The Panic Virus

by Seth Mnookin; Dan B. Miller

  41 The Times did run a correction in which the paper acknowledged that the story’s subhead “erred in saying of a possible link that Dr. Halsey ‘thinks it’s an issue worth investigating.’ ”

  CHAPTER 14

  MARK GEIER, WITNESS FOR HIRE

  The week of May 10, 2004, word began to leak out of Washington that the Institute of Medicine committee that had been asked to conduct a broad “immunization safety review” three years earlier had completed its final report. After hearing rumors that the CDC was “preparing [itself] for a bomb,” Sallie Bernard, Liz Birt, Lyn Redwood, and some of the other SafeMinds stalwarts set to work preparing three separate statements: one if the committee recognized a link between vaccines and autism, one if it did not, and one if it split the difference.

  On Tuesday, May 18, at eight a.m., the 199-page report was released. It began:

  This eighth and final report of the Immunization Safety Review Committee examines the hypothesis that vaccines, specifically the measles-mumps-rubella (MMR) vaccine and thimerosal-containing vaccines, are causally associated with autism. The committee reviewed the extant published and unpublished epidemiological studies regarding causality and studies of potential biologic mechanisms by which these immunizations might cause autism. The committee concludes that the body of epidemiological evidence favors rejection of a causal relationship between the MMR vaccine and autism. The committee also concludes that the body of epidemiological evidence favors rejection of a causal relationship between thimerosal-containing vaccines and autism. . . .

  In addition, the committee recommends that available funding for autism research be channeled to the most promising areas. The committee makes additional recommendations regarding surveillance and epidemiological research, clinical studies, and communication related to these vaccine safety concerns.

  There was no question which of its prepared responses Safe-Minds would be releasing to the media. “The IOM has not only compromised its integrity and independence but also failed the American public,” their statement read. The “flawed, incomplete” report “put America’s children at risk” and “violat[ed] nearly every tenet of medical science”; the only explanation for the shoddy work of the unsalaried thirteen-person committee—which included the director of Brown University’s Center for Statistical Science, the co-director of Yale University’s Center for the Study of Learning and Attention, and the chair of the Department of Family Medicine at the University of Washington’s School of Medicine—was that its members had been “bought and paid for by the CDC.” The press gave the organization plenty of opportunities to get this message out to the public: Redwood served as the report’s main critic in stories by the Associated Press, The Boston Globe, The Hartford Courant, the San Francisco Chronicle, and the Scripps Howard News Service; Mark Blaxill, a Boston-based consultant who was also one of SafeMinds’ leaders, talked to The New York Times, the Orlando Sentinel, and The Washington Times; and Bernard was quoted in a Reuters story. Reporters also reached out to Barbara Loe Fisher, who pointed to the breadth of the committee’s work as a reason to distrust its conclusions. “This report went beyond any other report, and this is why I felt it was political,” she said. It was a perfect example, she explained, of why only research done by scientists not involved with the “public health community” could be relied on to produce honest work.

  Fisher’s statement reinforced what the parents in SafeMinds had been unintentionally acknowledging with their reactions to any article or study that ran counter to their theory: They had never for a moment doubted that vaccines had caused their children’s illnesses. Instead of trying to collaborate with the scientific community, their real goal was to get officials to admit they’d been wrong—and when they didn’t, that refusal only served to confirm their suspicions of a broad, international conspiracy.

  With compromise no longer an option, the war on any organization, institution, or individual that was even tangentially involved with vaccines or vaccine safety became increasingly personal. The attacks were particularly vicious toward Marie McCormick, a professor of maternal and child health at the Harvard School of Public Health, who had served as the IOM committee’s chairperson.

  McCormick was an unlikely bête noire for SafeMinds members and their allies. At the outset of her tenure as the committee’s chair, she’d had serious reservations about what she’d perceived as the hasty way the country’s vaccine program had been expanded in the 1990s. “I don’t think it was well described to the general population,” McCormick says. “I’m not even sure it was well described to physicians. I remember the grand rounds on hep B, thinking, You don’t have the evidence for what you’re talking about—that it was going to prevent STDs. The vaccine hadn’t been around long enough to prove that.” And unlike many of her peers, McCormick did not have a knee-jerk aversion to Andrew Wakefield’s work; in fact, the ideas behind his 1998 MMR paper had intrigued her when it first came out. “I’ve seen what [autism] looks like, and I thought, Isn’t it interesting that someone is finally taking seriously the concerns of these families about the sort of functional problems that these kids have. I also work in a developmental clinic for premature infants, and often the most distressing part of dealing with these kids are things like feeding disorders or bowel disorders. . . . I thought it was nice that someone was finally paying attention.”

  Despite her open-mindedness, soon after the committee was formed, SafeMinds members began to refer to McCormick as “Church Lady,” a reference to the sanctimonious Saturday Night Live character played by Dana Carvey in the 1980s. At one point, Liz Birt was given a voodoo doll of McCormick so she could stick it “full of pins.” Another time, after McCormick had told reporters that thimerosal had not been “proven to be dangerous,” Lyn Redwood wrote in an e-mail, “I am out for blood here. It takes a lot to get me really pissed off, and she has done it.”

  Ironically, when the committee was being assembled, the IOM’s strenuous effort to avoid even the appearance of a conflict of interest had led to concern on the part of the vaccine establishment that committee members wouldn’t be able to understand the nuances of the issues under discussion. “Part of the criticism of the vaccine advisory panels such as the ACIP”—the CDC’s Advisory Committee on Immunization Practices—“was that manufacturers, developers, people with a vested interest in vaccines were on the panel,” McCormick says. As a result, the IOM decided that committee members were not allowed to have ever worked on the development or evaluation of any of the vaccines under review or to have testified on the issue of vaccine safety, nor could they have had ties to any vaccine manufacturers or any of their parent companies. The result was a panel that included pediatricians, neurologists, geneticists, public health experts, and ethicists—but no immunologists or vaccinologists. “The fear [in public health circles] was that we were setting a precedent for other vaccine advisory committees,” McCormick says. There is a good chance that she was precisely the type of person the vaccine establishment had been worried about.

  In the days after the IOM review was released, it became clear that, similarly to SafeMinds, many journalists had filed their stories before they’d taken the time to understand the report’s contents. As a result, the news coverage overwhelmingly ignored the lopsidedness of the evidence in favor of a “he said, she said” paradigm that treated all perspectives as equally valid. In support of its conclusions, the committee cited reports written by dozens of researchers from universities and governmental agencies around the world. In contrast, the evidence that had been submitted to the committee favoring a causal connection between autism and vaccines consisted primarily of one unpublished paper by Mark Blaxill and the collected works of two men who lived together in suburban Maryland: Mark Geier and his son, David.42 Unlike virtually everyone else who worked in the field, the Geiers’ studies were not done in collaboration with other scientists. Instead of using cutting-edge equipment, the Geiers worked in the basement of their two-story home in what Mark Ge
ier insisted was a “world-class lab—every bit as good as anything at the NIH.” (A New York Times reporter who visited the Geiers’ workspace in 2005 described it as “a room with cast-off, unplugged laboratory equipment, wall-to-wall carpeting and faux wood paneling.” Mark Geier’s training is as a geneticist; his son majored in biology in college but has not completed advanced degree work in any field.)

  By the time the IOM report was published, Mark Geier had been a mainstay of the anti-vaccine world for almost twenty years. In a 2002 paper titled “The True Story of Pertussis Vaccination: A Sordid Legacy?” he showed just how closely he adhered to its preferred narrative. Like many other people, the Geiers traced America’s current anti-vaccine movement to the early 1980s, when the focus of “Vaccine Roulette” on “children with severe injuries reported to be associated with whole cell pertussis vaccine” had spurred self-appointed citizen watchdogs to “band together” in order to “educate parents about the potentially harmful effects of childhood immunizations.” “Vaccine Roulette” also marked a turning point in Mark Geier’s life: It was then that he first learned of the team of attorneys who eventually “convinced him that he also needed to testify against the vaccine manufacturers.” (“Once the ice was broken,” the Geiers wrote, “other expert witnesses came forward to testify against the manufacturers.”)

  By the spring of 2004, Mark Geier had appeared in close to one hundred vaccine-related lawsuits. His willingness to appear on the witness stand might have bolstered his standing with activists, but it didn’t help his reputation in legal or scientific circles. For more than a decade, judges had been ruling that his testimony was so unreliable as to be worthless. In a 1987 case, Geier testified that his calculations showed that a version of the DPT vaccine made by Wyeth Pharmaceuticals was four times more toxic than the next most toxic pertussis vaccine on the market. As it turned out, he was wrong by 1,200 percent; as the result of his mistake, a $15 million jury award to the family of a severely retarded girl was summarily dismissed. (A year later, Geier blithely characterized his blunder as “what we call an order of magnitude error—that is, when I did the calculation, I must have missed a zero.”) In a 1991 case, a judge ruled that Geier “clearly lacks the expertise to evaluate” the conditions he was testifying about. The following year, one judge questioned Geier’s “qualifications, veracity, and bona fides,” while another wrote, “Were Dr. Geier an attorney, he would fall below the ethical standards for representation.” In the course of two separate trials in 1993, Geier was labeled “seriously intellectually dishonest” and his testimony was determined to be “not reliable, or grounded in scientific methodology and procedure.” And in 2003, after his testimony in a case before the Vaccine Court, a Special Master named Laura Millman wrote in her ruling that Geier likely did not meet the AMA’s “minimum statutory requirements for qualification” as an expert witness. Millman also noted that Geier’s role as “a professional witness in areas for which he has no training, expertise, and experience” appeared to contradict the AMA’s ethics guidelines, which state, “The medical witness must not become an advocate or a partisan in the legal proceeding.” She ended her appraisal of Geier’s work with advice for plaintiffs: “Petitioners must seriously consider whether they want to proceed with a witness whose opinion on neurological diagnosis is unacceptable.”

  The IOM review provided still more examples for why the Geiers’ papers were often regarded as substandard: Their procedures were “nontransparent,” their findings were “uninterpretable,” their experiments had “significant methodological limitations,” and their results were so extreme as to be patently “improbable.” (In one study, the Geiers claimed that 98 percent of the variation in autism rates between the mid-1980s and the mid-1990s was due to thimerosal. They didn’t explain how this jibed with their simultaneous support of the theory that the MMR vaccine, which did not contain thimerosal, also caused autism.)

  A close reading of the two men’s work made even those assessments seem restrained. In one of the papers they’d submitted to the IOM, the Geiers had relied on the U.S. Department of Education’s (DOE) tally of special education services to show that there’d been an increase in autism diagnoses in children born between 1990 and 1994 as compared to those born in 1984 and 1985. In doing their calculations, the Geiers failed to take into account that 1990 was the first year “autism” was given its own category by the DOE. They also didn’t acknowledge that in 1987, the DSM had scrapped the narrowly defined category of “infantile autism” in favor of a more expansive one for an “autistic disorder.” The description for the former took all of seventy-six words. It required that patients fulfill each one of the six listed characteristics to receive a diagnosis, one of which was an onset date before the age of three. The latter required 698 words to explain, and required patients to fulfill eight out of a total of sixteen total characteristics. Crucially, children no longer needed to be diagnosed before they were thirty months old; instead the age of onset could occur anytime in “infancy or early childhood.”

  In another paper, the Geiers referred to figures for the total number of vaccines distributed during a given time frame as if they were the total number of vaccines that had actually been administered. (To get a sense of why that’s a bad idea, think of the times when you’ve bought cold medicine or pain relievers in preparation for a worst-case scenario. Now take a look in your medicine cabinet and see how many of those bottles you’ve actually used up.) They also claimed to have determined that children who’d received a version of the DPT vaccine manufactured with thimerosal had higher rates of autism than those who’d received a version of the vaccine without the preservative—but they gave no indication as to how they’d concluded which children had received which version.

  In light of Laura Millman’s characterization of Mark Geier as a witness for hire, there was one last factor, unrelated to the actual substance of the Geiers’ work, that should have raised reporters’ antennae: All eight of their studies on thimerosal and MMR had been completed since the beginning of 2003, just as the single largest vaccine-related lawsuit in history was getting under way. It also went unobserved in news circles that the millions of dollars in damages at stake in the thousands of claims pending before the Vaccine Court may have had something to do with why the IOM report had generated so much outrage.

  42 The committee also considered Wakefield’s 1998 Lancet paper; however, the fact that it was “uninformative in assessing the hypothesized causal association between MMR vaccine and autism” precluded it from factoring into members’ deliberations. The report did note, “[S]ubsequent studies have not been able to replicate or prove this hypothesis.”

  CHAPTER 15

  THE CASE OF MICHELLE CEDILLO

  In 1986, when Congress established the framework for a compensation fund to simultaneously help people with vaccine-related injuries and shield manufacturers from potentially crippling lawsuits, it did so with the assumption that a streamlined, broad-reaching program would give some stability to the country’s immunization efforts. In order for the program to work according to plan, plaintiffs—or petitioners, as they were referred to—had to be able to navigate the system with relative ease. To that end, the law established a specially formulated table of injuries ranging from chronic arthritis and persistent bleeding to anaphylactic shock and paralytic polio. So long as a given injury occurred within a defined period of time following the administration of a vaccine, the court would assume causality without petitioners needing to provide further evidence. This “no fault” standard was the most significant way in which the Vaccine Court functioned differently from the traditional legal system.43 The trade-offs for this relaxed evidentiary standard and the court’s promise of relatively quick resolutions were the program’s predetermined limitations: Awards for death and “pain and suffering” were capped at $250,000; the maximum reimbursement for legal costs was $30,000; plaintiffs who accepted the court’s rulings lost the right to file suit in civil court; cl
aims had to be filed within three years of an injury’s onset; and plaintiffs who rejected the court’s decision faced severe restrictions on the conditions under which they could file future civil suits.44

  In its first decade of operation, the program was, by almost every benchmark, a success. By 1999, when the average number of cases filed had settled to around 130 a year, immunization rates had stabilized, wholesale vaccine prices had declined, and none of the remaining vaccine manufacturers had exited the market. During that entire time, the court had addressed claims for more than a dozen different purported injuries. Autism was not one of them. It wasn’t just that the disorder was not listed as a table injury and hadn’t shown up in the VAERS reports that served as an early warning system for vaccine reactions—it hadn’t even been the subject of any significant discussions. That began to change after the publication of Andrew Wakefield’s 1998 paper on the MMR vaccine: In 1999, a single autism case was filed. In 2000, there was one more. In 2001, the year “Autism: A Novel Form of Mercury Poisoning” came out, there were twenty-three.

  Then, after Lyn Redwood teamed up with a Dallas-based law firm that specialized in class action lawsuits, the floodgates opened.45 “Attorneys began to start focusing on it,” Redwood says. She had hoped the issue could be resolved without the interference of lawyers, but, she says, litigation became the only option after “federal agencies and professional associations” refused to be “up front with the American public.” With lawyers advertising in magazines and on infomercials, and autism Listservs and conferences abuzz over vaccine claims, thousands of parents began preparing their cases. The result was unlike anything the system had ever seen before: Suddenly, more autism claims were being filed each week than were usually filed for every other type of injury combined in a month or more. On July 3, 2002, Gary Golkiewicz, the Chief Special Master on the Vaccine Court, stated the obvious when he announced that the vaccine compensation program was facing an “unusual situation.” Already, three hundred claims had been filed that year alleging a connection between vaccines and autism—and, Golkiewicz wrote, lawyers had informed him that they were planning on filing anywhere from three thousand to more than five thousand additional claims “during the next several months” alone. (By the end of 2004, a total of 4,321 autism cases had been filed.) In order to ensure that families could resolve their cases in a timely manner, Golkiewicz wrote, the plaintiffs and the government had agreed to address the “general causation” theories—“i.e., whether the vaccinations in question can cause autism and/or similar disorders, and if so in what circumstances”—as part of a single “Omnibus Autism Proceeding.” After the causation phase of the trial had been decided, the results would be “applied to the individual cases.” Because of the Vaccine Court’s purposely diminished burden of proof, a positive ruling in the causation phase would in all likelihood have resulted in compensation for thousands of other families with autism-related claims.