Development and reliability testing of the clinical pharmacist recommendation taxonomy.

Study Objective: To evaluate the reliability of a newly developed taxonomy--the Clinical Pharmacist Recommendation (CPR) taxonomy--to classify clinical pharmacy interventions.

Design: The CPR taxonomy was developed and refined in three phases. In each phase, reviewers independently reviewed recommendations made by a clinical pharmacist-physician team and categorized them into mutually exclusive categories: priority, problem, and response. Interrater reliability was assessed for all categories during each development phase.

Setting: Primary care clinics of a Veterans Affairs Medical Center.

Patients: Fifty-three patients enrolled in the Veterans Affairs Enhanced Pharmacy Outpatient Clinic (EPOC) trial.

Measurements And Main Results: Interrater reliability was assessed using the kappa statistic. A total of 423 recommendations were evaluated during the three testing phases. In the final testing phase, agreement was moderate for pharmacotherapy problem subcategories (kappa = 0.57), substantial for pharmacotherapy problem primary categories (kappa = 0.64), and almost perfect for response categories (kappa = 0.85). Taxonomy completion time/patient averaged 4.6 minutes (range 1-11 min).

Conclusion: The CPR taxonomy provides a reliable method to systematically evaluate clinical pharmacy recommendations based on the therapeutic problem identified and specific action recommended to resolve the problem.