Validation of the lie/bet screen for pathological gambling on two normal population data sets.

The validity of the Lie/Bet Screen was tested on two community population samples, one adult (n=2,014) and one adolescent sample (n=3,237), in Norway. With positive responses on at least one of the questions on Lie/Bet Screen used as the cutoff point the screen showed high both sensitivity and specificity. The negative predictive value was also high, but the positive predictive value was comparatively lower. A prediction of probable pathological gambling or "At-risk gambling" based on both Lie/Bet questions identified a valid screening in the two samples (0.54% in adults, 5.6% in adolescents). Compared to the use of the full DSM-IV this is pretty close, with the figures 0.45% and 5.22%. It is concluded that the Lie/Bet Screen may function as a good screening device for pathological gambling plus At-risk gambling in normal community samples.