Is gambling involvement a confounding variable for the relationship between Internet gambling and gambling problem severity?

Internet gamblers have more problems gambling than land-based gamblers, but recent studies showed that Internet gamblers are involved in a higher number of gambling activities, which may confound the relationship between Internet gambling and gambling problems. The present study aimed to test whether the relationship between Internet gambling and gambling problems persisted when including variables related to gambling involvement as predictors, namely time spent gambling and diversity of gambling formats. Data from a large sample of French adolescents (n=9910) were used. Associations between disordered gambling/money spent gambling with Internet gambling were performed using generalized linear models, not controlling and controlling for diversity of gambling formats and time spent gambling. The results showed that Internet gamblers had significantly more problems than land-based gamblers. The relationship decreased when diversity of gambling formats and time spent gambling were controlled separately, and became non-significant when they were both included in the model. To conclude, time spent gambling and diversity of gambling formats rather than Internet gambling should be considered a detrimental gambling behavior. They seemed to capture different aspects of gambling patterns. This study was a step forward in changing the conceptual model of problem gambling, with gambling involvement as a main variable. Time spent gambling and diversity of gambling formats predicted problem gambling.Internet gambling did not predict problem gambling above gambling involvement.Gambling involvement was a main variable to understand problem gambling.