Following previous studies,16 we used two categories of problem gambling: moderate risk of gambling problems (PGSI score 3–7) and severe risk (PGSI score 8 or more). Statistical analyses were performed using SPSS version 21. To determine the clinical correlates of problem gambling in bipolar disorder, people rated as being at moderate or severe risk of problem gambling were compared with those at no or low risk (PGSI score <3) on a range of demographic and clinical variables using chi-squared tests, Mann–Whitney U-tests (owing to significant non-normal distributions of continuous variables) and multivariate binary logistic regression (using enter method). All tests were two-tailed and tested at a threshold of statistical significance of P<0.05.
