FDA MEETING REPORT: INSULIN VIALS WILL HAVE TACTILE MARKINGS

by Ed Bryant

Early in 1992 I was contacted by a blind diabetic, who informed me that he and his fellows, perfectly capable of accurately drawing up their insulins, had no reliable way to distinguish between insulin types. All insulins, fast-, intermediate- or long-acting, could only be told apart by reading print on the label. This, I was reminded, placed many in grave danger, as consequences of vial misidentification could be severe.

I conducted a national survey: Was change needed? Should insulin vials have tactile markings, to help blind diabetics, those losing vision, the hurried, the elderly, the young, busy medical professionals, overworked pharmacists, and the rest of us? Survey results were clear and unequivocal--change WAS needed. For the best of reasons, safety and independence, insulin types should be identifiable by touch.

Representing the Diabetes Action Network of the National Federation of the Blind, I campaigned for this goal: to make tactile-marked insulin vials available. I wrote letters to the U.S. insulin manufacturers (Eli Lilly and Company, and Novo Nordisk Pharmaceuticals Inc.) and to the Food and Drug Administration (FDA). Hundreds of you, VOICE readers, joined me in the letter-writing campaign. Many nurses and other medical professionals wrote in support, acknowledging the print was so small they had trouble reading it, and that misdosage mistakes were made in hospitals. Pharmacist error was covered, too (and this actually happened to me!).

We finally persuaded the FDA to call a meeting of interested participants: Insulin manufacturers, diabetes educators, drug packaging/labeling firms, diabetes associations and organizations of the blind. The first meeting took place on October 19, 1995, at FDA headquarters in Rockville, Maryland.

At that meeting, participants agreed that non-sighted insulin vial identification was a necessity, and that such coding should be factory-applied, durable, and sufficiently prominent that blind diabetics with neuropathy could use it. The insulin manufacturers were to come to the next meeting (to be held in three months' time) with both short-term prototypes and long-term proposals.

Bad weather and scheduling conflicts forced postponement of the second meeting, which was not held until April 10, 1996. By the close of that meeting, Lilly and the FDA were ready to agree on a set of one through four tactile bars on the label, as a means of distinguishing insulin classes. But Novo Nordisk asked for more time "in which to test alternative prototypes." We agreed to meet again, some time in July 1996.

For reasons still unclear, Novo Nordisk was not ready by the July deadline, and did not transmit its findings to the FDA until the end of December. FDA officials attempted to schedule the "rematch" in March or April of 1997, but scheduling conflicts delayed the final meeting until June 3, almost one year late.

At the June meeting, the insulin manufacturers presented their test findings. Lilly related how their researchers had sought out blind diabetics with differing degrees of neuropathy (mild, moderate, and severe), then tested their success in distinguishing dot codes, vertical lines, and horizontal lines on the vial label. Lilly found that although a few individuals had neuropathy too severe to recognize any system, a series of wide horizontal bars provided greatest accuracy: over 98% successful tactile recognition. (Note: This finding mirrors the consensus of ALL consumer groups present at the meeting.)

When the question of tactile-label durability was raised, Lilly related how they had tested the bars under a wide variety of conditions, including long-term immersion in alcohol, and experienced no failures. A representative from CCL Label, a national company that makes vial labels for pharmaceuticals, confirmed that his company could guarantee durable tactile bars on vial labels. I pointed out that even if there were the rare label failure, the system would be more reliable than at present, where blind insulin users are stuck with rubber bands or tape.

Novo Nordisk, who had tested tactile prototypes with blind diabetics but not considered the impact of neuropathy on label conformation, raised a number of objections. In spite of Lilly's tests and CCL Label's guarantees, in spite of the FDA's satisfaction with the system, Novo Nordisk continued doubting its appropriateness and reliability. A very early tactile-label prototype, that had failed "torture tests" over a year ago, was put forward as "evidence" that the four-bar system was not sufficiently durable. Perhaps Novo's label-supplier in Denmark is having difficulties--I find it interesting that Lilly's supplier has mastered the problem.

All organizations present except Novo Nordisk accepted the four-bar system as presented. A Lilly representative told me his company was "readying their production line." The FDA stated that once they had completed the approval process, one company could proceed without waiting for consensus from its competitor. Because the FDA "didn't want Novo Nordisk to feel they were being treated unfairly," they suggested Novo could raise the issue at the International Diabetes Federation (IDF) meeting, to be held this July in Helsinki, Finland. If they raised sufficient objection, and there was strong support at that meeting, FDA might reopen the discussion process.

The FDA promised that on August 1 they would fax meeting participants their final determination. They stated we would go with the system as agreed, unless there is substantial international objection.

The above disturbs me greatly. Although it is unlikely the IDF would support Novo Nordisk in the face of the evidence, outside the USA Novo Nordisk is a very major provider of funding to diabetes agencies and foundations. Also, such a decision allows the IDF to "sit in review" of FDA policymaking, a serious surrender of FDA's statutory responsibilities to an international body. I hope it doesn't come to that.

To review: At the meeting, we agreed on the following system of four horizontal tactile bars on the insulin vial label: One bar = fast-acting insulins such as Humalog; two bars = Regular insulins; three bars = any mixed insulin (70/30 or 50/50); and four bars = longer-acting insulins (NPH, Lente, Ultralente).

Once final approval is granted, Lilly estimates it will take between six and 18 months for the tactile-marked insulin vials to reach pharmacists' shelves. Insulins have a shelf-life of two years from date of manufacture, so it may be as long as two years from the start of tactile- labeled insulin production before all the older, unmarked vials are off pharmacists' shelves, though the bulk will be replaced far sooner.

It has taken a long time, but hopefully the next report the VOICE carries about tactile-marked insulin vials will be a review of the first to reach production, and the impact they have on our safety, independence, and diabetes self- management.