Can Computerized Physician Ordering Create Errors?
Koppel R, Metlay JP, Cohen A, et al. Role of computerized physician order entry systems in facilitating medication errors. JAMA. 2005;293:1197-203.
Adverse drug events are a frequent etiology of inpatient morbidity and prescribing errors are the most frequent source. Computerized physician order entry (CPOE) is touted as a potential remedy for some types of adverse drug events. Few studies have investigated the potential for novel medication errors generated by a change to CPOE from conventional ordering. Koppel et al. present a quantitative and qualitative study of medication errors caused or exacerbated by a CPOE system. Interviews, surveys, and focus groups were the primary means of data collection. Housestaff who typically enter more than 9 orders per month were the primary study population, but data collection also included pharmacists, nursing staff , information technology managers, and attending physicians. The study was conducted in a tertiary-care teaching hospital between 2002 and 2004 utilizing a CPOE system in place since 1997. The CPOE system utilized is described as “monochromatic” and having “pre-Windows interfaces.” While not integrated with all hospital functions, the system was integrated with pharmacy and nursing medication lists. Researchers grouped errors into two broad categories: (1) information errors (fragmentation and systems integration failure) and (2) human-machine interface flaws (machine rules that do not correspond to work organization or usual behaviors). In total, 22 types of errors were recorded.
An example of an “information error” is assumed and incorrect dose information based on viewing doses intended only to describe pharmacy stocking practice―i.e. assuming that because the pharmacy stocks a 10mg dose of a medication, 10mg is an appropriate “minimally effective” dose. A “human-machine interface error” example is selecting an incorrect patient for ordering due to properties of the CPOE screen, such as the patient name not appearing on all screens. There are several important limitations to this study, but perhaps most important is the inability to generalize this data to other settings with potentially different physician users and software. Also important is a lack of description regarding physician user training and/or correlation of errors with amount of training or frequency of use, considering that the study population was defined as housestaff who may only use the system for 9 orders each . Despite these limitations, the study represents a requisite component to the growing trend toward the complete electronic record―namely, the use of objective investigations to study the safety and effectiveness of CPOE and the electronic record to promote the most optimal implementation and evolution of this new clinical tool.
Single-Dose Azithromycin for Acute Otitis Media
Arguedas A, Emparanza P, Schwartz RH, et al. A randomized, multicenter, double blind, double dummy trial of single dose azithromycin versus high dose amoxicillin for treatment of uncomplicated acute otitis media. Pediatr Infect Dis J. 2005;24: 153-61.
Acute otitis media is a common comorbid condition in pediatric inpatients. Patients at risk of having AOM with drug-resistant Streptococcus pneumoniae can be treated with high-dose amoxicillin as a first-line therapy according to recent American Academy of Pediatrics (AAP) recommendations. Despite this recommendation, there is evidence of reduced in vitro activity of amoxicillin against β-lactamase-producing Haemophilus influenzae and Moraxella catarrhalis, as well as a lack of data from controlled and blinded studies demonstrating efficacy, adverse events and compliance for high-dose regimens. Azithromycin has in vitro activity against the 4 pathogens of clinical significance in AOM, and studies have shown that a single-dose regimen of azithromycin by the oral route is pharmacokinetically feasible, safe, and comparable in success rate to 3- and 5-day azithromycin regimens. With these considerations in mind, Arguedas et al. designed this study to compare single-dose (30 mg/kg) azithromycin with high-dose (90 mg/kg/day) amoxicillin in uncomplicated AOM.