ABSTRACT
Background
This was a 4-year mirror-image study of adult patients diagnosed with bipolar disorder (BD) assessing the effects on treatment continuation and hospitalisation between aripiprazole 1-month (A1M), risperidone-LAI (R-LAI) and the monthly and 3-monthly formulations of paliperidone palmitate (PP1M, PP3M). We aimed to evaluate and compare the use of A1M, R-LAI, and the monthly and 3-monthly formulations of paliperidone palmitate (PP1M, PP3M) by using the change of number and length of hospitalisations 2 years before compared to 2 years after initiation of LAIs for continuers and discontinuers. Secondary outcomes were: (1) discontinuation rates at 2 years and reasons per LAI, (2) time to discontinuation per LAI, and (3) time to first hospitalisation per LAI.
Results
A total of 122 BD were included; 74 continued LAI treatment at two years. Reasons for discontinuation were poor compliance (50%), ineffectiveness (43.2%), and tolerability issues (13.6%). Both time to individual LAI discontinuation and time to first hospital admission were significantly lower in the R-LAI group. There was a significant overall reduction in the number and length of hospitalisations two years before and after LAI initiation, although multivariate logistic regression analysis showed that A1M, PP1M and R-LAI were associated with an increased risk (OR = 1.89, 95% CI = 1.54–3.68, p = 0.015; OR = 1.63, 95% CI = 1.29–2.77, p = 0.022; OR = 3.08, 95% CI = 1.48–6.05, p = 0.008, respectively) of bed usage compared to PP3M. Last, study completers showed a considerable drop of 79% in number of hospital admissions and 83% in bed days (p = 0.001) as opposed to non-completers.
Conclusions
Study findings suggest that long-acting antipsychotics such as A1M, PP1M, and particularly PP3M are associated with high retention and lower hospitalisation rates after 2 years of treatment in patients with BD.