can casinos identify problem gamblers?

  1. valarie j. fane Reply

    In today’s society of lawsuits gone wild, I believe the casinos would be very hesitant to implement such an intervention unless the patrons would somehow agree to this (in writing) prior. It is a well-meaning intervention intended to save persons from themselves; however, I’m not sure how it would work.

  2. Theresa A. Mataga Reply

    I found the article interesting as I usually wouldn’t or couldn’t leave the slot machines until I was broke.
    I began to think in a different light when Internet gambling reared its ugly head. Then I began to think about the affect Internet gambling would have on children between the age of 10 and 18 who play games on the computer on a daily basis.
    How would the children cope with gambling addiction?
    The children will not understand what was happening to them and they will not be able to cope with the problem. Youth gambling addiction on the Internet is going to be an epidemic sooner than you think.

Leave a Reply

Your email address will not be published. Required fields are marked *