At-risk gambling in patients with severe psychiatric illness and in community subjects matched for age and sex.

In the recent literature the rates of gambling in psychiatric patients have been compared only indirectly with those found in community samples and no study has so far matched a clinical sample with community controls. We selected 875 outpatients attending two community mental health centers and 3.500 community subjects, matched for age and sex. At-risk gambling was defined according to the four categories of the Canadian Problem Gambling Index (CPGI) scores: 0 no-risk, 1-2 low-risk, 3-7 moderate-risk, 8+ high-risk. Data were also collected on substance, alcohol, and tobacco use. Patients were diagnosed with schizophrenia, bipolar disorder, unipolar depression, cluster B personality. At-risk gambling was significantly higher in psychiatric patients compared to community subjects. In the univariate multinomial logistic regression analysis, high-risk gambling was associated with lifetime substance use and being unmarried, moderate-risk with age at onset of alcohol use and lifetime tobacco use, and low-risk with higher education. In the multinomial logistic regression analysis high risk-gambling in psychiatric patients was four times that of community controls, while in substance users high-risk gambling was two times that of non-users. The results suggest that screening for gambling could improve the care of psychiatric patients who suffer from a comorbid behavioral addiction.