Pathological gambling is a disabling disorder experienced by about 1% of adults. We randomised 233 participants (41.6% women) 1:1:1 to nalmefene (20 or 40 mg) or placebo. In analyses performed using an intention-to-treat (ITT) population, nalmefene failed to show statistically significant differences from placebo on primary and secondary outcomes. Post hoc analyses of only participants who received a full titration of the medication for at least 1 week demonstrated that nalmefene 40 mg/day resulted in significantly greater reductions on the primary outcome measure. These findings suggest that medication dosing may be an important consideration in achieving symptom control.