Voluntary self-exclusion (VSE) programs enable problem gamblers to engage in a break from casino-based gambling. The current study analyzed the effects of a VSE program in British Columbia, Canada on problem gambling symptoms and the comparative reductions in problem gambling symptoms when participants abstained from gambling, continued to participate in non-casino based gambling, or attempted to violate their exclusion contract. 269 participants completed two telephone interviews over a 6-month period. Participants were administered the Problem Gambling Severity Index (PGSI). Substantial reductions in PGSI scores were observed after 6 months. Program violators had significantly smaller PGSI Difference Scores by Time 2 compared to those who continued to gamble outside of the casino and those who completely abstained from all gambling. There were no significant differences between those who gambled informally and those who abstained. A multiple regression identified that while access to counselling and length of enrollment also contributed to the reduction in PGSI scores, violation attempts were most strongly associated with smaller reductions in symptoms of problem gambling. These results imply that some gamblers can successfully engage in non-casino based forms of gambling and still experience reductions in symptoms of problem gambling. Future analyses will explore characteristics associated with group membership that may help to identify which participants can successfully engage in non-casino based gambling without re-triggering symptoms of problem gambling.