From: LindaB@psbc.org Sent: Wednesday, July 17, 2002 9:54 AM To: fdadockets@oc.fda.gov Subject: [Docket No. 02N-0204] Comments on [Docket No. 02N-0204] General Questions Related to Drugs and Biologics: 1. Which medical products should carry a bar code? For example, should all prescription and over-the-counter (OTC) drugs be bar coded? Should blood products and vaccines carry a barcode? The Puget Sound Blood Center collects, prepares and distributes many medical products that may be subject to this regulation: blood components for transfusion and further manufacture, tissue allografts for transplant, and other human cells, tissues, and cellular and tissue-based products (e.g., stem cells, cord blood, islet cells). Most of these products currently carry a barcode. The reason that these products are bar coded is two-fold: for internal process control and external process control. Internally, the bar coded information helps to automatically correlate product labeling with manufacturing records via a unique identification number. Additionally, other sets of information may be bar coded to assure the accuracy of additional information (such as ABO, expiration date, and component attributes). The bar codes provide a vehicle for computer-assisted checks of labeling, production, and distribution histories. Virtually every critical step in the production process is verified by scanning information from the bar codes into the computer system. New labels are generated and applied as product qualities change, and again, the information is verified via the barcode. Process control is fundamentally aided by use of bar codes. Externally, many of our hospital clients scan the information from the blood product bar code into their computerized hospital systems. This provides them with an up-to-date inventory and distribution histories. While few are using further automated tools to assure that the appropriate blood component is transfused to the correct patient, many have expressed interest in such systems. To our knowledge, none of our tissue clients use the bar codes for computer-assisted allograft tracking or distribution. As a centralized transfusion service providing compatibility testing for more than 15 different hospitals, the Puget Sound Blood Center endeavored decades ago to standardize the barcode symbology used. Even before the FDA promoted Codabar as the standard symbology for blood components, we agreed amongst the hospitals we serve to use Code 39. Consequently, our region is fairly well integrated. However, several clients failed to implement automated systems or ultimately used different coding schemes to accommodate other models. This handicaps the utility of the bar codes or requires the introduction of "wedges" or translation programs. Most recently, ISBT 128 was heralded as the universal symbology for blood components. Partly due to the Gulf War and military interests in standardizing blood labeling internationally, there was substantial momentum for moving forward with this model. Within the Puget Sound Blood Center, extensive research, design, and equipment modifications have since been made to accommodate the anticipated advance as well as over ten thousand dollars in licensing fees. However, the further licensing fees imposed on hospitals and the requisite modifications to their systems have presented real limitations to the utility of such a costly change. In a survey of our hospitals conducted last year, only about 30% (11 out of 31 respondents) were willing and ready to move forward with ISBT 128. The Puget Sound Blood Center finds bar coding to be absolutely necessary internally. Externally, the benefits of bar coding would be furthered if integrated with hospital systems so that blood and tissue products could be assigned to the patient's electronic medical record. Verification steps or quality control measures could then be automated that aided the human element of transfusion and transplantation medicine. Agreeing on a standard is essential in order for this to happen. 2. What information should be contained in the bar code? What do you consider to be critical bar code information that will reduce medical product errors? If data exists, please provide it for the record. What information would be helpful but not necessarily critical, for reducing medication errors? Provide data. Blood Components are identified with the following: Information on Component label Presentation Bar Code Used By Useful in reducing medication errors Unit Number Bar coded and Eye-readable Blood Centers Some hospitals No nursing homes No outpatient transfusion services No Plasma Derivative manufacturers Yes ABO/Rh Bar coded and Eye-readable Blood Centers Some hospitals No nursing homes No outpatient transfusion services No Plasma Derivative manufacturers Yes Expiration Date Bar coded and Eye-readable Blood Centers Some hospitals No nursing homes No outpatient transfusion services No Plasma Derivative manufacturers Yes Component Name Some aspects Bar coded, All Eye-readable Blood Centers Some hospitals No nursing homes No outpatient transfusion services No Plasma Derivative manufacturers Yes Facility Registration Number Bar coded and Eye-readable Few Blood Centers Few Transfusion Services No nursing homes No outpatient transfusion services No Plasma Derivative manufacturers No Additional information (directed, autologous, released uncrossmatched) Eye-readable only N/A Yes Facility identity Eye-readable only N/A No Instructions Eye-readable only N/A No Volume or pool size Eye-readable only N/A No Prescription Only Eye-readable only N/A No Anticoagulant preparation Eye-readable only N/A No For blood transfusion, other critical information sources are the Transfusion Report attached to the blood unit that identifies which patient is to receive the blood component and the patient's wristband. To reduce medical errors in blood transfusion, it is our recommendation that bar coding be carried through to these information sources. Tissue Allografts are identified with the following: Information on Component label Presentation Bar Code Used By Useful in reducing medication errors Donor ID and Deposit# Bar coded and Eye-readable No distributors or hospitals Yes Expiration Date Eye-readable only No distributors or hospitals Yes Tissue Code (product name and qualities) Bar coded and Eye-readable No distributors or hospitals Yes Size Bar coded and Eye-readable No distributors or hospitals Yes Additional information (directed, autologous) Eye-readable only N/A Yes ABO/Rh Eye-readable only N/A Yes Testing Statements Eye-readable only N/A No Facility identity Eye-readable only N/A No Handling Instructions Eye-readable only N/A No There is no standard for tissue codes or format. We are aware that some tissue banks are able to read each other's bar codes and may use them for tracking and distribution purposes. It is important to note that for many tissue products, the risk an error poses to the patient may be less severe as compared to an incompatible blood transfusion. Most tissue products are not specifically matched to a patient and are exchangeable. While still undesirable, most tissue allograft errors that we are aware of are observable and result in replacement with the appropriate product. However, certain tissue, cellular and tissue-based products (e.g., stem cells, cord blood, islet cells) do carry substantial risk if not directed to the specific patient. These products are often HLA-matched and not interchangeable thus raising the risk if erroneously administered to a patient or not provided to the correct one. Proposed Good Tissue Practices require that tissue banks maintain records showing who received tissue products. Bar coded information may be retained in the patient record, more as a reflection of what was transplanted than as an automated verification of what was ordered. Again, to reduce errors, the bar coded information must be carried through to the patient record via a computer-assisted system. For tissues, there are substantial issues beyond bar coding that must be addressed to achieve this ideal. 3. Considering current scanners and their ability to read certain symbologies, should the rule adopt a specific bar code symbology (e.g., reduced space symbology (RSS) and 2-dimensional symbology)? Should we adopt one symbology over another, or should we allow for "machine readable" formats? What are the pros and cons of each approach? In order for the bar coding to be useful for the manufacturer, distributor and healthcare professional, the information contained in the bar code must be interpretable and meaningful. "Machine readable" formats which allow for automated translation of standard data sets seems most appropriate. In this way, certain data elements in defined formats could be introduced to any system designed to accept them. The focus should be on issues of patient safety and product traceability. Reflecting on experiences with ISBT 128, both a universal symbology and standard data elements were prescribed. The failure to yet implement this approach derives from the substantial expense and effort to retrofit existing systems in the manufacturing and distribution setting in order to provide or accept the required data elements in addition to acquiring equipment suited to the new symbology. One can only surmise that few hospitals will actually utilize the new method based on the fact that a minority are using computerized systems for blood inventory, few for tissue, and the expense posed to those that do is daunting. We also feel it important to emphasize that barcodes may not be the only mechanism to reduce medication errors. Other suggestions have included transponders or exclusion devices to prevent administration. These methods merit investigation as well. 4. Assuming that we require bar codes on all human drug products, where on the package should the bar codes be placed? Are there benefits to placing bar codes on immediate containers, such as the bottles, tubes, foiled-wrapped tablets, and capsules, found inside prescription or OTC product cartons? Is there a way to distinguish whether certain containers with a bar code will have a more significant effect on preventing errors than others? Minimally, the product should be bar coded on the final container. To reduce transfusion errors, the blood product must be corroborated against the Transfusion Report and the patient (preferably against an attached patient wristband). An automated mechanism to assure that these three pieces of information match is desirable, which would require bar codes on the Transfusion Report and patient wristband. In our 50 years' experience, there are 3 different sources for error in transfusion: 1) The pre-transfusion sample is drawn from Patient A then erroneously labeled as being from Patient B. This usually occurs when samples are labeled at a different point in time than they are drawn. Were there a system to absolutely correlate the sample to the patient, this opportunity for error would be eliminated. 2) The wrong patient record is accessed when compatibility testing is performed leading to an incorrect conclusion about the suitability of the blood for transfusion. This may occur if multiple samples or patients are being handled simultaneously or simply through a transcription error. Were a system in place to access the patient's record from a bar code on the sample, this opportunity for error would be eliminated. 3) The blood component intended for patient A is transfused to patient B. Any of the above errors could lead to this eventuality or there could be an independent error whereby patient A is not recognized to be patient B. Most commonly, this seems to occur if one of the three pieces of information is not utilized for comparative purposes. Were a system in place that required input from each of these sources and automatically identified discrepancies, opportunity for this error would be eliminated. For tissue allografts, the external container is bar coded. The internal sterile container information is usually only eye-readable. This should be sufficient so long as the external container is compared to the patient record and patient identity. 5. What products already contain bar codes? Who (i.e., hospitals, nursing homes, outpatient clinics, retail pharmacies, etc.) uses these bar codes and how? As with all comments, if data exists, please provide it for the record. Please refer to the response for Question 1. Thank you for the opportunity to comment. Sincerely, Linda S. Barnes QARA Director Puget Sound Blood Center *** IMPORTANT NOTICE *** This e-mail message, including any attachment, has been scanned for viruses. The content is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the appropriate recipient, please contact the sender by reply e-mail and erase all copies of the original message.