Adverse drug reaction monitoring in psychiatry out-patient department of an Indian teaching hospital.

objectivesAdverse drug reactions (ADRs) to psychotropic agents are common and can lead to noncompliance or even discontinuation of therapy. There is paucity of such data in the Indian context. We deemed it worthwhile to assess the suspected ADR profile of psychotropic drugs in an ambulatory setting in a public teaching hospital in Kolkata. MATERIALS AND

methodsA longitudinal observational study was conducted in the outpatient department (OPD) of the concerned psychiatry unit. Twenty consecutive patients per day, irrespective of their psychiatric diagnosis, were screened for suspected ADRs, 2 days in a week, over 15 months. Adverse event history, medication history and other relevant details were captured in a format as adopted in the Indian National Pharmacovigilance Programme. Causality was assessed by criteria of World Health Organization-Uppsala Monitoring Center (WHO-UPC).

resultsWe screened 2000 patients (68.69% males, median age 34.4 years), of whom 429 were suspected of having at least one ADR; 84 cases had insufficient evidence about causality (WHO-UMC causality status "unlikely") and were excluded from further analysis. Thus, 17.25% (95% confidence interval: 15.59-18.91%) of our study population reported ADRs with at least "possible" causality. Of 352 events recorded, 327 (92.90%) were "probable" and the rest "possible". None was labeled "certain" as rechallenge was not performed. Patients received a median of 3.2 psychotropic drugs each. Thirty-three different kinds of ADRs were noted, including tremor (19.60%), weight gain (15.34%) and constipation (14.49%). Among the incriminated drugs, antipsychotics represented the majority (57.10%), with olanzapine topping the list.

conclusionsThis study offers a representative profile of ADRs to be expected in psychiatry out-patients in an Indian public hospital. Establishment of a psychotropic drug ADR database can be a worthy long-term goal in the Indian context.