For decades, clinicians have used handwritten or verbal orders for patient care, and unit secretaries, nurses and other medical personal have transcribed them. However, as the Institute of Medicine's (IOM) 1999 report “To Err Is Human” noted, traditional processes leave room for misinterpretation and error. In addition, recent governmental regulatory requirements have emphasized a paperless system that uses computerized provider order entry (CPOE) and electronic health records (EHRs).
CPOE has a promising, yet unproven, future as a tool for improving overall patient care and safety. Currently, the strongest evidence supports CPOE for improvement of cost savings and health care efficiency and reduction in medication errors, while little research has been done to show CPOE reduces mortality. A relatively small proportion of hospitals uses CPOE, however, so its full benefit may still be unknown. According to the Agency for Healthcare Research and Quality at the U.S. Department of Health and Human Services, four percent of hospitals nationwide have a CPOE system. Even in those that have adopted CPOE, less than half of physicians enter at least 80% of their orders electronically.
We believe that as hospitals begin to implement and enforce use of CPOE, more research and investigation will be done that will ultimately support its benefits. CPOE and clinical decision support (CDS) help clinicians and others to provide standardized and evidence-based practice, which we believe has significant impact on patient outcomes, including on mortality.
CPOE and errors
The IOM estimates that at least 1.5 million people in the U.S. are harmed annually by medications. A 2009 Studies in Health Technology and Informatics article estimated that nearly 400,000 adverse drug events (ADEs) occur in hospitals each year. ADEs also have a large financial cost, with Medicare paying about $882 million annually to treat medication errors, according to a 1995 Journal of the American Medical Association (JAMA) study.
One study, conducted at a 700-bed academic medical center in Chicago and published in Archives of Internal Medicine in 2004, found that more than 60% of prescribing errors involved incorrect medication dosing. Another large cross-sectional study conducted in Sweden and published in BMJ in 2001 found that 13.6% of 962,000 prescriptions distributed from pharmacies were written with at least one potential drug interaction. CPOE can help with these issues, as it allows physicians to write orders in a clear, standardized format and uses software to check for medication errors like drug allergies, drug dosing errors, and drug-drug interactions.
Indeed, research has shown that CPOE implementation helps reduce errors and their associated costs, while increasing efficiency. A 1998 JAMA study found implementation of a CPOE system reduced allergy-related drug reactions by 56%. (Drug allergies represent the highest proportion of medication errors in inpatient settings.) Another study conducted in 2006 and published in the Journal of the American Medical Informatics Association examined the cost-benefit analysis of implementing a CPOE system for a 720-bed tertiary medical center, and found it would take $11.8 million to initiate and support a hospital-wide CPOE system, and eventually save the hospital $28.5 million over a nine-year period. A randomized trial conducted in 2008 and published by the Agency for Healthcare Research and Quality compared laboratory, radiology, and pharmacy turnaround time between a computer system with CPOE software and a control arm without the CPOE program and found that turnaround time with the CPOE system was significantly reduced.
CPOE and mortality
The effect of CPOE on patient mortality is controversial. CPOE systems have previously been associated with an increase or no change in hospital-wide mortality rates. For example, a 2005 Pediatrics study found an increase in mortality rates associated with CPOE implementation at The University of Pittsburgh School of Medicine, from 2.80% before CPOE implementation to 6.57% afterward. However, a more recent study in the July 2010 Pediatrics found that mortality rates decreased by 20% after implementation of a CPOE system at a 303-bed hospital at Stanford University School of Medicine's quaternary children's hospital.
Clearly, more research on the effects of CPOE on mortality in hospitalized patients is needed. Hospitals should continue to evaluate such effects, in addition to medication error rates for patients who are dependent on time-sensitive therapies. We believe CPOE technology holds great promise as a tool to reduce human error during health care delivery, and hospitalists will play a major role as leaders in CPOE implementation and research.