Aripiprazole is a new antipsychotic with a different mode of action to established second generation antipsychotics. We aimed to study patients who were prescribed aripiprazole in routine clinical practice, to identify patients who had a good clinical response.

From a data set of 21,000 electronic patient records (starting in 2002), we retrospectively identified all secondary care psychiatric patients started on aripiprazole (n=180). We assigned an improvement score of Clinical Global Impression to these records to measure the effectiveness of aripiprazole. We examined demographic and clinical correlates of patients who improved (CGI scores <5) versus those who did not improve (CGI≥5).

Adequate records for analysis were available for 120 patients. 77 patients (64%) had a CGI 1-4 (minimally to very much improved). 43 patients (36%) had a CGI≥5 (no change to very much worse). The discontinuation rates were 17% (improved group), and 43% (no change to worse group) Those who did well could not be distinguished in terms of age, sex, mean duration of record availability (approx 700 days), diagnosis (>80% psychosis), duration of contact with services, or initial dose of aripiprazole (10mg). Patients who improved with aripiprazole were less likely (p<0.01) to be treatment resistant (previous or subsequent treatment with clozapine). Discontinuation was primarily due to agitation (29%) followed by inefficacy (23%) and worsening psychosis (10%).

Aripiprazole was clinically effective in around two-third of patients. Favourable response was associated with lack of treatment resistance. Agitation followed by inefficacy were the commonest reasons for discontinuation.