From: Bruce L. Lambert, Ph.D. [lambertb@uic.edu]
Sent: Friday, July 12, 2002 5:11 PM
To: fdadockets@oc.fda.gov
Subject: Comments on Bar Codes [Docket No. 02N-0204]
To Whom it May Concern:

Generally speaking, I am in favor of the use of bar codes on FDA-regulated medications, blood products, and devices. I believe that bar codes have the potential to reduce several types of medication errors. However, they are not a panacea and the FDA should be well aware of their limitations:

1. Bar codes will not prevent doctors from prescribing the wrong medication as the result of recall errors involving similar drug names. Bar codes can only prevent or catch transcription, dispensing, and administration errors.

2. Bar code technology, like any technology, must be integrated into a complex social system, and this process of integration can be highly problematic, often causing as many problems as it solves.

See: 

Patterson, E. S., Coehlo, D. A., Woods, D. D., Cook, R. I., & Render, M. L. (2000, May 26-28). The natural history of technology change: How introducing bar coding changes medication administration. Paper presented at the Fifth Conference on Naturalistic Decision Making, Tammsvik, Sweden.

3. Bar code readers are not perfect, and studies of bar code readers in realistic situations (e.g., grocery stores) reveal high error rates. For example, this is from the web site of a bar code label manufacturer:

"Accuracy
Bar coding increases accuracy by reducing the likelihood of human errors from manual entry or miscommunication from misread or mislabeled items. Bear in mind that print quality is of utmost importance to ensure that the scanner has the ability to read the bar code. Without this capability, the accuracy and productivity gains of bar code technology are virtually non-existent." 

Also keep in mind that, due to the very high volume of prescriptions dispensed in the US annually, even very small error rates translate into fairly large absolute numbers. For example, assuming 4 billion prescriptions are 
dispensed annually in the US, a 1 in 1 million error rate will still produce 4,000 errors.

See also:

http://www.ftc.gov/opa/1998/9812/scanner.htm
http://www.ftc.gov/reports/scanner2/scanner2.htm
http://www.ftc.gov/bcp/conline/pubs/products/scanffc.htm

Although I recognize that grocery stores are different from hospitals and pharmacies, the high error rates in real-world use of bar code scanners in groceries gives some insight into what could happen when the theoretical error rates (1 in a million) are put to the test in real world settings.

Respectfully Submitted,


Bruce L. Lambert, PhD
Department of Pharmacy Administration
University of Illinois at Chicago
833 S. Wood St. (M/C 871)
Chicago, IL 60612-7231

phone: 312-996-2411
fax: 312-996-0868