Assessing the appropriateness of physician prescribing for geriatric outpatients. Development and testing of an instrument

In a randomized, controlled trial evaluating the impact of clinical pharmacists' consultations on physicians' prescribing decisions, a seven-member physician/pharmacist panel developed an instrument to characterize and quantify the full range of drug-prescribing problems. The instrument was tested for validity and reliability.

The instrument was applied in reviewing prescribing for 236 geriatric outpatients.

To ensure internal validity of the instrument, five panel meetings were devoted to reaching a consensus on procedures, categories, criteria, and scoring. Each case was evaluated independently by two reviewers and each drug in the regimen was evaluated for drug-drug interactions, dosage, schedule, allergic reactions, therapeutic duplication, use of drugs without established diagnoses, and inappropriate drugs for the patient's clinical conditions; discrepancies were adjudicated by the panel chairperson. Self-reported adverse effects of 60 study patients who were interviewed, and hospital readmissions of all study patients were used to test the external validity of the instrument.

Inter-rater reliability improved over time, reaching a high of 97.1 percent after six months. A positive association was determined between the prescribing scores and the number of reported adverse effects (rho = 0.28, p = 0.02). Although the number of patients with drug-related hospital readmissions was too few to establish a significant association between the prescribing scores and readmissions, the findings were expected.

Given current requirements for drug utilization reviews in hospitals, health maintenance organizations, and Medicaid programs, this instrument may prove to be very useful.