Factors associated with pathological gambling at 10-year follow-up in a national sample of middle-aged men (2007)

Background: The present analyses will expand on previous reports by considering the impact of eight psychiatric disorders and genetic vulnerability to problem (P) and pathological gambling gambling (PG). Methods: Diagnoses of DSM-III-R life-time P and PG were derived in 1992 and past-year P and PG in 2002 from 1675 individual twins from the Vietnam Era Twin Registry. Logistic regression was used to predict past-year P and PG as a function of socio-demographics and life-time co-occurring psychiatric disorders including gambling problems measured in 1992. Co-twin analyses accounted for familial contributions to past-year gambling problems. Results: High school or greater educational attainment was associated with less likelihood of current P and PG. With the exception of alcohol dependence and generalized anxiety/panic, all disorders studied remained associated significantly with an increase risk of past-year P and PG after adjusting for 1992 gambling symptoms. Past-year P and PG was associated significantly with the number of pathological gambling symptoms reported in 1992. After controlling for genetic and family environmental factors, one or more 1992 symptoms were associated with 2002 symptoms. Conclusions: Education and substance dependence, mood and antisocial personality disorders were associated with current gambling. A history of PG symptoms is the strongest predictor of past-year problem gambling. (Author' s abstract)