Printer Friendly

Joining the bar code revolution.

THE COST OF LABOR accounts for more than half of all operating expenses in today's clinical laboratory. Instituting bar coding is one way to reduce labor costs while achieving high levels of productivity, accuracy, and efficiency.

Once an obscure data collection technique, bar coding has emerged as the number one alternative to manual data entry in high-volume transaction processing. It makes life easier for lab staff; many hospitals have found that bar coding actually boosts employee morale. Laboratorians derive satisfaction from knowing they are working with leading-edge technology.

One time-honored goal of clinical laboratory testing is to identify specimens positively and permanently. The ideal specimen has an unmistakable ID from the moment of acquisition to receipt of the result report. Yet in most hospitals the ID is not foolproof. Even where bar codes are used, they are not always applied to the tube until the specimen is sent to the lab. This system is inefficient and greatly susceptible to human error.

At our large not-for-profit hospital, as with a growing number of other facilities, the bar code label that positively identifies the specimen throughout testing is affixed at the point of care. Our method virtually eliminates the misidentification of specimens, which risks errors at various points on the test path. Such errors often have grave consequences for patients.(1)

* Reading machines. It was the development of automatic analyzers that spurred an early proposal for averting specimen ID problems: a machine-readable identification system (MRIS). An MRIS developed in 1963 mated computer-style punch cards with automatic analyzers. That proposal died with the punch cards, according to Arthur E. Rappoport, M.D., the first chairman of the computer committee of the College of American Pathologists and a leading advocate of automatic specimen identification.

Optical character recognition (OCR), available since the early 1970s, uses a highly stylized type font (face) that can be read by both human operators and machines. OCR was plagued from the start by a relatively high error rate. An example is the frequent rescanning required in department stores, where cashiers often must resort to entering product numbers manually. Another disadvantage of OCR is that it requires a relatively high level of operator skill. Consequently, this computer-related system never succeeded in the clinical lab.

Another idea, pioneered in a Cincinnati supermarket in 1972, used a series of concentric rings read by scanners. Although bull's-eye symbols never caught on, related cost-benefit analyses and system refinements proved useful in creating the system that did: bar codes.

* Vertical lines. By 1985, it seemed inevitable that bar coding would undergo applications for the clinical laboratory.(2) Nevertheless, it took some years for the next step to come to pass. Lack of progress was being lamented in print as recently as 1989.(3)

What accounts for the spectacular rise of bar codes to ubiquity in the clinical laboratory over a brief period? For several reasons, these symbols represented an improvement over manual key entry and OCR. This article presents a brief history of the technology and a discussion of printer hardware.

* Punching keys. The widespread adoption of computers has provided many opportunities to automate data acquisition, storage, and retrieval.

At first, data were entered by manual keying. Then a special deck of cards was literally key-punched and read into the computer. This system has disadvantages.

One drawback is that neither technique is done in real time. For example, a time lag takes place between specimen identification and computer entry.

A second shortcoming is the human involvement required for each transaction. The error rate for manual keying is approximately one error per 300 characters. Many more opportunities for mistakes are introduced through the multiple steps of the process: recording the data on paper, gathering the paper records, and transcribing them via keyboard into the computer, and, when used, punch cards.

* Automatic entry. Automated data capture and entry technologies, including OCR and bull's-eye rings, were developed in an effort to address the drawbacks of manual systems. In automatic techniques, a single entry results in the capture of a data stream; scanning one bar code can call up an entire patient record. Although a human operator may be involved in the transaction, the actual data capture event is machine mediated.

Industrial bar code applications date back to the late 1960s. At that time, railroads in North America adopted a tracking system, developed by Sylvania, that used reflective red, white, and blue bars on the sides of railroad cars. Insufficient investment in training, maintenance, and equipment led the system to be abandoned in 1974.

* Supermarkets et al. Bar coding evolved quickly in the supermarket industry, where the need for automatic inventory control and checkout is great and financial backing for research is available. In 1973, after its abortive experiment with bull's-eye codes, the industry chose the Universal Product Code (UPC) for use in North American stores. UPC bar codes are used today on virtually every grocery item and consumer product.

Throughout the 1970s, bar coding grew in popularity as related hardware dropped in price. The technology was hampered at first by lacking a standard symbology (language). Equipment manufacturers cooperated to develop today's most commonly used bar code symbologies: Code 39, Code 128, Interleaved 2 of 5 (I 2 of 5), and Codabar.

In 1984 the Health Industry Bar Code Council (later named the Health Industry Business Communications Council) approved Code 39 as the standard. That paved the way for the introduction of bar coding in health care.

Many bar code applications in the clinical laboratory involve space constraints. Small items such as specimen tubes lack the "real estate" required for early versions of bar codes. Researchers eventually succeeded in reducing the amount of space needed for placement of a symbol.

High-density codes were introduced in 1987. As new standards for bar codes evolved, manufacturers recognized the need for refinements. Advances included more efficient reading devices and higher precision and greater uniformity in symbol printing. The use of bar codes in limited-space applications was hampered by the poor quality of the early printers that were used for point-of-care applications.

In Code 39, the first widely used standard, the number of characters available per inch was insufficient for wide-spread use in the clinical laboratory. The rigorous technical requirements demanded by the laboratory called for more compact (denser) symbologies. This change was needed to conserve precious space on labels that were to be attached to slides and small test tubes.

* Current technology. In five crucial ways, technology has brought us to the brink of a bar code revolution.

Analyzers can boast increasingly sophisticated bar code capability.

Printers can produce small, scannable codes.

Laboratory information systems provide bidirectional interfacing. By reading the bar code, the analyzer learns that the labeled specimen is the one on which to perform work downloaded from the LIS.

Workflow management is increasingly valued as a way to enhance productivity.

New technology is embraced more readily than ever by hospital administration and laboratory management as a way to streamline operations.

* Attributes. Other factors have added to the attractiveness of bar coding. The shrunken budgets of the 1990s force laboratorians to live with contracted resources and to use ingenuity in their quest for efficiency and productivity. Technologies such as bar coding, which provide easily demonstrable savings, have been welcomed.

Many health care institutions have implemented total quality management techniques emphasizing process improvements aimed at preventing errors. The reduction in manual steps allowed by bar coding helps achieve TQM.

The cost savings that accompany improved efficiency are welcomed by administrators. Thus the management climate in both hospitals and labs is ready for innovative approaches such as bar code specimen identification.

* Analyzer capabilities. Ideally, analyzers use the LIS-assigned accession number in a form and symbology that will support optimal performance. This is starting to be the case, but wasn't always. Manufacturers have been equipping their analyzers with increasingly sophisticated bar code readers, some of which can read more than one symbology. Certain instruments, such as the Coulter STKS (Coulter Electronics, Hialeah, Fla.) and Paramax (Baxter Healthcare, Irvine, Calif.), require a bar code to input the accession number.

At first, bar code labels were placed on specimen tubes in the lab just before the tube was loaded in the analyzer. The reason for this extra step was the lack of a good, wide-carriage printer that could produce collection lists side by side with matching bar codes capable of being applied at the point of collection. Now one and only one label, a bar-coded one, is put on the tube at the time of test order.

Another extra step arose when some vendors built analyzers that required the entry of a special ID--internal to the analyzer and not identical to the laboratory accession number--by keying or bar coding. This step has largely been eliminated.

* Technical difficulties. Fundamental shortcomings have limited the successful implementation of bar code technology in the clinical laboratory.

High-density codes. Dot matrix and other inexpensive printers can seldom maintain the near-photographic quality necessary for printing the codes on small medical labels. The ideal is greater than 200 dots per inch (dpi).

First-read rates. Printers have not always produced labels that achieved acceptable first-read rates (greater than 99%) on automated instruments.

Symbologies. Few printers have been capable of producing lines of labels bearing more than one symbology.

Appropriate software. Sophisticated bar code driver (software) programs applicable to existing laboratory systems have not always been readily available.

Custom software. Affordable custom programming has been hard to find--although some LIS vendors now include bar code packages with their systems and some bar code printer vendors provide interfaces to existing LISes.

Speed. Printers formerly were not capable of producing labels fast enough to meet peak workflow.

Wide stock. Too few printers accept the 8.5-inch-wide stock needed to print bar code labels and collection lists side by side to keep collection lists as short as possible to facilitate handling by phlebotomists.

Prevalence. Understandably, labs hesitated to be pioneers. Until fairly recently, there were relatively few documented cases in which a lab successfully printed bar code labels and integrated them into routine workflow. Fortunately, technology has advanced greatly, leading labs to take advantage of it--especially as they have observed successful implementation elsewhere.

* Specifications. For each bar code symbology, a set of specifications must be followed during printing. Failure to do so will produce symbols that will yield low first-read and high substitution-error rates.

If many labels fail to scan when placed on the analyzer, technologists will lose faith in the system and begin to hand-key accession numbers into the instrument. Thus in designing a bar code system every effort should be made to create a scan rate greater than 99%.

* Printer categories. Various printer types handle bar codes in different ways. A discussion follows.

Dot matrix. Designed for text, these printers can reproduce bar codes when connected to a properly programmed driver. Because of their need for multiple ribbon passes, however, it is virtually impossible for them to sustain the symbol quality needed in the laboratory.

When the ribbon is new, the dots print larger than the print head needles. As the ribbon ages, the dots shrink in size. As long as the dots overlap sufficiently, a usable bar code can be printed. As the dots become smaller, overlap decreases and the bar code read rate rapidly falls to unacceptable levels. One-time-use film ribbons, which would solve the problem, are prohibitively expensive.

Dot matrix printers are capable of printing Code 39 at a density of four characters per inch. Conserving "real estate" on small laboratory labels is thus a problem.

Laser. Contemporary laser printers can turn out high-quality bar codes consisting of 200 to 300 dpi. Since most laser printers are sheet fed, however, their use in the laboratory is limited. Most laboratory applications demand continuous-feed, fan-fold, or roll-label stock.

Thermal transfer. The adoption of thermal transfer technology for lab printers has been somewhat limited. Its principal advantage is in printing on plain paper rather than special label stocks. The crucial consideration is to make sure the label and ribbon work well together. Otherwise the printing is prone to smudging or being wiped off in handling. It may also come off in a water bath. In addition, thermal transfer printers with wide throats have had paper and ribbon alignment problems that require adjustments to maintain print quality. The thermal ribbon becomes an additional consumable to order, store, keep track of--and pay for. These expenses may be mitigated by the ability to use low-cost paper.

Direct thermal. Most lab applications now use direct thermal technology. Advances in direct thermal paper stock have paralleled the development of sophisticated printers, allowing bar codes with 200-dpi resolution to be produced. Barrier-coat papers that resist smudging are available. Another innovation is freeze-proof adhesives.

Because the operational life of thermal print heads is limited, they are considered a consumable. While direct thermal printing saves money in not requiring ribbons or toners, it does call for the more expensive label stock.

Intelligent printers. These machines represent the most significant advance in laboratory bar code printing. Instruments such as the MedPlus 2400F (MedPlus, Cincinnati, Ohio), a direct thermal printer used in the bar code specimen identification system at Bethesda Hospitals, Cincinnati, were specifically engineered for applications in the clinical laboratory.

These printers take a data stream from the LIS, analyze it in their internal microprocessors, select the relevant data, configure an appropriate bar code label, and print the bar code in the location best suited to the analyzer that will be used. The printer provides optimal bar code symbology for that analyzer.

* Computer peripheral. Picture the bar code printer as a sophisticated, programmable peripheral device that has substantial data-handling capability. It interfaces well with the LIS and is flexible in label design. Because the interfacing software is contained on a cartridge, changes in label layout or symbology are easily made--usually without involving the LIS vendor.

Understanding bar code technology grants the savvy laboratory manager or technical expert an unusual degree of autonomy. A primer and glossary are provided in the article that begins below.


1. Rappoport AE. A hospital patient and laboratory machine-readable identification system revisited. J Med Syst. 1984; 8(1-2): 133-156.

2. Rappoport AE. Machine readable identification systems: If bar code works in the supermarket, it should be great for medicine. Pathology. 1985; 39(2): 39-40.

3. Wielert M, Aller RD, Pasia OG. Bar code breakthroughs slow to turnup. CAP Today. June 1989; 3(6): 6-7.

Dr. Kasten is associate director of pathology and laboratory services, Schrand is laboratory computer system manager, and Disney is laboratory information system analyst, Bethesda Hospitals, Cincinnati, Ohio.
COPYRIGHT 1992 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Bar Codes, part 1
Author:Kasten, Bernard L.; Schrand, Pam; Disney, Mary
Publication:Medical Laboratory Observer
Date:Dec 1, 1992
Previous Article:POLS adjusting to life under CLIA.
Next Article:What are all those lines and spaces? Understanding bar code technology.

Related Articles
Heighten efficiency with an integrated bar code system.
Strategic planning for an integrated bar code system.
Creating a bar code chemistry system.
We carved out bar coding goals - and reached them.
Wristbanding for positive patient ID.
Reading between the lines: using bar-code technology is a smart way to keep track of business data.
Efficiency gains with bar coding and interfaced instruments.
Bar coding stirs greater interest.
Using bar codes in the lab.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters