syphilis in the early 1900s and diabetes mellitus in the 1950s and 1960s. It may be exceptionally difficult for the individual physician to decide whether a detected untoward event occurring during a patient's clinical course is a complication of the disease or the treatment. The FDASRS does not ask the physician to decide the cause of the ADE; instead, it seeks suspicions or possible attributions. An isolated case may be irreparably confounded by other factors making attribution impossible. When many such similar cases are reported, trends or clusters can be examined. Therefore, if a strong suspicion exists that the drug is involved in producing the event, the case should be reported. More than one third of physician respondents detected and attributed an ADE to drug therapy during the previous practice year, and 34% of these events were serious. Only 5% of all observed ADEs were reported directly to the FDA-SRS. Those physicians who reported the ADEs that they observed appeared to be less busy than their counterparts, prescribed fewer drugs, spent more time in hospital practice, and were more likely to be from the primary care specialties. Inpatient care remained significantly associated with reporting and can be explained by the broad definition of reporting within the questionnaire. One of the reasons that physicians gave for not reporting was that the event was not serious. This valid reason, consistent with other reports,' appropriately spares the SRS. Reporting all common side effects and nuisance symptoms would overburden the SRS and hamper it in efficiently responding to the signals of serious risk. In general, a serious event can be defined as one that leads to or prolongs a hospitalization, contributes to or causes death, or is associated with cancer or a congenital anomaly. Another more important reason physicians gave for not reporting was that the event was already known and documented. This response indicates a basic misunderstanding about the SRS. While it is true that its primary purpose is the generation of early signals of new drug problems, the utility of the SRS is not limited to that function. It can also serve as a monitoring system to detect changes in the nature or frequency of ADEs. These changes can result from a shift in the natural history of the disease, the development of alternative therapies, the use of newer agents in combination with the drug, and the aging of the population with resultant alterations in receptor sensitivity, pharmacodynamics, and pharmacokinetics. Any one of these changes can influence the risk-benefit ratio for the drug. Therefore, all ADEs, new or documented, particularly for newly marketed drugs, should be reported. A major reason for not using the FDA-SRS was the unavailability of the FDA 1639 form when it was needed. This finding supports the FDA's attempt to test innovative reporting mechanisms through a series of pilot projects. A project with the Rhode Island State Health Department is placing abbreviated reporting forms in every physician office. Another project at the Maryland State Department of Health is providing a telephone reporting service. The first-year experience with these projects has resulted in a fourfold increase in Maryland and an eightfold increase in Rhode Island of physician reporting. Other FDA-supported ADE reporting projects have begun in Mississippi, Colorado, and Massachusetts. The attitudinal response of the physicians is encouraging for it demonstrates that the majority of physicians have appropriate doubts about drug safety and appreciate the link that they provide between patient and drug. Furthermore, they accept this link as a professional responsibility requiring no reimbursement. There are some concerns about the burden of additional government forms and a perception, specifically on the part of some nonreporters, that the act of reporting increases their own liability. The FDA-SRS is focused on the public health identification of that component of drug risk that is unpredictable, uncommon, and serious. Physician misjudgment or error with prescription drugs is not of interest nor are product liability issues. In fact, as noted on the official reporting form, the submission of a report does not constitute an admission on the physician's part that the drug caused the event. Furthermore, the name of the reporting physician is held in confidence by the FDA and is not subject to release under the federal Freedom of Information Act. Finally, the act of reporting in itself demonstrates the highest level of professional responsibil ity in voluntarily contributing to the public health. Drug safety cannot be guaranteed at the time a drug is marketed because the extent and nature of exposure is limited, but after marketing, the timely definition of drug risk is certainly achievable. Adverse drug event reporting plays a large role in achieving this objective. Reporting cannot be made mandatory. Isolated ADES often present as little more than suspicions: there are no measures comparable to the culture or serologic titer in infectious disease. The public must rely on the responsi bility of the physicians in whom they have placed their trust. Medical school curricula need to address the issues of drug development and the critical role physicians play in monitoring for drug safety. The professional responsibility for reporting suspected ADEs should be modeled for students, interns, and residents by attending physicians. State medical societies should evaluate the level of reporting within their professional community and encourage physicians to participate in the reporting system. Adverse drug event data are routinely collected by hospital pharmacy and therapeutics committees or quality assurance committees and, in many cases, just as routinely internally reviewed and filed. Physician staff members who serve on these committees should insist that such data be communicated to the FDA-SRS and see that a mechanism for doing so be established. Research on which this publication is based was performed pursuant to contract 223-85-4267 with the Food and Drug Administration, Department of Health and Human Services. The authors are grateful to John Juergens, PhD, Food and Drug Admin istration, for reviewing the manuscript. References 1. Faich GA: Adverse drug reaction monitoring. N Engl J Med 1986;314:1589-1592. 2. Rossi AC, Knapp DE: Discovery of new adverse drug reactions. JAMA 1984;252:1030-1033. 8. Rossi AC, Knapp DE, Anello C, et al: Discovery of adverse drug reactions. JAMA 1983:249:2226-2228. 4. Griffin JP, Weber JCP: Voluntary systems of adverse reaction report. ing: Part 2. Adverse Drug React Acute Poisoning Rev 1986;1:23-55. 5. Faich GA, Knapp D, Dreis M, et al: National adverse drug reaction surveillance: 1985. JAMA 1987:257-2068-2070. 6. Inman WHW (ed): The United Kingdom, in Monitoring for Dru Safety Lancaster, England, MTP Press Ltd, 1980. 7. Koch-Weser J Sidel VW, Sweet RH, et al: Factors determining physician reporting of adverse drug reactions. N Engl J Med 1969:280:20 26. 8. Wardell WM, Tsianco MC, Anavekar SN, et al: Postmarketing sur veillance of new drugs: I. Review of objectives and methodology. J Clin Pharmacol 1979;19:85-94. 9. Milstien JB, Faich GA, Hsu JP, et al: Factors affecting physician reporting of adverse drug reactions. Drug Inf J 1986;20:157-164. 10. Fleiss JL: Statistical Methods for Rates and Proportions, ed 2. New York, John Wiley & Sons Inc, 1981, chap 9. 11. Griffin JP, Weber JCP: Voluntary systems of adverse reaction report. ing: Part 1. Adverse Drug React Acute Poisoning Rev 1986;4:213-230. Adverse Drug Events-Rogers et s Mr. SHERIDAN. Well, sir, again, we did not assume in our analysis that all cases of PM/DM would have indeed been reported to the Collagen Corp. We have taken steps to try to encourage rheumatologists in this country-and we have searched other data bases. We have searched the literature to try to obtain information on all the patients that we can possibly obtain information on. But, as I said, sir, there are two problems, not just the reporting of PM/DM cases for people who have collagen; there is a problem with the identification and reporting of all PM/DM cases, since it's such a difficult disease to detect and properly classify. And then, also, even if that is done, not all cases are reported to the CDC. Mr. WEISS. Our analysis of your statistical analysis indicates that there is an assumption of 100 percent reporting, but we will be submitting some technical questions to you to try to clarify that. [Questions and the FDA response are in app. 2, pp. 273-276.] Mr. SHERIDAN. We will be happy to answer those, sir. Mr. WEISS. Isn't it true that if you assumed only half the PM/ DM cases that occurred within a year of collagen injections were reported, a rather generous estimate, the number of cases of PM/ DM would be considered statistically significant? Mr. SHERIDAN. Sir, I would like to defer those questions to our expert and submit answers in writing. I can give you a general sense of the basis of our disagreement with the Texas Department of Health. We disagree strongly with their conclusions. There are two basic issues at stake here, and that is whether or not the patients that have been identified should be included in an analysis because of other confounding factors that may Mr. WEISS. Well, we will wait for your response to the technical questions that we submit. Mr. SHERIDAN. Thank you, sir. We will be happy to submit them. Mr. WEISS. The Texas Department of Health has concluded that collagen is significantly associated with PM/DM. The Collagen Corp. and the FDA have refuted their findings thus far. If they eventually convince you that they are correct, what action would FDA take? Mr. SHERIDAN. If there were an established link between PM/ DM and the use of collagen for cosmetic purposes, we would clearly have to reevaluate the risk/benefit ratio associated with this device. Now, I can't independently give you an answer without talking to our scientists and to our advisory panels, but I believe that, given that the indication is purely cosmetic, that it might require that we withdraw approval for the product. But, again, we would have to seek advisory panel opinion and discussion with our scientists. Mr. WEISS. FDA determined that the Collagen Corp. failed to report at least one-third of the serious adverse reactions to the medical reporting system. The Collagen Corp. disagreed. Last fall, FDA and the company officials worked out several compromises regarding reporting requirements. One decision was that abscess formation did not require a report since it is discussed as a possibility on the package insert. Is that correct? [Relevant documents are in app. 5, p. 373.] Mr. SHERIDAN. Sir, I believe that is correct. It gets very complicated. The MDR requirement is very intricate. I think it has to do with Mr. WEISS. OK. Well, anyhow, if in fact it is correct, it is correct. A second decision was that scars lasting less than 6 months do not need to be reported. Is that correct? Mr. SHERIDAN. My recollection-I am not expert-is that that was an initial position taken by the Collagen Corp., sir. Mr. WEISS. And it was not agreed to by FDA? [This is also discussed in the letter in app. 5, p. 374.] Mr. SHERIDAN. MDR is not my area, and I would prefer to answer those questions Mr. WEISS. Who here has it as their area? Mr. LEVITT. I don't believe that anybody at the table here is an expert in the MDR. Mr. WEISS. Well, would you please give us responses to these questions in writing. Mr. LEVITT. We will be happy to. [The response is in app. 2, pp. 276-277.] Mr. WEISS. The cosmetic benefits of collagen treatments usually last 3 or 4 months; is that correct? Mr. SHERIDAN. There are different periods of duration-different durations for correction, depending upon the nature of the defect that is being corrected. Some corrections actually last a very short time, while others endure for a fairly substantial amount of time. Mr. WEISS. The question is "usually." Usually, the cosmetic benefits for wrinkles and lips last for 3 and 4 months; is that correct? Mr. SHERIDAN. The indication on the collagen label says that duration can be expected for periods of 6 to 18 months. And you raise an interesting point-you make an interesting question-because we would like to reevaluate the durability of this procedure by asking Mr. WEISS. Now, again, are we talking about something that is within your area? Mr. SHERIDAN. Yes, indeed. Yes. Mr. WEISS. And you are not aware of the fact that the benefits are usually, for wrinkles and lips, usually 3 and 4 months? Mr. SHERIDAN. Well, as you well know, the device is not approved for lip augmentation directly into the muscle. So I suppose you are referring to corrections around the vermilion line of the lip, or just above the mucous membrane. Indeed, the corrections for wrinkles in that area are fairly durable and will last longer than the corrections for problems, say, in the globular area between the eyes and above the nose, for example. Mr. WEISS. Yet an abscess on the face or scars on the face lasting almost 6 months are not considered serious enough to be reported as an adverse reaction, by agreement; is that right? Mr. SHERIDAN. Again, sir, on the MDR questions, I am not competent to answer. Mr. WEISS. We have a pamphlet put out by the Collagen Corp. about their products. They don't really warn the consumers about the potential risks except in the very small print in the back. What other kinds of information are available to consumers or their doctors about the dangers of collagen? [The Collagen Corp. pamphlet follows:] On The Cover Sunny Griffin is a 47-year-old mother, building contractor, former TV correspondent and model. The unretouched cover photographs show Sunny before and after receiving injectable Zyderm and Zyplast® Collagen treatments. As you can see in the bottom photograph, many of her facial lines and wrinkles have virtually disappeared. Read the rest of this brochure to find out how you, too, may experience the same results. Collagen is the natural protein that cushions and supports your skin. When age, disease or injury wears down this underlying layer of support, a line or scar often forms. But today there's a unique and safe way to actually fill in these lines and scars. You can replenish your skin's own collagen with injectable Zyderm and Zyplast Collagen. Medical research at Stanford University led to the development of injectable collagen over a decade ago. Since then, this natural substance has helped over 300,000 men and women look their best. How It's Different Creams soften your skin's texture. Surgery removes excess skin. But injectable collagen adds a whole new dimension to your skin care program. It's the only nonsurgical treatment designed specifically to eliminate facial wrinkles and scars. |