RESEARCH QUESTIONS
Is Internet gambling a common occurrence among college students? Do Internet gamblers have high rates of pathological gambling (PG)? Is Internet gambling associated with poor mental health?

PURPOSE
The two main purposes were: to be the first to examine rates of Internet gambling in a large sample of college students and to compare prevalence rates of PG and indices of mental health in Internet and non-Internet gamblers.

HYPOTHESES
Internet gambling would be fairly common among college students, and Internet gamblers, especially those with frequent Internet wagering, would have high rates of PG. Internet gambling would be associated with poor mental health.

PARTICIPANTS
One thousand three hundred fifty six students (42% males; average age = 21 years) were recruited from three American university campuses between March 2005 and May 2006.

PROCEDURE
A research assistant asked students walking by or before class started to complete a brief questionnaire. The questionnaire included basic demographic items, as well as questions regarding lifetime participation in several forms of gambling (including Internet gambling). Gambling severity and global physical and mental health were also assessed. Additional questions asked about recent gambling frequency and expenditures, and desire for gambling treatment.

MAIN OUTCOME MEASURES
The South Oaks Gambling Screen (SOGS) assessed gambling severity, with scores > 5 indicative of lifetime probable pathological gambling. The General Health Questionnaire assessed global physical and mental health, with items asking how one had been feeling in the past few weeks. Higher scores reflected poorer health.

KEY RESULTS
Overall, 77% of participants were not Internet gamblers, 17% were infrequent Internet gamblers (i.e., they had gambled ‘‘1–10 times’’ or ‘‘more than 10 times but never weekly’’), and 6% were frequent Internet gamblers (they gambled “weekly or more than once a week but never daily” and “daily”). Thus, almost 25% of participants admitted to having wagered on the Internet at least once in their lives. Internet gamblers were more likely to be male and Caucasian than non-Internet gamblers.
Gambling-related problems. Internet gamblers were more likely to endorse lifetime gambling problems (i.e., higher SOGS scores). About one third of participants who had ever gambled on the Internet were classified as probable pathological gamblers (approximately 12% of the entire sample were lifetime probable pathological gamblers). A larger proportion of Internet gamblers, especially frequent Internet gamblers, wagered more than $100 in the prior two months than non-Internet gamblers. When all types of gambling were considered, it was found that frequent Internet gamblers wagered more often in the prior two months than those in other groups.
Health. Probable pathological gamblers had poorer global physical and mental health than non-pathological gamblers PG was associated with poor mental health, even after controlling for campus and demographic characteristics known to impact mental health ratings (e.g., age and gender). Even after taking PG status and basic demographics into account, Internet gambling was a significant predictor of poor mental health, with greater Internet gambling frequency associated with the worst mental health ratings. Finally, when asked whether they were interested in learning about problem gambling and ways to reduce gambling, 61% of frequent Internet gamblers responded affirmatively, versus 38% of infrequent and 19% of never Internet gamblers.

LIMITATIONS
The directionality of the relationship between Internet gambling and PG could not be determined. Although the GHQ was a reliable and valid indicator of global health status, its focus was more on emotional rather than physical health. Further, the SOGS, used to classify students as probable pathological gamblers, has been criticized for potentially overestimating rates of PG relative to DSM-based instruments. The study was conducted in only three university settings in a single state in the United States, and recruitment utilized a convenience strategy rather than one of random selection. It is thus unclear how generalizable the results may be to other populations.

CONCLUSIONS
Internet gambling is common among college students, and it is associated with distress, even among non-regular Internet gamblers. Prevention, early intervention, and treatment efforts should target college students who wager on the Internet so as to reduce harm associated with such gambling.