Antiretroviral Medication Errors among Hospitalized HIV-Infected Adults

0 Clinical Infectious Diseases 2012;55(11):1585-6 The Author 2012. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions , please 1 Assistant Professor, Temple University School of Pharmacy , 3307 N Broad St., Philadelphia, PA 19140 CORRESPONDENCE - TO THE EDITORYehia and colleagues [1] recently described a current estimate of antiretroviral (ARV) medication errors among hospitalized HIV-infected patients at one institution. Among a total of 308 admissions with ARV medications prescribed in 2009, there were 145 ARV medication errors in 110 admissions on the first day (29%) and 22 errors in 21 of 308 admissions on the second day (7%). They conclude that ARV medication errors are typically corrected within 48 hours. At Temple University Hospital (TUH), a 740-bed tertiary care teaching hospital in Philadelphia, Pennsylvania, we found similar rates of ARV medication errors. From April 2010 to March 2011, 290 HIV-infected patients admitted with orders for ARV medications were analyzed prospectively until discharge, contributing to 994 patient-days of follow-up. A total of 84 errors were identified in 75 patients, resulting in 26% of patients admitted with at least 1 medication error. Consistent with data from Yehia and colleagues [1], most errors occurred with a protease-inhibitor (PI)-containing regimen (82%). At TUH, an HIV clinical pharmacist reviews a list of all ARVs dispensed in the hospital on a daily basis. The pharmacist evaluates the patients ARV regimen and ensures that (1) the regimen contains agents from at least 2 different classes (2) all doses and frequencies are accurate (including renal dosage adjustments) and (3) there are no contraindicated ARV-drug interactions. Although not specifically measured, nearly all pharmacy recommendations were accepted and errors were often corrected within the same day. It is very likely that these errors would have persisted beyond 48 hours had there not been interventions made by the clinical pharmacist, although since they are corrected this cannot be assessed. Heelon and colleagues [2] have shown that the duration of ARV prescribing errors was decreased when a clinical pharmacist evaluating ARV medication orders intervened to resolve errors (15.5 hours versus 84 hours, P < .0001). In the study conducted by Yehia and colleagues [1], two clinical pharmacists specializing in infectious diseases similarly reviewed all medication orders to identify ARV medication and drug interaction errors. Therefore, it is not surprising that error rates dramatically decreased on the second day of hospitalization. It would be interesting to know if these results can be reproduced in a hospital that does not have specialized services such as clinical pharmacists trained in infectious diseases who proactively review ARV medication orders. The conclusion made by Yehia and colleagues [1] that most ARV prescribing errors are corrected within 48 hours is unlikely to be generalizable to hospitals without these resources. In fact, the title and abstract were slightly misleading; it implied that ARV errors were corrected within 48 hours without any specific interventions. The fact that clinical pharmacists trained in infectious diseases have the task of correcting these errors should have been mentioned in the abstract, since the Yehia and colleagues [1] essentially evaluated their impact and could serve as a template for institutions interested in addressing the issue of ARV errors. Potential conflicts of interest. J. C. G. is a consultant for Cubist Pharmaceuticals, is on the speakers bureau for Pfizer and Optimer, and received study grants from Merck & Co. and Cubist Pharmaceuticals. C. W. H. has no potential conflicts of interest. Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. (...truncated)