Medication usage patterns in patients with human immunodeficiency virus infection: a comparison of patient-reported medication usage with medical chart review

Patients infected with the human immunodeficiency virus (HIV) receive numerous medications from multiple providers. As a result, it is important that medication usage is properly documented in each patient's medical record. Lack of adequate documentation may confound a provider's assessment of drug efficacy, potentially leading to an increased incidence of drug interactions and adverse effects. The objective of this study was to determine if discrepancies exist between patient-reported medication usage and that documented in the medical record by healthcare providers. Data were obtained using structured telephone surveys and medical chart review. Study participants were recruited from the University of California, San Francisco Medical Center AIDS Clinic. Results obtained for 41 patients demonstrated discrepancies between patient-reported medication usage and that documented in the medical record ranging from 9 to 92 percent, depending on the class of drug. The largest differences were observed with the "as-needed" class of drugs: benzodiazepines (92 percent), morphine (60 percent), and codeine (56 percent). Differences were also noted for scheduled medications: ketoconazole (54 percent), clotrimazole (45 percent), acyclovir (38 percent), zidovudine (15 percent), and pentamidine (9 percent). These observed discrepancies reaffirm the need for accurate exchange of information between provider and patient to promote the most effective, rational, and safe drug therapy. Careful reviews of medication usage at each visit and use of pharmacy-based medication profiles are potential mechanisms to improve documentation of medication usage in HIV-infected patients.