ASHP Foundation announces research results

At the midyear meeting, several researchers spoke about their Foundation-funded research projects. Here are preliminary details of two of them.

• Medication errors and adverse drug events in the ICU. In this study, a researcher found that a direct observation approach to error detection may catch different rates and types of adverse drug events (ADEs), particularly potential ADEs, than chart reviews and voluntary/solicited incident reports.

First, institutional review board approval and informed consent were obtained. Then specialty residents and critical care pharmacists used the direct observation approach to record all activities related to the entire medication use process that might involve medication errors and ADEs in two studies, involving one adult and one pediatric intensive care unit (ICU). The observation was straightforward and not disguised.

Each study had a pilot followed by four four-day observation phases. A nurse evaluator was available for resolving potential disagreements in classification.

In the adult trial, in which patients took a median of eight scheduled and four as-needed medications, observers recorded 185 incidents. Of these 185, 132 were clinically important medical errors, 110 of which were potential ADEs (all of which were considered preventable), and 22 were actual (preventable) ADEs. This is a higher ratio of potential ADEs to actual (preventable) ADEs than reported by other research that used chart review and solicited incident reporting techniques, says Brian L. Erstad, PharmD, professor at the University of Arizona College of Pharmacy in Tucson. However, the stage at which most of the preventable ADEs occurred — the prescription and transcription stages — was consistent with previous investigations.

The observation approach catches potential errors that would have been missed unless they were directly viewed, Erstad said in a presentation of the results. Such an approach, however, would not be possible without the support of the hospital administration.

Fewer ADEs were identified in the pediatric trial, Erstad says. Both trials also revealed system problems. The full details about the study design and results will be published at a later time in the journal Critical Care Medicine.

• SODid (Solid Oral Dose Identification) Study: Current Practices of Pharmacists and Physicians. This study tested the ability of pharmacists and physicians to identify three common medications. The results underscored the need for a standardized imprint coding system.

Solid oral dose forms aren’t always easily identifiable once they are taken out of their original container. In his presentation of this study’s results, Andrew C. Seger, RPh, PharmD, showed an example of how patients sometimes turn to the Internet to try to identify an unknown medication. Along with his colleagues, Seger, a senior research pharmacist at the department of general medicine at Brigham and Women’s Hospital in Boston, wanted to see how successful pharmacists and physicians are when encountering an unidentified solid oral dose form. The study also looked at current resources available to help with such identification.

The researchers randomly selected 50 pharmacists and 50 physicians to test their ability to identify three medications: simvastatin (Zocor), a top-10 branded medication; lorazepam, a generic white tablet; and naproxen sodium (Aleve), an over-the-counter medication. These drugs are commonly prescribed and used in both markets where the study was conducted. The researchers also surveyed participants regarding their views and experiences on current identification resources used in everyday practice and potential alternatives. Both the observation and survey pieces of the study were administered in the practitioner’s natural setting.

The physicians and pharmacists correctly identified the oral dose form in 190 (63%) of 300 observations. Only 24 pharmacists (48% of all pharmacists) and 18 physicians (36% of all physicians) correctly identified all three of the medications, while five pharmacists (10%) and 10 physicians (20%) identified none. Simvastatin was correctly identified in 77%, lorazepam in 63%, and naproxen sodium in 51% of the observations. The average identification time was 3.65 minutes.

The resources used most often were various electronic sources for the pharmacists (52%) and the print version of Physician’s Desk Reference (PDR) for the physicians (46%). Overall, 77% of the practitioners expressed dissatisfaction with the current system, and 91% favored a more standardized approach.

The full study results and discussion will be published at a later time.