Tracy Patton had just arrived at a community theater rehearsal in August 2000 when she felt such a searing explosion in the back of her head that it knocked her to her knees.
A Dose of DenialHow drug makers sought to keep popular cold
and diet remedies on store shelves after
their own study linked them to strokes
By Kevin Sack and Alicia Mundy
At the hospital in Louisville, Ky., doctors said Patton, then 37, had suffered a catastrophic stroke, and they predicted she wouldn't survive the night.
Patton defied the odds. But nearly four years later, she is so overwhelmed by simple tasks that she must post a "personal hygiene checklist" in her bathroom to remind herself to brush her teeth and flush the toilet.
At 15, Tricia Newenham was full of promise when she suffered her stroke in October 2000 while hanging out in her bedroom with a cousin. A Down East Mainer from a family of woodsmen and lobstermen, she had been named her middle school's student of the year and was on track to become the first Newenham to attend college.
She spent a month in a coma, and emerged totally blind and profoundly mentally impaired. When reminiscing about her former self, about her prom dates and nights at the movies, she dissolves into inconsolable sobbing, condemned to remember just enough of what her life was like then to understand how much less it is now.
Only hours before these devastating strokes, each victim had washed down a seemingly innocuous over-the-counter cold medicine, one of billions of doses consumed annually nationwide. The medicines contained phenylpropanolamine, or PPA, the active ingredient in scores of popular nonprescription decongestants and diet aids until November 2000, when the Food and Drug Administration declared PPA unsafe and asked drug companies to stop selling it.
By then, the drug industry had spent more than two decades fending off growing evidence of a possible link between PPA and hemorrhagic stroke. But Patton and Newenham were among hundreds of PPA consumers who suffered attacks after a landmark study -- sponsored by the drug industry itself -- concluded in October 1999 that the use of PPA was associated with an increased risk of that deadliest form of stroke.
Recently obtained internal company documents show that rather than alerting the public during cold season, drug makers launched a yearlong campaign to keep the results quiet and stall government regulation. By the time the FDA acted, 13 months and hundreds of strokes later, the companies had reformulated their brand names with little interruption in sales. The market for PPA has been estimated at $500 million to $1 billion annually.
In the interim, Americans continued to purchase PPA products right off the shelf and assume they were safe.
The Times reviewed thousands of pages of documents produced through discovery in PPA lawsuits and obtained from the FDA through a Freedom of Information Act request. The documents demonstrate that the pharmaceutical industry consistently challenged any notion that PPA could be dangerous and dismissed evidence to the contrary. They also show that the manufacturers assured the public that PPA was safe even as some FDA scientists and industry officials were raising concerns.
As early as 1982, an FDA report warned that PPA had "the ability to cause cardiovascular effects, cerebral hemorrhage and cardiac arrhythmias." Two years later, a memo from the medical services department at Sandoz Pharmaceuticals, which made the PPA products Triaminic and Tavist-D, referred to PPA as "an agent known to cause hypertension and stroke."
Yet the drug companies accelerated their marketing of PPA, winning FDA approval to sell prescription PPA products on an over-the-counter basis and introducing flavorful new formulas for children.
Upon learning that the 1999 study had found a stroke link, the drug makers opened a relentless assault on its methodology and on the integrity of the Yale University researchers who conducted it. They did so despite having paid for the five-year, $5-million study themselves, approving its protocol and handpicking investigators who had previously expressed skepticism about a link between PPA and stroke.
Some documents show that the companies hoped to survive the 2000 cold season without pulling PPA products. Rarely do the internal memos indicate concern by corporate officials that PPA might pose a threat to the public.
Early in November 2000, for instance, two weeks after an FDA advisory panel concluded that PPA could be hazardous, an official with Bayer, which made Alka-Seltzer Plus with PPA, drafted a proposed "PPA Crisis Action Plan." Its stated objectives: "Delay mandatory implementation of FDA recommendation. Blunt PR impact by highlighting questionable study conclusions as they pertain to cough/cold products. Begin shipping PPA-free product as soon as possible."
Terry O. Tottenham, a lawyer for Bayer, said the plan was not implemented. But records and interviews show the industry largely followed that course.
The FDA eventually recommended the withdrawal of more than 100 PPA products, including popular cough and cold brands such as Robitussin CF and Dimetapp, and appetite suppressants such as Dexatrim and Acutrim. FDA officials said they did not move faster because the industry's efforts to discredit the Yale results effectively delayed the delivery of a final report.
"There were obvious concerns that we weren't getting the data because it was being held up by the people who sponsored the study," said Dr. Charles J. Ganley, director of the FDA's Division of Over-the-Counter Drug Products.
Left in the DarkOnce the FDA stepped in, the manufacturers issued press releases, posted notices on websites and wrote letters to doctors, pharmacists and retailers advising them of the agency's action. But after the products were withdrawn, neither the companies nor the FDA mounted major advertising or direct-mail campaigns to warn Americans they might have dangerous products in their handbags and homes.
A survey commissioned by plaintiffs in lawsuits over PPA estimated that 3.5 million U.S. households still possessed PPA formulations a full 15 months after the withdrawals were requested in November 2000.
The wife of a Mississippi stroke victim says she purchased Alka-Seltzer Plus with PPA from a convenience store in April 2001, six months after it was supposed to have been removed. Her 45-year-old husband took two packets over two days, began convulsing, and now is confined to a wheelchair, according to his lawyer. Both the retailer and the wholesaler have testified in depositions that they were never alerted by Bayer or the FDA. The couple are suing Bayer and Double Quick Inc., the retailer.
Even now, the industry's attacks on the study it commissioned are its primary defense against more than 2,500 lawsuits filed by plaintiffs who say they suffered strokes shortly after taking products with PPA.
Spokespersons for each of the major manufacturers of PPA products said in interviews or written responses that they continued to believe PPA was safe, despite the study's findings.
"We did not believe then, nor do we believe now, that PPA was dangerous," wrote Tottenham, the Bayer lawyer. "In fact, we believe the PPA in those medications was safe and effective and did not, in any way, cause or contribute to hemorrhagic stroke or any other circulatory disease, when used as directed." Tottenham said that there was "no valid scientific evidence" of an association and that the Yale study, when "properly analyzed, does nothing to change this conclusion."
The Consumer Healthcare Products Assn. (CHPA), the leading nonprescription drug trade group, declined to comment for this article.
But its former longtime president, James D. Cope, who took a previously scheduled retirement shortly after the study's completion, said the drug companies would have been well-advised to accept the findings. "Industry was convinced that if a proper study would be done, the results would come out where they wanted them: safe and effective," he said. "It didn't. And since it didn't, and they designed the best study they could, I think they have to live with it."
The manufacturers include some of the largest drug companies in the world -- Bayer; Novartis, which absorbed Sandoz in a merger; Wyeth, which makes Dimetapp and Robitussin CF; GlaxoSmithKline, which makes Contac; and the former Thompson Medical Co., which made Dexatrim until Chattem Inc. bought the line late in 1998.
Doctors and scientists cannot say for sure that PPA caused Patton, Newenham or any other victim to have a stroke. Many of those who suffered strokes after taking PPA were particularly susceptible to blood pressure spikes because they suffered from one of several conditions affecting tens of millions of Americans.
Some had hypertension, which afflicts one in five Americans. Patton was among the estimated 3% of Americans with cerebral aneurysms, which can pop if blood pressure rises suddenly. Newenham had an arteriovenous malformation, a circulatory defect that can make vessels prone to rupture. Neither Patton nor Newenham knew about their conditions before their strokes.
If PPA had been a lifesaving drug, the benefits might have more clearly outweighed the risks to such a large segment of the population. But though it unclogged countless stuffy noses and helped millions of dieters shed a few pounds, PPA was neither essential nor irreplaceable.
What made it worth fighting for were its sales, estimated at several billion doses a year. In a deposition, Robert G. Donovan, the former head of Sandoz Consumer Health Care, said the profitability of the company's two PPA products was about 75% to 80% of revenue.
Years before the Yale Hemorrhagic Stroke Project, clinical studies and individual cases published in medical journals began to raise concerns about PPA.
Sandoz was among several companies that considered reformulating their cold products with comparably effective ingredients, primarily pseudoephedrine, which had been used safely for years and gained final FDA approval in 1994. But pseudoephedrine cost more than PPA. And it had a bitter taste that was more difficult to mask, an important consideration for the pediatric market. The manufacturers concluded there was no reason to switch.
"There appears to be little, if any, upside business potential in reformulating from both a consumer and professional standpoint," wrote the marketing director of pediatric cough and cold products for Sandoz in a 1988 memo included in litigation records.
"Few consumers are aware of [over-the-counter] cough/cold products' active ingredients," he wrote. "Fewer still would be aware of any safety issues with PPA."
In depositions, many industry officials have tried to shift responsibility to the FDA.
"My assumption was that if there was an issue of safety, supported by sound evidence, that the Food and Drug Administration would exercise their responsibility and take the product off the market," Donovan testified. But FDA officials said that until they received the final results of the industry study, there was not sufficient evidence of PPA's dangers to take it off the market or demand prominent stroke warnings on labels.
The agency does not require manufacturers to report cases of adverse reactions to certain drugs, like PPA, that have long been available over the counter. That makes it virtually impossible to track potentially dangerous trends as they develop.
Some companies provide such data voluntarily, but research has shown that underreporting is widespread.
As a result, there are no reliable figures for how many strokes may have been associated with PPA use. But in 2000, FDA epidemiologists estimated that between 200 and 500 hemorrhagic strokes a year could be attributed to PPA in people 18 to 49.
Although products with PPA are no longer for sale, Sen. Ron Wyden, an Oregon Democrat who held hearings on PPA safety as a congressman in 1990, said in an interview that he hoped FDA officials would learn from the PPA experience.
"In 1990, there was already essentially a decade of evidence that PPA could be causing health problems, and it was 10 years after I opened up these hearings that PPA came off the shelf," he said. "There are real human consequences of slow, stalling tactics. Loved ones don't come back, disabilities don't disappear, just because PPA is off the shelf after years of foot-dragging."
The 'Big Soldier'Tracy Patton could never have imagined such a tentative life.
Though only 5 feet 2, she was the first player off the bench for her Scottsburg, Ind., high school basketball team when it went to the state semifinals. She held the town's high school long jump record -- 16 feet, 8 inches -- and won a track and field scholarship to college. Before her stroke, she mowed the yards at seven of her father's rental properties every week.
Whenever Patton faced adversity as a child, when she skinned a knee or was dragged to the dentist, her father would tell her to "be a big soldier," and she took his instruction to heart. Within her family and in her job as the director of a foster care agency, she had a reputation for determination and grit.
Patton says that on the drive to nearby Louisville for her play rehearsal that night, she took a single tablet of Tavist-D to unclog her sinuses. She hemorrhaged an hour later.
At the hospital, Patton's doctor told family members to expect the worst, and advised them to say their goodbyes. "She'll probably bleed out," he said. "We'll try to make her as comfortable as we can." To their astonishment, just before sunrise, she began to wake from her coma.
The surgeons cut a scythe-shaped incision in her head, repaired her aneurysm with two metal clips and inserted three metal plates in her skull. She lost 85% of the vision in her left eye, leaving her to view the world as if through Vaseline-smeared glasses.
These days, Patton struggles with simple math and the concept of time. She keeps a printed inventory of her own odd behaviors: "I have been found talking to pictures on the wall. I have 'thanked' myself after I have poured myself a drink." One day, she thought she saw a reindeer driving a car.
Patton, who is single, tries to present a patina of normalcy, but her daily routine taxes her strength. She has recurring nightmares, often about brains. "Shopping for brains, brains of all shapes and colors and sizes, brains in jars," she said. "And then there's brain surgery. They want me to go in and look at other people having brain surgery. I wake up in a panic."
Patton still holds down her job, where her office computer has been programmed with large-print type. But she acknowledges she is able to work only because a colleague has assumed many of her responsibilities. "She does it all and acts like I did it," Patton said. "We hide it really well."
On the day she came home from the hospital, a month after her stroke, Patton celebrated her 38th birthday with family and friends. She held an ice pack to her head as her sister, Kim, read her cards aloud and handed her gifts -- a scented candle, a collectible race car, and, most utilitarian, a collection of hats. Then Kim presented the cake she had baked in the shape of a brain. "Can you imagine how many times it took me to make gray icing?" she asked with a laugh.
But the day's most meaningful gift came from her father. It was a simple sliver of silver, and he had had it engraved: "#1 Big Soldier."
Hidden DangersResearchers estimate that hemorrhagic strokes, which occur when blood vessels rupture and bleed into the brain, kill more than a third of all victims within a month and leave more than a third of survivors severely disabled. They account for only 12% to 17% of the 700,000 strokes in the U.S. each year, according to the American Stroke Assn., but play a disproportionate role in making stroke America's third leading killer.
Phenylpropanolamine, which was first synthesized around 1910, is one of a class of drugs that stimulate the sympathetic nervous system, not unlike amphetamine and cocaine. It also can constrict blood vessels and increase the force of heart contractions. By narrowing blood vessels in nasal mucous membranes, PPA allows air passages to open up. The very same mechanism can, according to the FDA, produce transient spikes in blood pressure.
Whether PPA can raise blood pressure to dangerous levels has been the subject of extensive debate. After dozens of studies and case reports, the cumulative weight of scientific evidence suggests that PPA is not necessarily dangerous to everybody, but that it can trigger lethal reactions in some
Studies found that the degree of blood pressure elevation attributable to PPA depended on the type and quantity of product taken, whether it was taken in combination with stimulants like caffeine, whether the patient was otherwise susceptible to high blood pressure, and whether it was consumed while upright or lying down, when the impact apparently is greatest.
Of particular concern was that PPA could produce adverse effects in quantities that were only two or three times the FDA's recommended dosing limits -- 150 milligrams per day for decongestants and 75 milligrams per day for diet drugs.
The FDA takes the position that over-the-counter drugs, while not risk-free, should be "relatively hard to get in trouble with," said Dr. Robert J. Temple, the agency's associate director for medical policy.
Some researchers, many of them bankrolled by drug manufacturers, have held that PPA has minimal effects on blood pressure. Most prominent is Dr. George L. Blackburn, whose chair in nutrition medicine at Harvard was endowed by the founder of Thompson Medical Co., which created Dexatrim. Another is Dr. John P. Morgan, who was Thompson Medical's part-time medical director and who defended PPA in a 1985 text that was heavily underwritten by the company.
Morgan acknowledged in an interview that he was paid "an enormous amount of money" over the years by Thompson. "I know some people thought that influenced the research, but that was not the case," he said. He added that he still did not believe PPA caused strokes. "But I've always been afraid I'd be proved wrong," he said. "Maybe it does."
When the FDA began evaluating PPA in the 1970s, the agency's expert panels recommended it be categorized as safe and effective, while also noting reports of "idiosyncratic reactions of central nervous system stimulation and/or blood pressure rise." There was enough uncertainty about safety to keep the agency from issuing a final ruling, placing PPA in a regulatory limbo that allowed its continued marketing.
As the body of research about PPA's potential hazards expanded, the industry maintained there was no health risk. In 1989, the Proprietary Assn. -- a forerunner of the CHPA -- declared in a statement that PPA was safe, and asserted that "clinically recommended doses of PPA produce no clinically significant changes in blood pressure, heart rate or EKG."
But even as the companies were publicly defending PPA, one internal report after the next referred to its potential dangers.
In 1989, for instance, a manager in the Sandoz medical services department wrote that both PPA and pseudoephedrine were "viable for use" in over-the-counter products but that "each have had dire outcomes in small doses." The manager then added: "It isn't only abuse or overdoses which cause problems. Adverse effects are rare but can be serious."
By 1996, a confidential Sandoz "safety update" warned: "It is conceivable that the intake of drugs containing this active ingredient predisposes to or even causes cerebrovascular accidents."
Some manufacturers had begun exploring the possibility of reformulating their brands with other ingredients in the early 1980s. But at Dorsey Laboratories, a division of Sandoz, officials calculated in 1983 that a switch to pseudoephedrine, which was more than twice as expensive as PPA, would cost $1.4 million. Furthermore, a company marketing survey concluded that awareness of PPA's possible side effects was not yet widespread among physicians and pharmacists.
"Based on the reaction of these professionals, reformulation to pseudoephedrine is not an urgent matter," wrote a staffer in Dorsey's marketing research division.
Birth of a StudyIn 1990, prompted by reports of PPA's dangers and its growing use in diet drugs, then-Rep. Wyden held his hearings. Some of the testimony was devastating.
After enumerating a laundry list of ailments, including cerebral hemorrhage, that had been associated with PPA in medical literature, Dr. Thaddeus E. Prout, then the chairman of medicine at Greater Baltimore Medical Center, issued a challenge: "I defy anyone to find another unregulated drug that has such a record of disaster." Prout, an authority on drugs, blamed the FDA for not moving against PPA. "Thousands of people will suffer as a result of our negligence," he predicted.
From the FDA's vantage point, the only way to truly determine whether PPA users were at disproportionate risk of stroke was to conduct an extensive epidemiologic study. But since the agency does not itself sponsor research for drug reviews, it had to rely on PPA manufacturers to investigate the safety of their own products.
There was incentive for the companies as well. A long-term study would keep PPA on the market, at least temporarily.
The industry sponsors -- Thompson Medical and Ciba Consumer Pharmaceuticals (then the maker of the diet drug Acutrim) -- were involved in all major methodological decisions, as was the CHPA. They chose investigators who were highly credible with the FDA, but who also were known to be skeptical of any connection between PPA and stroke.
Lawrence M. Brass, the Yale neurologist initially enlisted for the study, said in an interview that he told the industry sponsors in an early meeting that he "really didn't see a lot of evidence for an association."
Similarly, the sponsors approved Dr. Louis C. Lasagna, who had endorsed PPA's safety in a 1988 textbook, as chairman of the study's three-member oversight committee.
In April 1994, as the investigators and the companies applied the finishing touches to the study's design, Timothy R. Dring, Ciba's assistant director of regulatory affairs, wrote to one of the Yale researchers that the protocol "will serve our research purposes quite admirably."
At the same time, Dexatrim's maker, Thompson Medical, broke with the rest of the industry by disclosing on labels that there had been reports that stroke and other conditions "might be associated" with PPA.
It seemed a purely defensive move. Thompson's president, Daniel N. Horwitz, stressed in a letter to the FDA that the company "in no way believes that these reports are correct." And CHPA officials made it clear that they disagreed with the decision.
The companies not only continued to sell their products, but also introduced new PPA brands, like Bayer's Alka-Seltzer Plus children's cold medicine with "fizzy fresh cherry taste."
Unexpected FindingsTo the astonishment of the drug companies, the study conducted by its handpicked researchers using their industry-approved protocol produced bombshell results.
An examination of 702 stroke cases in people 18 to 49 years old identified 27 victims, mostly women, who had taken PPA shortly before their attacks. Their experience was compared to a control group of 1,376 people.
The most significant finding was that women who had taken an appetite suppressant with PPA were 16 times more likely to have a stroke within three days than those in a control group who had not taken any.
The study also found that the risk of stroke was three times greater for women who had used PPA products for the first time in the last 24 hours. "First use" meant that the subject had not taken a PPA product in the previous two weeks. All the first-users in the study had taken PPA cough and cold remedies.
In their final report to the FDA, the investigators concluded that "the association of PPA with risk for hemorrhagic stroke is present for both customary indications for PPA (as a cough-cold remedy and an appetite suppressant)."
For all subjects -- regardless of gender, the type of product taken or the timing of ingestion -- the risk of stroke increased by 50%. But that finding, like some others, fell short of statistical significance.
Upon learning the results on Oct. 17, 1999, the members of the study's oversight committee -- including Lasagna, who had previously written so confidently about PPA's safety -- unanimously approved the findings and instructed the researchers to notify the FDA immediately.
The next day they did, placing telephone calls to officials at the agency and to the leaders of the trade association. Industry officials feared that the study would prompt FDA action against all forms of PPA, and began mounting a counteroffensive.
Even before the Yale results were disclosed, the industry had started considering worst-case scenarios. "We need to be thinking offense here not just defense," wrote John Incledon, vice president of the respiratory business unit of Wyeth's Whitehall-Robins Healthcare division, in an e-mail to a colleague on Oct. 6, 1999. "The timing of this is ideal for news stations to pick up on it in droves. It's Yale, it's cold season and it's in children's products."
Once the study was finished, the FDA asked the doctors to summarize their findings in a letter. Dr. R. William Soller, then the senior vice president of the CHPA, wrote to one of the researchers to insist that any report to the FDA stress PPA's overall safety record. The next day, he wrote again, this time warning that the Yale investigators may have violated their contract by communicating results to the FDA before fully vetting them with the industry sponsors. Two days later, a lawyer for Novartis accused the Yale doctors in a letter of committing "a serious breach" by prematurely disclosing the results.
The doctors were taken aback. "We contend that we were in full compliance with the contract," Dr. Walter N. Kernan, one of the Yale researchers, said in an interview. "Neither the university nor the investigators were going to allow anything to get between us and what we perceived as the best interest of the public."
The FDA, meanwhile, told the researchers that any communication with the agency would have to be public. It quickly scheduled an open meeting of its Nonprescription Drugs Advisory Committee, a panel of experts whose recommendations were almost always endorsed by the FDA commissioner.
But the doctors said their data needed further refining and analysis before public release. And the industry told the FDA it could not prepare a public defense so quickly.
In the end, the panel's meeting was pushed off by more than 10 months, with much of that time marked by parrying between the Yale doctors and the industry over the research methods they had agreed to five years earlier. In the meantime, the public remained unaware of the study's findings, and products with PPA stayed on the market.
"I think the companies wished that we'd never reach the final stage," said Dr. Ralph I. Horwitz, a clinical epidemiologist who helped lead the Yale team, "because the longer the process was underway, they were able to avoid finally dealing with the consequences of the research."
Horwitz, who is now dean of the Medical School at Case Western Reserve University, said the delay left issues of consumer safety in suspension. "I do think that had they been able to see the study findings as soon as the data were available, that the FDA would have felt obliged to issue a warning in the interest of public health and safety," he said.
The FDA's Ganley said the agency could not act without a final report. "Why would we put out something like that when we didn't really have the data in hand?" he asked. "We would be criticized, and justifiably so."
Circling the WagonsAs the clock ticked toward some form of FDA regulation, the companies enlisted a battalion of consultants to analyze the study. They hired a public relations firm, Ruder Finn, which suggested in a draft memo that the industry argue that PPA-related jumps in blood pressure were "within the range of increases associated with routine daily activities, such as climbing stairs or mowing the lawn," according to correspondence from CHPA's Totman.
One Bayer memo about a meeting of an industrywide PPA task force said it had been noted that the FDA's Temple had an unfavorable position on PPA and that "efforts should be made to steer the media away from him."
At Chattanooga, Tenn.-based Chattem, which had just bought Dexatrim, President A. Alexander Taylor II wrote to Soller that he feared "an avalanche of negative publicity" if the study were to become public before it could be reviewed by the industry sponsors.
Taylor appealed to his home-state senator, Tennessee Republican Bill Frist, a physician, to intervene with the FDA. Taylor began a Feb. 4, 2000, letter by mentioning that a prominent Frist donor served on the Chattem board of directors and that the company's chief executive officer was the brother of a Frist county reelection chairman. "I am a very proud contributor to your current campaign," Taylor continued.
"A few more months of study before any public release of this data will not harm the public health, and may benefit it," he wrote.
Frist never responded. "It fell on deaf ears," Taylor said in a deposition.
As the months passed, the companies saw that it was time to reformulate.
In December 1999, Novartis had kick-started its efforts to remake Triaminic with pseudoephedrine. Wyeth also set a quick timetable for reconstituting Dimetapp, then slowed down as the FDA postponed its advisory committee meeting.
"It appears that we have a bit more time than we originally expected on a decision from the FDA on PPA," wrote Wyeth's senior product manager for Dimetapp in an April 2000 memo. "By launching in January 2001 instead of September 2000 (the peak of the cough/cold season), we will be able to better manage open stock and display inventories."
In May, a draft memo from Wyeth's Whitehall-Robins division addressed whether retailers should return stocks of Robitussin CF when they began receiving shipments of PPA-free replacements. The answer: "NO. Again, the decision to reformulate was voluntary. Therefore, current stock is perfectly safe and effective for use. We will not be accepting returned product for the new formula."
David E. Dukes, a lawyer for Wyeth, stressed in an interview that the companies and the FDA were still evaluating the Yale study at that point. "At the senior levels, where decisions were being made about Dimetapp and Robitussin CF, safety was the primary concern," he said.
Meanwhile, even the consultants brought in by industry found it difficult to dismiss the study's conclusions.
Dr. James Lewis, a University of Pennsylvania epidemiologist who had been commissioned by Bayer, wrote that the Yale researchers showed "fairly convincing evidence of an association between PPA use and the risk of hemorrhagic stroke."
Like other consultants hired to critique the study, however, he also guided his client toward a strategy for attacking the fragility of the findings. With such a small number of PPA-related stroke cases, he wrote, any methodological bias could affect the statistical significance of a result.
The industry seized on that approach. "We came out and said we don't like our own study," said Cope, the former CHPA president. "It's predictable: If the results don't come out the way you like, since there's no such thing as a bulletproof study, you point out the weaknesses in it."
Under increasing pressure from the FDA, the Yale doctors submitted their final report on May 10, 2000. Though they stopped short of saying PPA could cause hemorrhagic stroke, they later emphasized that "causation is one explanation for that association."
The FDA staff quickly sided with the researchers. An FDA statistician praised the study as one of the best he had reviewed in the last decade. A pair of FDA epidemiologists were so persuaded that they took the unusual step of recommending that PPA-containing appetite suppressants be banned as over-the-counter products.
"The study demonstrated a statistically significant increased risk of hemorrhagic stroke among both appetite suppressant users and first time users of PPA as a cough/cold remedy," their report concluded. The recommendation for a ban infuriated industry officials, who complained it would unfairly sway the advisory panel, now scheduled to meet on Oct. 19.
At the advisory committee meeting, Dr. Noel S. Weiss, a University of Washington epidemiologist who had been hired by the CHPA, honed in on the disparity in the study's findings for different types of products. "Why an association with appetite suppressant drugs and not for colds and such when the typical doses given for colds are higher than for appetite suppressants?" he asked.
But the attack on the study was so broad that it ultimately undercut the industry's cause, FDA officials said. "They essentially trashed the study, even though they are the ones who designed it and financed it," Ganley said.
The advisory panel voted unanimously, with a few abstentions, that PPA could not be considered safe. A few weeks later, the agency told manufacturers it expected to classify PPA as not safe for over-the-counter use and asked that they voluntarily discontinue marketing the products. The evidence of an association, while not conclusive, had been persuasive.
"If someone wanted you to swear it must be true, I wouldn't do that," said the FDA's Temple. "But we thought it was enough. Look, if the drug were lifesaving, if it was something of immense value and had no replacements, you'd have to think about it. But it wasn't."
Stolen YouthTricia Newenham suffered her stroke the week before the advisory committee met. As her neurosurgeon prepared to remove the damaged tissue from her brain, he warned her parents she might not survive her eight hours on the table. Three days later, before a second operation to remove a blood clot, her mother had her baptized, just in case.
Tricia made it through, but she would never be the same.
Before her stroke, she stood a gangly 5 feet 6 and weighed 106 pounds. She was becoming a woman, and had developed a serious case of the boycrazies (login: kissable98; password: puckerup). She never left her house in Steuben, Maine, without her hair and makeup just right. She had given up basketball and softball and swimming in vanity over her skinny legs.
Now 18, Newenham weighs 196 pounds, the consequence of medication and a captive life in which junk food provides a rare escape. Once a promising artist who had imagined a career in design, she passes the time rocking gently in a blue recliner, listening to the drone of soap operas.
Newenham can brush her teeth and wash her hands, but she does not always know when to stop. Clinging fast to her femininity, she insists on shaving her own legs, but leaves them scabby with cuts.
She hears a lot and comprehends much, but she has difficulty processing thought and even more in communicating it. When asked a question, she struggles to find the most economical answer, usually a word or two, often a guttural "I don't know," with little intonation.
Is she in physical pain? "No." Is she angry? "No." Is she sad? "No." Is she frustrated? "Yeah."
What are her favorite things to do? "Sleep and eat." What does she like to eat? "Chop suey." What else? "I don't know."
What does she remember about her old life? "Boys."
Is she content or bored with her life? "I get bored." What does she wish she were doing? "I don't know. Going out with friends."
And then she begins to cry. And cry and cry.
While Newenham lingered in a coma after her stroke, the doctors advised her parents to consider a nursing home. Even if Tricia came to, she would not be the daughter they had known.
But ever so slowly, Tricia started waking up. First, a tear would form in the crease of her eye. Sometimes, when her mother talked to her, Tricia would reward her with a muffled, choking cry.
Patricia Bybee told the doctors she was taking her daughter home. "I just sensed that she was in there," she said. Two months later, Tricia had spoken her first word: "M-m-m-om." Three months after that, she could walk.
When the paramedics came to the house on the day of the stroke, they asked Bybee what medications Tricia had taken. "Only Triaminic," she said. It was months later, after the FDA had asked drug companies to withdraw their PPA products, that she learned of the possible connection between cold medicines and stroke.
Bybee, a nurse's assistant, and her husband, Tim, a plumber, went deep into debt to pay off medical bills. They say a settlement reached last year with Novartis will relieve their debts and cover the cost of Tricia's attendance at a school for the blind near Boston.
After three years, they have slowly adapted to their new reality, and lowered their expectations.
"We would give anything to have her back the way she was," said Tim Bybee. "But we have grown to love her and accept her the way she is."
Softening the Blow The withdrawal of PPA was a tough loss for the companies, but they had succeeded brilliantly in minimizing the impact. The study's long duration bought five years to market PPA products without restriction. And the post-study delay provided time to formulate products without PPA, which began shipping within days of the withdrawal.
Throughout, the manufacturers remained defiant.
As late as Oct. 25, 2000, six days after the FDA advisory committee voted that PPA was unsafe, Novartis' director of medical affairs sought to reassure doctors about the safety of PPA, particularly for children, who he pointed out were not included in the Yale study.
"As you know, hemorrhagic stroke is exceedingly rare in the pediatric population," Dr. Geoffrey Ross wrote.
"Particular to the pediatric population, PPA plays a significant role in reducing missed school days and in promoting effective symptom relief."
The researchers submitted a scientific article on their PPA study to the New England Journal of Medicine, reporting a statistically significant risk of stroke for women taking PPA diet drugs and a possible association between decongestant use and stroke in women. The journal considered the findings important enough to post the article on its website Nov. 6, six weeks before it came out in print.
The same day, the FDA issued a public health advisory that consumers should not use PPA products. Health officials in a number of countries, from Canada to Malaysia, quickly followed suit.
Finding herself stuck with millions of dollars' worth of useless pills Chattem's Dexatrim brand manager suggested to a colleague in an e-mail -- produced by the company for litigation -- that they sell the inventory to a sampling and promotion company known as Box of Brands.
Andrea M. Crouch, the vice president of toiletries marketing, responded enthusiastically. "I think using Box of Brands is an excellent idea!" she wrote. "I see no downside -- so what if they divert! I will be surprised if they don't have a problem with PPA but let's go for it!"
Crews Townsend, a lawyer for Chattem, called the proposal "part of the brainstorming process" and said the company never sold its Dexatrim inventory to Box of Brands. Instead, he said, it was destroyed.
At Wyeth, gallows humor set in. When a speech was drafted for an executive to deliver at an awards banquet several weeks before Christmas 2000, it included a line about the need to dispose of large quantities of Dimetapp and Robitussin CF: "When you're sorting this year's holiday gift list by 'naughty and nice,' don't think 'lump of coal,' think 'product with PPA.' " By the time a final draft had been prepared, that opening had been replaced by an off-color joke about the 2000 presidential campaign, according to a copy provided by lawyers for Wyeth.
Virtually all of the firms took one-time accounting losses because of the withdrawals -- $80 million for Wyeth, $54 million for Bayer, $50 million for Novartis and $8 million for Chattem, according to corporate filings and statements.
But for the most part, the losses were short-lived. Thanks to the quick reformulation and shipment of PPA-free Dimetapp, domestic sales of the product rose 50% in the first quarter of 2001. Meanwhile, 2001 sales of Dexatrim Natural, the non-PPA version of Chattem's diet aid, "more than made up for lost sales" due to the discontinuation of Dexatrim with PPA, according to an annual report.
The Battle ContinuesChattem announced in December that it would seek to settle most of its 332 PPA lawsuits. It had already agreed last year to pay $3.5 million to Jennifer Villarreal, a Texas hairstylist who collapsed at a gym on March 6, 2000, allegedly after taking a Dexatrim pill.
The company also said it had reached settlements with insurers who alleged in lawsuits that Chattem had applied for liability policies without disclosing the Yale results. Townsend, the Chattem lawyer, said the settlements did not indicate that the company had changed its belief that Dexatrim with PPA was a safe and effective product.
Novartis agreed to make payments to Tricia Newenham and her family last year on the condition that the terms remained confidential, according to her mother. Novartis would not answer questions about Newenham, and neither Novartis nor the other companies would disclose details about other settlements.
Thousands of plaintiffs, including Tracy Patton, have lawsuits pending against PPA drug makers, but only a few of those cases have made it to trial. This year, juries in New Jersey and California rejected claims by plaintiffs that Novartis products had caused their strokes. In both trials, the victims had had a variety of risk factors for stroke, and their strokes occurred before the Yale study was completed.
"Our company is extremely gratified that two diverse juries, from opposite ends of the country, reached the same conclusions after an exhaustive airing of all the evidence," said Nancy Fitzsimmons, vice president for global communications of Novartis Consumer Health.
The industry's efforts to discredit the Yale study have continued in court, with defense lawyers trying to poke holes in enough of the study's PPA-related stroke cases to negate its findings.
The dispute has become increasingly personal. Drug company lawyers have hypothesized a conspiracy that has the Yale doctors manipulating data to strengthen their chances of finding an association. Such a finding, the industry lawyers suggest, would help the researchers win promotions and publication in a top journal.
"Those investigators were very vulnerable to human frailty, to human desires and human error; driven by a desire to succeed, to obtain recognition," said Jan E. Dodd, a lawyer for Novartis, in her closing argument in the California trial.
In constructing their theory, defense lawyers have focused on a midcourse shift by the researchers in defining when a stroke begins. In one case, the investigators set the onset of a stroke one hour after the patient had taken PPA, even though the patient had begun suffering a headache -- potentially an early sign of stroke -- six days earlier. The companies have made much of a 1998 e-mail sent by Yale researcher Kernan to his colleagues saying the strategy of shifting the time of onset "effectively increases our likelihood of finding an association" between PPA and hemorrhagic stroke.
Kernan said in an interview that he was just stating the obvious, and not suggesting that data could be tweaked to produce a desired result.
"Sloppy research does not get you promoted at Yale," he said. "Lack of integrity gets you fired. The investigators had no stake in the outcome of the research."
Brass, the Yale neurologist, joked that given the drug industry's financial support for friendly scientists, the researchers would have had more incentive to skew the study in PPA's favor. "If we were thinking payoff, we should have shown PPA as safe," he said. "I would have had a chair. Yale would have a whole division dedicated to the study of obesity. But we don't fix studies. We just do them."
Some in academic medicine worry that the industry's attacks on the Yale doctors might discourage researchers from pursuing similar studies. "With the amount of hassle and harassment that they had to endure, I'm sure the next time they're asked to undertake something like this, they'll wonder if it's worth the cost," said former FDA Commissioner David A. Kessler, now the dean of medicine at UC San Francisco.
The Yale doctors, said Horwitz, clearly were naive about the industry's willingness to assail their integrity. "I love arguing about the science," he said, "but this is outrageous, especially considering who was making the accusations. It was their study."
Kevin Sack is a Times staff writer based in Atlanta; Alicia Mundy is a special correspondent based in Virginia.
Researcher Janet Lundblad in Los Angeles contributed to this report.
Built for Speed?
The Good Drug Guide
MDMA: Utopian Pharmacology
Ephedra in Herbal Supplements