New Drugs: a Dose of Reality — the Press

MEDICAL REPORTING tice to compensate physicians for time NEW DRUGS: spent," but would not say how much. Markenson said money went to the hospital. 'The contract was between me, Searle, and A DOSE OF REALITY the hospital," he said. 'The money pays for the time involved, nurses, physicians, spe cial equipment needed. It's not all profit to The Press Too Often Plays Up the Positive the university or the investigator, but there is some profit or nobody would be doing it." ate last year NBC Journalists could have began trumpeting the asked harder questions virtues of a new about the long-term L"superaspirin," the effects of Celeb rex. For drug Celebrex, which is jointly example, a University of marketed by G.D. Searle and Pennsylvania study pub Pfizer Inc. and would soon lished in January in the become the fastest-selling new Proceedings of the National drug ever. On December I, the Academy of Sciences dis Nightly News reported that a cussed a possible risk of Food and Drug Administration blood clotting, which can' arthritis advisory panel that result in heart attacks and reviewed the drug had recom strokes. The study, which mended FDA approval. The recommended larger clini segment featured Dr. Joseph cal trials, was funded by Markenson, identified as "with Searle, thollgh not men the Hospital for Special tioned in company press Surgery in New York," saying releases about Celebrex. that Celebrex, the first of a new NBC quoted Dr. Markenson on Celebrex without noting his link to the its maker (Searle says the study class of pain-relief drugs, reached "highly specula would "revolutionize the industry because ed" on a warning label, it then featured tive conclusions," though Dr. G. A FitzGer it's a whole new group of drugs that are Markenson giving a reassuring message. ald, its senior author, says it raised "a flag going to be safe." Finally in February, Nightly News revis that people hadn't thought about before.") The next day on the Today show Dr. ited Celebrex, with Tom Brokaw declar NBC did briefly mention potential cardiovas Steven Abramson, the physician who head ing that in clinical trials Celebrex "has cular risks on its January show, bui;not on ed the advisory panel, downplayed the proved to be very effective at treating the its February show. A Nexis search turns up drug's potential side effects. 'There may be pain without any side effects." only scattered mentions of the study, most some ulcers," he said, "though it's much NBC's stories were fairly typical of the of them on 10cal1V. less than the other drugs" that Celebrex reporting on Celebrex: the press tended to The Wall Street Journal did some digging was being compared with, the nonsteroidal highlight the positive findings, often failed of its own. Using the results of a Freedom of anti-inflammatory agents (NSAIDS), such to report that there were unknowns, and Information Act request, the Journal report as aspirin and ibuprofen. slid past potential negatives. And some ed April 20 that Celebrex had been linked to In January, after the FDA approved times, it did not disclose that its sources had ten deaths and eleven cases of gastrointesti Celebrex, Dateline NBC brought back financial ties to the drug company. For nal hemorrhages during its first three Markenson, saying he had closely moni example, NBC failed to report that Marken months on the market. (The FDA respond tored one of the patients featured on the son had helped test the drug as a clinical ed that the product did not pose "some spe show through the drug trials. While investigator for Searle. A Searle spokesman cial risk" at this time. Searle said the drug i;S Dateline noted that the FDA "has insist- conceded that it is "standard industry prac- was "performing as expected.") z 10 SEPTEMBER/ OCTOBER 1999 COLUMBIA JOURNALISM REVIEW Still, there were red flags on Celebrex a partial picture. It measures the risk of that might have led to more balanced sto the adverse outcome in those who receive ries about it. For example: treatment divided by the risk In the con • The FDA approved the drug only for trol group. For example, in one trial of rheumatoid arthritis and osteoarthritis, Fosamex, 2 percent of the women who not for acute pain, as the drugmakers had took a placebo experienced hip fractures, also wanted because, according to the while 1 percent in the group who took the FDA, there wasn't enough evidence to drug did not. In relative terms, this is a 50 show it was effective for that use. percent reduction. • Despite the company's wish to avoid a But in terms of absolute risk - the risk warning label, the FDA required essential of the adverse outcome in the control ly the same label it requires for all NSAID group minus the risk in the treatment products, noting the potential for gastroin group - the difference is only a 1 per testinal bleeding. centage point reduction. That sounds far • When the FDA approved the drug, it less dramatic, and is not likely to appear in noted that "additional studies in many company press releases. But it may be thousands of patients would be needed to more useful to a woman who is weighing see whether Celebrex actually causes the potential risks and long-term effects. fewer serious gastrointestinal complica 'There are many circumstances in which tions than other NSAID products." relative risk doesn't convey the full pic As Scott-Levin, a consulting firm, points ture," says Dr. Mark Chassin, chairman of out in its publication, Pharmaceutical the department of health policy at Mt. Quarterly, Celebrex got a "lukewarm FDA Sinai School of Medicine in New York. endorsement." Yet with some help from The Harvard researchers found that of the press, its sales have surpassed even the 124 stories that quantified benefits, 83 Viagra, bringing in some $600 million in percent used the more dramatic relative its first six months. framing, and 3 percent used absolute fram The press often does a poor job of report ing. Only 14 percent presented both, which ing on new drugs. The results of a study would give a reader or viewer the most conducted at Harvard Medical School - tools to make a judgment. and funded by The Commonwealth Fund "The distortion of numbers is our and the Harvard Pilgrim Health Care Foun biggest bane for both physicians and dation - suggest that the media coverage of patients," says Dr. Robert Rangno, associ Celebrex was hardly unusual. ate professor of pharmacology and thera The Harvard findings, which have been peutics at the University of British Colum presented at two public conferences, bia. Rangno tries to help physicians inter looked at media coverage of Fosamex, a pret the drugs properly, and his ideas also bone-building drug; pravastatin, a choles apply to journalists: terol-lowering agent; and the use of aspirin • If the long-term consequences of taking for preventing cardiovascular disease. the drug are not known, say so. Look at n a sample of 207 television and news the FDA review documents, which are paper stories about new drugs aired or sometimes on the FDA's Web site. published between 1994 and 1998, the • Pay attention to warnings in FDA news I Harvard scientists found that 53 per releases and to The Medical Letter On cent of them did not mention potential risks Drugs and Therapeutics, a well-respected and side effects. And 61 percent of the sto newsletter. On Celebrex, the Letter said ries that quoted an expert clinician or study short-term studies showed it caused with a financial tie to the drug companies, fewer ulcers than older drugs, but cau failed to reveal that link. tioned that "whether serious gastroin The study pointed out another serious testinal bleeding will occur less frequent deficiency in the coverage - how benefits ly with celecoxib [Celebrexl remains to of the drugs are often framed in the press. be established." For example, the researchers found that • Realistically interpret the benefits and when Fosamex was coming onto the mar risks, and don't rely solely on relative ket in 1996, all three network evening framing. news programs reported that women who • Ask the drug companies about financial took the drug had 50 percent fewer hip ties to expert clinicians who are made fractures, or that the drug cut the risk of available to discuss a new drug. fractures by about half - a figure that -Trudy Liebennan indicates what is called relative risk. Lieberman is health policy editor for Con Relative risk, while accurate, gives only sumer Reports and contributing editor to CJR. CO LUMBIA JOURNALISM REVIEW SEPTEMBER/OCTOBER 1999 .

Load more