Computer Entry for Scripts May Halve Medication Error Rates

Hospitals that process prescription orders through a computerized provider order entry (CPOE) system likely reduced their medication errors by 48% in 2008¬, according to a new meta-analysis. Given the adoption rate of the new technology by American hospitals, that translates to an estimated 17.4 million medication errors averted that year.

David C. Radley, PhD, MPH, from the Institute for Healthcare Improvement, Cambridge, Massachusetts, and colleagues published the results of their study online February 20 in the Journal of the American Medical Informatics Association.

The Institute of Medicine estimated in 2007 that the average hospital patient is subject to 1 medication error each day. Dr. Radley and colleagues sought to provide a baseline national estimate of how many medication errors were averted when hospitals adopted CPOE. Rather than hand writing the name of a medication, which can be misinterpreted, CPOE allows clinicians to choose the medicine and dose from drop-down menus aided by automatic prompts that flag potential harmful drug interactions. The researchers focused on 2008, which is the year before Congress provided $20 billion to help build a national health information technology infrastructure aimed at improving safety and health care quality.

Dr. Radley and colleagues performed a meta-analysis of published studies to estimate the benefit of CPOE systems. They calculated that CPOE use dropped medication error rates by approximately 48%.

Using data from the 2006 American Society of Health-System Pharmacists and the 2007 American Hospital Association Annual Survey, the researchers estimated that hospitals placed nearly 1.76 billion medication orders in 2008. Extrapolating from the American Hospital Association surveys, Dr. Radley and colleagues found that approximately 26.1% of those orders in acute care hospitals were processed using CPOE.

"At the rate of CPOE adoption and implementation in 2008, our findings suggest that medication errors were reduced by ~12.5% (bounds 10.6–14.4%). This equates to ~17.4 million (bounds 0.09–27.1 million) fewer medication errors over a 1-year period than would be expected without CPOE," the authors write.

If more hospitals adopted the computerized technology, the number of averted medical errors would be even more dramatic, they note. They add, however, that it remains unclear whether reducing medical errors also reduces patient harm from medications.

Study limitations include methodological inconsistencies in the studies, with variances in medication error definitions and detection. In addition, the hospitals included in the studies were disproportionately large, urban, academic medical centers; medication error rates there may vary from the rest of the nation's hospitals. Finally, the research team assumed that the effect of CPOE adoption on medication errors is constant as adoption rates increase.

"Our rigorously developed meta-analytic estimate is in keeping with the earlier, evidence-based heuristic of Bates et al, and, more importantly, is a much-needed addition to our knowledge of the effect of CPOE on medication errors," the authors conclude.

Support for this study was provided by the Agency for Healthcare Research and Quality. The authors have disclosed no relevant financial relationships.

J Am Med Inform Assoc. Published online February 20, 2013. Full text