Abstract:
Aims Schizophrenia is a chronic illness, with approximately two-thirds of patients experiencing relapses, often with rehospitalisation. Treatment with antipsychotic medications reduces the relapse rate. Despite half a century of antipsychotic drug availability, doubts remain regarding the translation of research findings into day-to-day practice or into clinical practice guidelines. This study therefore aimed to explore this efficacy–effectiveness debate by examining prescribing correlates of rehospitalisation in a large cohort of treated patients. Method Four hundred and fifty-one inpatients discharged with diagnoses of schizophrenia or related disorders in three distinct New Zealand districts between July 2009 and December 2011 were tracked until December 2013. Utilising a national mental health database, rehospitalisation rates and duration were thus obtained for two years following discharge. Discharge variables including treatment history were obtained from clinical records and individual clinicians. Results In contrast to treatment guidelines, relatively many (34%) were prescribed multiple antipsychotics and fewer (20%) than expected received clozapine. Māori were prescribed clozapine more frequently (24%) than non-Māori (13%). Compulsory treatment was associated with the use of more long-acting injectable medications than in voluntary patients. Clinician characteristics did not predict prescribing patterns. Nearly half (44%) of the cohort were rehospitalised within two years. Those with a longer (> 3 weeks) index admission (HR = 0.53, p = 0.001) were less likely to be rehospitalised, as were older patients (> 50 years) (HR = 0.58, p = 0.04). Those subject to compulsory treatment appeared more likely to be rehospitalised (HR = 1.3, p = 0.06) and spent more time rehospitalised (p = 0.05). Antipsychotic types, routes and dosages were not significantly associated with rehospitalisation, except in the case of clozapine (HR = 0.61, p = 0.01). Conclusion Observed prescribing practice aligned with existing guidelines, except for antipsychotic polypharmacy and clozapine underutilisation. Only the latter appeared to be ethnically influenced. Rehospitalisation rates were higher for patients under the age of 50 and for those with shorter index admissions. Other than the beneficial effect of clozapine, the type and route of prescribed antipsychotics did not significantly affect rehospitalisation rates. This study does not support any claimed advantages of second-generation over first-generation antipsychotics.