The Vaccine Identification Standards Initiative
National Immunization Program / U. S. Centers for Disease Control and Prevention
Conference Call Summary
April 2, 1998
3:00 P.M. - 4:30 P.M. EST
|Bruce Weniger (chair)||NIP / CDC|
|Joshua Schwartz||NIP / CDC|
|Karen Chaitkin||CBER / FDA|
|Suzy Feikema||NIP / CDC|
|Sonya Gupta||GA Div. of Public Health (for S. McKenna)|
|Del Carvell||GA Immunization Program|
|Gina Galliera||SmithKline Beecham|
|Susan Abernathy||NIP / CDC|
|Ron Filipski||Pasteur Mérieux Connaught|
National Drug Code (NDC) Issues
Suzy Feikema described several problems with using the NDC for vaccine identification for immunization registries. (See attached draft memo "Why not NDC codes.") Karen Chaitkin added that collaboration of companies in the production / distribution of products can also contribute to NDC confusion. Notably, the last company to touch a product before distribution uses its specific NDC on the packaging. Thus, the company encoded in the NDC may not be the company that manufactured the product.
Ms. Chaitkin noted that the FDA assigns the labeler code (the first four or five digits) for the NDC and the company chooses the product code and the package code. An FDA contractor oversees the use of NDC codes. Ms. Chaitkin planned to speak to the contractor and provide specific details to the working group about the process of assigning and maintaining NDCs.
Ms. Feikema noted that if accurate, easily updated tables of NDC data existed, the NDC should be useable. She cautioned, however, that the NDC is useful only for computers; it is not a useful tool to identify vaccines to humans, as it contains only nonintuitive numerals.
Susan Abernathy noted that the standards development organization Health Level 7 (HL7) has a Vocabulary Special Interest Group that is currently working on developing a new coding scheme/vocabulary to represent all pharmaceutical products, including vaccines. This process is expected to take several years. In the meantime, the National Immunization Program continues to maintain its codes for identifying vaccines products by their vaccine type and appropriate manufacturer. NIP advises immunization registries to use these codes in their systems.
Joshua Schwartz described that for bar coding, the NDC has been incorporated into a standard promulgated by the Uniform Code Council (UCC). The UCC/EAN-128 symbology can encode, in a single bar, the NDC, the expiration date, and the lot number. Thus, if the NDC numbers are to be used for vaccine identification in the VISI, the UCC has already solved the issue of how a single bar code could provide both the NDC and other useful information.
Bruce Weniger informed the group that finding a complete list of vaccines and their NDC numbers currently in circulation has been difficult. Representatives from SmithKline Beecham, Pasteur Mérieux Connaught, and Merck [plus John Kropas of Wyeth-Lederle Vaccines and Pediatrics in a subsequent private communication] all agreed to provide to the working group a complete listing of NDC numbers for their respective vaccines. The next step might be to solicit assistance from an appropriate unit at FDA, CDC, or one of its contractors to maintain a promptly-updated, easily and universally-accessible, and perhaps automatically-downloadable database of vaccine NDCs to update lookup tables for immunization/medical practice software developers. This might be done via a simple Internet web page.
Alternatives to the NDC
Dr. Weniger asked if anyone could suggest alternatives to the NDC to uniquely identify vaccines. None were suggested. Richard Zimmerman agreed that it would be beneficial if we could avoid "reinventing the wheel" on this issue. The consensus was that it was better to work to overcome the drawbacks to the already widespread use of the NDC in various national and international product identification standards systems than to try inventing an entirely new identifier.
Year 2000 and Bar Codes
Mr. Schwartz covered the issue of formatting for expiration dates considering the imminent Year 2000 computer problems. For bar coding, the UCC employs a standard expiration date format of YYMMDD. There appear to be no efforts to revamp the format for the Year 2000 issue, as it is the role of the applications software to decipher the expiration date from the above format.
Accordingly, Dr. Weniger suggested that the group accept the UCC standard for expiration dates. However, for the package labeling that depicts the expiration date in text for human users, Dr. Weniger suggested that the group select a feasible standard that removes the possibility for errors of interpretation. All agreed to the format YYYY-MON-DD, where numerals represent the year and day and the first three letters of the month represent the month. The NIP will employ this format in the next prototype.
Lot Number Length Issue
The number of characters in a lot number will affect the overall length (and height) of a bar code that includes lot number data. Dr. Weniger suggested that the group determine the maximum number of characters that a lot number might have so NIP could produce a representative prototype bar-coded sticker. Such a prototype would provide an estimate of the size of an actual sticker.
The original prototype sticker from NIP allowed for 12 lot number characters. Industry representatives on the call said that the lengthiest lot number was approximately eight digits, although most of the lot numbers currently used for vaccines contain four or five digits.
Mr. Filipski recommended erring on the larger size rather than smaller. The group agreed. Accordingly, future prototypes for bar coding will include space for 12 lot number characters.
Mr. Filipski expressed concern regarding the overall size of a peel-off sticker and questioned how the manufacturers could store such stickers within vaccine packaging.
Bar-code Size as Determinant of Sticker Size and Form Area
Dr. Weniger pointed out that the minimum amount of space to reserve on a vaccine administration record form (and thus the maximum size to allow for peel-off vaccine ID stickers) depended on both the number of characters to reserve for the lot number (now 12 alphanumerics, see above) and the minimum "X" dimension to be specified for the bar-coding. The "X" dimension for the UCC/EAN-128 symbology is the width of the thinnest bar of a bar-code, in thousandths of an inch, usually ranging from 7 to 40 thousandths. Higher quality paper, printers, and scanner devices allow for a smaller "X" dimension and a resultant smaller bar-code.
Manufacturer representatives indicated that since the bar-code would include lot numbers and expiration dates that would frequently change, printing would occur "online" during the packaging and labeling process. This might require a somewhat higher "X" dimension than "offline" printing technologies. The manufacturers agreed to provide the working group with their minimal requirements for the "X" dimension for bar-codes on both peel-off stickers and on vaccine primary and/or secondary packaging. These numbers would be used by the VISI to develop prototype labels that illustrate the maximum size needed for the stickers and the minimum size to reserve for them on the record forms.
Dr. Weniger pointed out that the prototype labels being prepared were just examples, and that, in general, the VISI should be as unprescriptive as possible. The VISI should specify the minimum standards necessary to achieve its goals of universal compatibility of the labels and forms with scanning equipment. The Initiative should encourage innovation by each company on the individual design and layout of their labels and packaging.
A participant made a recommendation to ensure that stickers were easy to peel away from the backing material. This prompted the suggestion of looking into the adhesive currently used for peel-off stamps by the U.S. Postal Service, as these stamps are extremely easy to peel off, yet bond tightly to paper envelopes.
Dr. Weniger asked if the industry representatives would look into the use of adhesive technologies at their companies for existing pharmaceutical peel-off stickers and provide text of potential specification standards (e.g., from any existing adhesive standards organizations) that might be adopted by reference for the proposed VISI labels standards. A label that does not permit removal from an inappropriate location and replacement elsewhere on the form within the first several minutes to hours after initial placement might be problematic. A test of the new Postal Service peel-off stamps subsequent to the call indicated too rapid a degree of permanent adhesion to paper that might not be suitable for vaccine stickers.
Future Conference Calls
The group agreed that the time of day of the conference call (3:00 P.M. EST) was acceptable for future calls and that the next call should occur in mid-May or mid-June.
Joshua G. Schwartz, MPH
Vaccine Development Fellow
Vaccine Safety and Development Activity
National Immunization Program / CDC (MS E-61)
1600 Clifton Road
Atlanta, GA 30333