OBJECTIVE:To evaluate outcomes of clinical use of risperidone long-acting injection (RLAI) and determine factors predicting continuation with treatment.
METHOD:This prospective, 3-year follow-up of consecutive patients started on treatment with RLAI in normal clinical practice between August 2002 and September 2003 obtained demographic and clinical data from case notes, prescription charts, and hospital computer records. To determine predictors of continuation, a proportional hazards regression (Cox) model was constructed.
RESULTS:The study included 211 evaluable patients. Over 3 years, 84% of subjects discontinued RLAI; 27.7% of these switched to oral risperidone. The Cox model showed that younger age (p = .001), longer duration of illness (p = .001), inpatient status at initiation (p = .002), and an RLAI dose of 25 mg/2 weeks (p < .001) predicted greater probability of discontinuation.
CONCLUSION:A small proportion of patients initiated on treatment with RLAI continued for 3 years. Outcome is likely to be improved by targeting RLAI treatment at specific patient groups and by using a dose of more than 25 mg/2 weeks.