COVID-19 Impact

You may see some delays in posting new content due to COVID-19. If you have any questions, please submit a message to PSNet Support.
 

PSNet: Patient Safety Network

 Cases & Commentaries

A Troubling Amine

Commentary By Elizabeth A. Flynn, PhD

The Case

A 43-year-old woman was admitted to the intensive care unit for symptoms of heart and respiratory failure. She was found to have severe mitral and tricuspid valve regurgitation. She responded well to medical therapy, and surgical valve repair was scheduled. During her initial evaluation, a jaw fracture was incidentally noted. Given the jaw fracture and her valvular disease, an oromaxillofacial surgeon recommended prophylactic antibiotic coverage prior to surgery. Penicillin, 500 mg orally four times daily, was ordered. On the second day of antibiotics, when the nurse compared the drug with the medication administration record (MAR), she noticed that the patient was receiving penicillamine (a non-antibiotic medication used in the treatment of Wilson’s disease and severe rheumatoid arthritis) instead of penicillin and alerted the pharmacy.

A pharmacist reviewed the original handwritten order and saw that penicillin was clearly prescribed. The pharmacist who entered the order into the pharmacy computer system had typed in the code “PENIC” and had received a drop-down box that displayed all formulations and dosages of both penicillin and penicillamine. That pharmacist had incorrectly selected penicillamine as the drug to be given. The final check of the medication (at the time the drug left the pharmacy) compared the drug product against the information in the pharmacy computer system but not against the original handwritten order. The patient suffered no ill effects from the error and received the course of penicillin as originally prescribed.

The Commentary

Penicillamine and penicillin are listed as look-alike/sound-alike drugs by the United States Pharmacopeia, along with approximately 1,100 other drug pairs.(1) The literature describes an incident similar to this case, in which a general practitioner wanted to prescribe penicillin V, 250 mg, but selected penicillamine, 250 mg instead. That case was a near miss—a pharmacist detected the error and intervened.(2) Similar incidents related to incorrect selection of a medication in a computer system have been reported.(3,4)

This case raises the issue of how well pharmacy computer systems aid in the prevention of medication selection errors. In one study, the pharmacy technician order entry error rate was measured at 2.5%. Most of these were missed orders, but 20 errors out of 246 involved incorrect medication selection.(5) In a study in 50 community pharmacies, order entry errors were responsible for 48 of 63 (76%) dispensing errors on new prescriptions.(6)

Pharmacy computer systems have been used to improve the medication system for decades. Thirty years ago, Means and colleagues (7) used direct observation to study the effect of a pharmacy computer system on medication administration errors detected but found little impact. This may be because early pharmacy computer systems were developed with a focus on operational, financial, and data management functions, rather than on preventing medication errors.(8) Modern pharmacy computer systems include order entry work flow, medication database information reliability, availability of enhanced drug name lettering, order verification method, clinical decision support (eg, drug interaction screening), and format of the MAR generated for the nurses, all of which may improve medication safety.(8) The Institute for Safe Medication Practices (ISMP) conducted a survey of safety features with pharmacy computer systems in hospitals in 1999 and 2005. The survey included a series of test orders with problems such as therapeutic duplication, drug-drug interactions, and overdosage based on age or body surface area. In the 2005 survey, only 4 of the 182 systems detected all of the unsafe orders, indicating little improvement since 1999 in systems’ ability to detect potential medication errors.(9)

Some, but not all, systems include advanced decision support such as dose-range checking, drug-laboratory results checking, and comparison to diagnoses. The value of linking pharmacy and laboratory data in order to reduce errors was well-described by Schiff and colleagues.(10) In the latest ISMP survey, 72% of respondents stated that their systems were linked to the laboratory systems, but only 42% alerted staff based on drug-related problems due to lab results.(11) Issues with noise or “alert fatigue,” can be dealt with by setting the alert sensitivity level at a point in which only the events with the potential for the most serious harm would appear.(12) ISMP reported that 71% of modern pharmacy systems generate screen alerts that have little or no clinical significance, but 60% of these alerts can be eliminated easily (ie, removing them with a single click).(11)

What can be done to improve the accuracy of medication selection? A study of the use of tall man lettering (eg, vinCRIStine and vinBLAStine) found that it was helpful in improving the accuracy of the selection of drugs that can be confused (sound-alikes and look-alikes).(13) The ISMP survey found that 57% of responding hospitals had computer systems that allowed the use of tall man lettering.(11) ISMP recommends tall man lettering for 33 medications, but penicillamine and penicillin are not currently included on this list.(14) Providing a link between the drug name and indication as a double check of correct medication selection likely would have prevented the error in this case, because penicillamine is used for the treatment of rheumatoid arthritis and as a chelating agent in heavy metal poisoning, but would not have been a recommended medication to treat an infection. A report from a hospital in Taiwan demonstrated that the addition of a link between certain drug names and indications for antiretrovirals and insulin secretagogues successfully prevented wrong drug errors.(15) An expert panel recommended displaying a list of medications indicated for a diagnosis on entry of the diagnosis. However, this feature has been recommended for prescriber order entry systems, not pharmacy systems, but it should be explored for pharmacy systems as well.(16) At this time, I am not aware of such diagnoses-medication links in commercially available pharmacy computer systems.

The use of enhanced lettering and links between drug name and indication will not prevent all errors associated with information systems, however. Research applying cognitive science and usability engineering methods (17) to identify design features that enhance order entry accuracy are needed, particularly for pharmacy systems. In the future, pharmacy systems may respond to the penicillamine order entered in this case with an alert such as “This surgeon has never prescribed penicillamine, but frequently orders penicillin. Are you sure you want to enter penicillamine?”

Take-Home Points

  • Pharmacists inspecting filled orders should use the original order as the basis for comparison and take the opportunity to verify correct entry of the order into the computer system.(18)
  • Verify information at each step in the process. This case illustrates the value of including double checks by everyone using the medication system–the nurse detected the error by doing a routine medication check.
  • Incorporate enhanced lettering in computer drug lists if not already available.
  • Physicians can help improve patient safety by providing an indication for each medication prescribed. Pharmacists assigned to enter orders into computers may not be familiar with the patient involved and appreciate having the indication available as they process the order.

Elizabeth A. Flynn, PhD Associate Research Professor Auburn University Harrison School of Pharmacy

References

1. USP Center for the Advancement of Patient Safety. Use caution–avoid confusion. USP QualityReview. April 2004. Available at: http://www.usp.org/pdf/EN/patientSafety/qr792004-04-01.pdf. Accessed September 12, 2006.

2. White A, Russ C. Recent prescribing near misses in practices. Bolton, England: Bolton NHS Primary Care Trust; September 6, 2005. Available at: http://www.bolton.nhs.uk/clinical/med_manage/circletters/formulary%20letter.doc. Accessed September 12, 2006.

3. Ash JS, Berg M, Coiera E. Some unintended consequences of information technology in health care: the nature of patient care information system-related errors. J Am Med Inform Assoc. 2004;11:104-112. [go to PubMed]

4. Kushniruk A, Triola M, Stein B, Borycki E, Kannry J. The relationship of usability to medical error: an evaluation of errors associated with usability problems in the use of a handheld application for prescribing medications. Medinfo. 2004;11:1073-1076. [go to PubMed]

5. Tierney M, McLurg D, Macmillan C. Transferring medication order entry from pharmacists to pharmacy technicians. Can J Hosp Pharm. 1999;52:240-243.

6. Flynn EA, Barker KN, Carnahan BJ. National observational study of prescription dispensing accuracy and safety in 50 pharmacies. J Am Pharm Assoc (Wash). 2003;43:191-200. [go to PubMed]

7. Means BJ, Derewicz HJ, Lamy PP. Medication errors in a multidose and a computer-based unit dose drug distribution system. Am J Hosp Pharm. 1975;32:186-191. [go to PubMed]

8. Daniels R. Pharmacy information systems. In: Brown TR, ed. Handbook of Institutional Pharmacy Practice. 4th ed. Bethesda, MD: American Society of Health-System Pharmacists; 2006:313-327.

9. Safety still compromised by computer weaknesses. ISMP Medication Safety Alert! Acute Care Edition. April 25, 2005. Available at: http://www.ismp.org/Newsletters/acutecare/articles/20050825.asp

10. Schiff GD, Klass D, Peterson J, Shah G, Bates DW. Linking laboratory and pharmacy: opportunities for reducing errors and improving care. Arch Intern Med. 2003;163:893-900. [go to PubMed]

11. Institute for Safe Medication Practices. National survey of safety features with pharmacy computer systems. 2005. Available at: http://www.ismp.org/survey/survey200505R.asp. Accessed September 12, 2006.

12. Stiening KK, Trimble JM, Merryfield DW. Computerized clinical alerts--reducing the noise. Abstract of meeting presentation at: American Society of Health-System Pharmacists Midyear Clinical Meeting; December 7-11, 2003; New Orleans, LA.

13. Filik R, Purdy K, Gale A, Gerrett D. Labeling of medicines and patient safety: evaluating methods of reducing drug name confusion. Hum Factors. 2006;48:39-47. [go to PubMed]

14. Name differentiation project. US Food and Drug Administration, Center for Drug Evaluation and Research Web site. Available at: http://www.fda.gov/cder/drug/MedErrors/nameDiff.htm. Accessed August 18, 2006.

15. Yu KH, Su SS, Kuo CC, Tsao HL. Reducing medication errors by linking drug names with patients' diagnostic codes. Am J Health Syst Pharm. 2006;63:808-809. [go to PubMed]

16. Bell DS, Marken RS, Meili RC, et al. Recommendations for comparing electronic prescribing systems: Results of an expert consensus process. Health Aff (Millwood). 2004; Suppl Web Exclusives:W4-305-317. [go to PubMed]

17. Kushniruk AW, Patel VL. Cognitive and usability engineering methods for the evaluation of clinical information systems. J Biomed Inform. 2004;37:56-76. [go to PubMed]

18. Kuyper AR. Patient counseling detects prescription errors. Hosp Pharm. 1993;28:1180-1181, 1184-1189. [go to PubMed]