View source for Electronic medical record, error detection, and error reduction: A pediatric critical care perspective

[[Electronic medical record, error detection, and error reduction: A pediatric critical care perspective]]

'''Brian Jacobs, MD'''

Pediatrc Crit Care Med 2007 Vol. 8, No.2 (Suppl)

Medication errors have been a major concern for health care organizations (HCOs) for the last decade. The Institute of Medicine (IOM) has recommended the use of automated systems for order processing and the elimination of handwritten clinical information by the end of this decade.   For this reason, HCOs has emphasized the adoption of  electronic medical records (EMRs) and computerized provider order entry (CPOE)s.  This has been considered to be an important approach toward improving the safety of health care not only of adults but also children (1). Consequently the implementation of EHRs has changed the way that medicine is practiced today.  In this article, the author (Dr. Jacobs) reviews some studies that have used computerized data extraction from the electronic medical record (automated event detection) with a particular emphasis of this method in pediatric practice. 

One area of particular concern in this article is the detection adverse drug events (ADE) due to the serious nature of these errors. The institute of Medicine has defined ADE as “injuries resulting from a medical intervention related to a drug “.(2)  Considering that ADE  represent 19% of all adverse events several approaches has been used in the detection and reduction of such errors.	

Dr. Jacob states that the implementation of EMRs with CPOE has been shown to prevent errors by generating complete, legible, unambiguous orders and electronic medication administration recorded together with the prescribers complete identification information.  Also, the enhancement of CPOE with clinical decision support systems (CDSS) has allowed for weight base dosing and dose range checking which is particularly important in the pediatric population.   This enhancement provides the opportunity for integrated information at the point of care, without interruption in the prescriber workflow, minimizing haphazard clinical decision situations such as overdosing. 

The author assesses the relationship of CPOE implementation and medical errors by examining 3 different studies.  Two of the studies were performed in pediatric institutions. While the study performed by Upperman et al. (3) showed an association of CPOE implementation with avoidance of ADE, Potts et al. (4) suggested the contrary.  Even though the design in this last study have been criticized, the author point out that is important not to over relied on technology which could perpetuate errors, if  the process is not carefully evaluated before, during after the implementation. (5) The author also reviews areas that could represent  a potential for safety deterioration after a CPOE implementation such as accidental selection of wrong patient for the entry of orders, alert fatigue, deterioration of physician-nurse communication, unrecognized discontinuation of important medication, incorrect entry by nurses of verbal orders from physician and unilateral  selection of the wrong medication(6).

Health care organizations have used several methods to detect errors. The author points out that in the past, they had mostly relied on voluntary event reporting by clinicians; however these tend to underestimate the actual adverse event by around 20-fold (7). Manual charting, which is another method to detect medical errors require specific triggers such as panic values, the transfer to a higher level of acuity, or the use of antidotes, to be documented.  This method, though very comprehensive, is very labor intensive and expensive, and has the possibility of ending up with sample bias.  For this reason HCOs have been using automated event detections to prevent errors for more than 20 years.  The author, however, suggests caution in the implementation of this method.  Institutions have to understand that for this method to be effective, the data in the information system needs to be complete and consistently entered to all the areas which provide with source of data in the EMRs, not only CPOE.  He also indicates that once these events are detected, the statistical significance needs to be analyzed.  
	
The author reviews the role of EMRs in automated adverse event detection through studies that involve medication errors, nosocomial infections and perioperative settings.  He presents two studies that used this method for medication error detection, paying particular attention to the one of Boiling et al. where administration of a reversal agent, naloxone, was used as a specific trigger to detect opiods overdose in children.  He also presented the studies by Evans et al. where the automated adverse event detection system detected 90% of the nosocomial infections, compared to the 76% detection rate of manual surveillance(8).  The use of automated adverse event detection methods has been also documented in procedures requiring anesthesia.  The author of this study reported that the automated system was able to detect 18.7% of adverse event compared to the 5.7% manual detection rate reported by anesthesiologists (9).  Dr. Jacob also presents a preliminary report by his group regarding the use of pager notification of CPOE –generated STAT (at once) orders for radiology and respiratory treatment for children admitted in his hospital.  He implies that their result showed a faster response for services when the CPOE is linked to pager notification.    

Dr. Jacobs does an excellent job in pointing out the benefits of implementing EMR with CPOE in an effort to reduce errors using computerized base systems to detected potential areas where manual detection can underestimate the errors.  Even though the author’s intention was to emphasize medical errors for the pediatric population, the article fell short of the topic, not intentionally, but for the lack of research in the area.  Most of the research until now has been directed towards intuitions dedicated to adult medicine which can not compare to the needs of the institutions that are dedicated solely to pediatric medicine.  Hopefully in the future, clinicians in the area of pediatrics will have the opportunity to not only receive specific benefits of automated event detection in the prevention of errors, such as alerts and reminders in medications, but in detecting nosocomial infection and procedure requiring anesthesia as well. 

'''References'''

1.Wang JK, Herzog NS, Kaushal R, et al. Prevention of pediatric medication errors by hospital pharmacists and the potential benefit of     computerized physician order entry. Pediatrics 2007; 119:e77-e85.

2.Institute of Medicine. Preventing medication errors. Washington DC National Academy of Science. Press 2006.  

3.   Upper JS, Stanley P, Friend K et al: The impact of a hospital wide computerized physician order entry on medical errors in a pediatric hospital.  J Pediatr Surg 2005; 40:57-9.

4.Potts AL, Barr FE, Greory DF, et al: Computerized physician order entry and medication errors in a pediatric critical care unit.     Pediatrics  2004;113: 59-63.

5.Seely CE, Nicewander D, Page R, et al: A baseline study of medication error rates at Baylor University Medical Center in preparation for      implementation of a computerized physician order entry.  Proc (Bayl Univ Med Cent) 2004; 17:357-362.

6.Koppel R, Metlay JP, Cohen A, et al: Role of computerized physician order entry systems in facilitating medication errors.  JAMA 2005; 29:      336-344.

7.Edlavitch SA: Adverse drug event reporting: Improving the low US reporting rates.  Arch Intern Med 1988; 148:1499–1503.

8.Evans RS, Larsen RA, Burke JP, et al: Computer surveillance of hospital-acquired infections and antibiotic use.  JAMA 1986; 256: 1007–1011.

9.Benson M, Junger A, Fuchs C, et al: Using an anesthesia information management system to prove a deficit in voluntary reporting of adverse     events in a quality assurance program.  J Clin Monit Comput 2000; 16: 211–217.

Elisdel M. Garcia-Bousquet

10-19-2007

[[Category: BMI-512-F-07]]