Customized Order Entry Sets Reduce Errors in Prescribing HIV Medication

Implementation of computerized customized order entry sets in a large urban hospital reduced HIV medication error rates by more than 35%.

Customized order entry sets (COES), a computerized program that displays dosing recommendations, identifies drug interactions, and reminds prescribers to order complete drug regimens, may help reduce errors in inpatient antiretroviral (ARV) medication prescriptions, according to a recent retrospective study.

Caryn Weiss, an adult nurse practitioner at the Montefiore Medical Center in Bronx, NY, and colleagues reported their findings on October 3, 2013, at IDWeek 2013, a joint meeting of the Infectious Diseases Society of America (IDSA), the Society for Healthcare Epidemiology of America (SHEA), the HIV Medicine Association (HIVMA), and the Pediatric Infectious Diseases Society (PIDS), in San Francisco, CA.

Previous research indicates that the HIV medication error rates may be as high as 84% of inpatient hospital admissions. According to Weiss and colleagues, prescribers may be unfamiliar with the complexity of the ARV regimens. Having specialized HIV pharmacists and informing prescribers through educational material and staff in-services may help reduce such errors; however, they may not be available at all facilities. In the current study, Weiss and colleagues investigated whether COES helped reduce dosage errors.

The researchers analyzed retrospectively the charts of 723 (before COES) and 661 (after COES) hospitalized adult patients for whom ARV was prescribed. Patient demographics, including age, CD4 count, viral load, and hospital length of stay, was similar between the two groups. Protease inhibitor (PI) therapy was the most prescribed treatment during both time periods, although there was a trend toward more non-PI-based therapies after the implementation of COES.

Overall prescription errors decreased 35% (from 38% to 24%) after COES implementation. The reduction in errors occurred across all treatment types, although PI-based therapies experienced the greatest improvements. Dosage, frequency, and drug interaction errors significantly decreased with the PI-based therapies after COES implementation. Additionally, the renally adjusted dosage of nucleoside reverse transcriptase inhibitors (NRTIs) improved after COES implementation. The proportion of incomplete treatment regimens across all treatment types was also reduced after COES.

Weiss indicated that COES may be a cost-effective method for prescribing HIV medication in centers that do not have HIV-dedicated pharmacists and inpatient HIV medical teams. For COES to be most effective in reducing prescription errors, Weiss indicated it should be updated with the most current medications and prescribing patterns for HIV treatment.

The authors do note some limitations to the study. Because the study took place at one treatment center, the results may not be applicable to other centers prescribing HIV therapies. In addition, the investigated prescription errors using a retrospective chart review, meaning that other variables, such as the ARV prescribing patterns, were not specifically controlled for during the two study periods.

The authors have nothing to disclose.