Polypharmacy in an Australian teaching hospital. Preliminary analysis of prevalence, types of drugs and associations.

A computer-based prescription retrieval system was used to study 21,521 prescriptions that had been provided to hospital patients who were receiving predominantly outpatient care. Over a three-month period 338 patients were found to be receiving 10 or more different drugs concomitantly. A further 338 patients were drawn at random from the same outpatient population for comparison. Age was linked significantly to polypharmacy (polypharmacy group: mean age, 63.7 years, SEM = 1.09; comparison group: mean age, 53.8 years, SEM = 1.00, P less than 0.05; chi 2 = 62.8, P less than 0.001). The relative risk of polypharmacy was related linearly to age. Admission to hospital was associated with increased prescribing rates in the polypharmacy sample (P less than 0.05), as was attendance at multiple clinics and multiple attendance at outpatient clinics (P less than 0.05 and P less than 0.05, respectively). Benzodiazepine agents were included in 63.7% of prescriptions in the polypharmacy group and in 37.3% of prescriptions in the comparison group. Non-prescription drugs were noted in 97.2% of prescriptions in the polypharmacy group and 58.0% of prescriptions in the comparison group, representing 34.7% and 27.3% of all items, respectively. In patients of less than 30 years of age agents for allergy/asthma/atopy contributed most to polypharmacy; agents that were associated with renal failure most in patients aged 31-50 years; and agents for cardiovascular disease contributed most in patients aged over 50 years. Our results suggest that a reduction in the use of non-prescription and psychotropic agents, heightened awareness of the dangers of polypharmacy and coordination and integration of over-all care and prescribing habits should reduce polypharmacy materially.