American Association for Clinical Chemistry
Better health through laboratory medicine
NACB - Scientific Shorts
NACB - Scientific Shorts (formerly NACB Blog)
By Corinne Fantz, PhD, DABCC, FACB
PermalinkRecommendFacebookEmailComments

The use of bar codes for identification purposes has made many health care processes more reliable and efficient. However, while the current laboratory standard is Code 128, it is certainly not the safest and most reliable symbology in use today.  Malfunctioning bar code printers, improperly sized wristband bar codes, and less than optimal bar code orientation and width have all been shown to be sources of bar code-related, patient mis-identifications (1).

 

In contrast to linear bar code symbologies which can be encoded with only one patient identifier, newer bar code symbologies, including 2D and matrix type bar codes, may be encoded with multiple patient identifiers, and contain more sophisticated checking algorithms such as error correction.  Furthermore, because redundant data can be embedded in these more sophisticated bar codes, up to 30% of the bar code can be destroyed and still produce accurate data when scanned.  These symbologies are already in widespread use in a number of industries including banks, the pharmaceutical industry and mail delivery services. 

 

Check bar coding processes for potential sources of errors

Probably the most effective solution to check the integrity of bar codes before releasing them into the patient care environment is to use commercially available verification devices.  These devices can be used to verify the adequacy of bar codes before placing them into service.  Bar code verifiers can be a good investment, especially if the user has frequent printer failures.  If there is flexibility in the bar code printer software, bar codes should be oriented such that printer failures generate white lines perpendicular to the bars of the linear bar code, preserving the integrity of the bar code data.  Neither of these solutions will prevent post-printing wear and tear that can lead to errors over time.  Required manual checks have been made policy to detect post-printing damage, particularly in high-risk healthcare environments, such as blood bank specimen labeling, but these checks are also not full proof (2), suggesting the real need for more reliable identification technologies and/or redundant data processing.

 

Adopt strategies to improve accuracy of bar coding process

Implementation teams should seek to include representatives from all areas intending to use bar codes, to garner the needs of all the various users.  For example, groups may have very different size constraints (space on tube or wristband). The team should also become familiar with important bar code specifications, making sure that any generated bar code meets minimum bar code scanner requirements, and this is true for all scanners throughout the healthcare system.  From a technological perspective, if there is only one type of bar code in your system, the scanners should be limited to read only that bar code symbology.  Procedures to maintain and monitor printer performance and detect malfunctions are also important to prevent errors.  Are operators notified when printers are in need of maintenance?  Are there procedures to prevent bar codes from being released to a patient care areas when printer maintenance has not been performed?

 

What promise does the future hold?

Ultimately, health care providers should seek to transition to alternative higher fidelity identification technologies.  However, without standards set for these newer technologies, there will likely be issues with compatibility of the scanners, instruments, and laboratory information systems, making implementation of these technologies complicated.  Use of 2D or matrix bar codes, for example, will require compatible scanners and software to read the more complex bar code symbols and multiple encoded data (as opposed to just one encoded identifier with 1D bar codes). 

 

Other technologies that may be used alone or in combination with bar codes for tracking and identification purposes such as Radio Frequency Identification Devices (RFID) will require an even more thorough investigation prior to implementation.  RFID, in addition to requiring special scanners, require linker software to connect the RFID tag ID to the patient.  RFID broadcasts information and this can be a double-edged sword.  While the broadcast information helps the user locate a RFID encoded specimen, the information may also be accessible to anyone else who happens to have the right equipment to read it.  Therefore, early adopters must also ensure that necessary software is Health Insurance Portability and Accountability Act (HIPAA) compliant (often this software is not)(3).

 

Laboratorians should lead the effort to develop standards for implementing new technologies for patient identification and work with manufacturers to ensure that these new technologies serve to further improve the safety of patient identification across the health care industry.

 

 

 

Please log in to leave a comment!

Posted by
On 3/10/2011

I also support Dr. Fantz's article. As a co-author on the related paper in Clinical Chemistry and as Director of Pathology Informatics at our institution, the issue of safe implementation of barcode technology has come up outside of the issues that resulted in our publication. Institutions which are planning to use the same barcode technology for different data elements (e.g. accession number vs. medical record number vs. visit number) may run into trouble with linear barcodes because generally these barcodes are placed on items with limited space. Given the low amount of data per unit area that linear barcodes can hold, this generally means that only one identifier can be used in these barcodes. This is interesting because in all other areas of healthcare that I can think of (and this includes interfaces between data systems), multiple identifier matches are required to ensure proper patient identification. Why should barcodes be an exception to this rule? Checking two identifiers dramatically reduces the possibility of misidentification. Furthermore, it is clear that barcodes will be scuffed, marked and/or otherwise damaged with normal wear and tear, regardless of where they are applied, and printers may also acquire subtle changes without warning. Therefore, additional safeguards should be in place to correct for this damage where safe and possible, alert the user that the barcode and/or printer is not operating effectively in understandable terms, and to significantly increase the layers of data integrity checks to ensure that the scanned data is accurate. I look forward to the CLSI review of this technology. This comment was approved by the NACBLOG editorial board. Please remember to add your name and affiliation!

Posted by
On 2/22/2011

I support Dr. Fantz' recommendations and conclusions that the present laboratory standard for bar codes, Code 128, may be prone to errors, and stated as much in the editorial I wrote in the same issue of Clinical Chemistry that accompanied Dr. Fantz' article. This issue of high error rates is especially true if linear bar codes are not used with a separate bar code quality assurance program to assure the cleanliness and performance of the print heads on the bar code label printers, verify the quality of the resulting bar codes, and other steps noted by Dr. Fantz in her blog and in her original article and my editorial respectively. As a result of this issue and the attention that has been called to bar code errors by Dr. Fantz, I asked the CLSI Area Committee (now called a Consensus Committee) on Automation and Informatics to discuss the development and publication of a prospective standard on two dimensional or matrix bar codes for use in clinical laboratories. That recommendation is on the agenda for discussion at the March 29 meeting of that committee. A big hurdle for the adoption of such a standard by CLSI is that most of the analyzers and other automated systems in use in laboratories today can only read linear bar codes. Thus, it could take considerable time for the vendors to begin installing the more advanced systems required to read two dimensional codes. Retrofitting these newer readers in existing equipment may not be possible. It could be many years to accomplish a transformation in our industry from linear bar codes to two dimensional codes. However, we have to start somewhere and a committee convened by CLSI to consider such a standard is the best place to start. Charles Hawker, PhD ARUP Labs, Salt Lake City, UT

About the Author
Corinne Fantz, PhD, DABCC, FACB
Corinne Fantz, PhD, DABCC, FACB 
 

Additional Resources

Mountain PJ, Callaghan JV, Chou D, Davis RR, Hawker CD, Knafel AJ, et al. Laboratory automation: bar codes for specimen container identification; approved standard. 2nd ed. CLSI document AUTO2-A2. Vol. 25. Wayne (PA): CLSI; 2005.

 

Palmer RC. The bar code book: reading, printing, and specification of bar code symbols. 2nd ed. Peterborough, NH, Helmers Publishing; 1991.

 

Hawker CD. Bar codes may have poorer error rates than commonly believed. Clin Chem. 2010 Oct;56(10):1513-4.

 

References

1. Snyder ML, Carter A, Jenkins K, Fantz CR.  Patient Mis-identifications caused by errors in standard bar code technology.  Clin Chem. 2010 Oct;56(10):1554-60. Epub 2010 Aug 11.

 

2. Ansari S, Szallasi A. “Wrong blood in tube”” solutions for a persistent problem. Vox Sang. 2010 Aug 25. [Epub ahead of print]

 

3. Sotto LS. An RFID code of conduct. RFID Journal 2005.