by Sonia Shah
9
The Emperor Has No Clothes:
The Vagaries of Informed
Consent
Neisseria meningitidis is generally a placid sort of bacterium. Sluggish and round, it tends to quietly reside in the back of peoples’ throats, minding its own business.
But every ten years or so, across a stretch of Africa from Ethiopia to Senegal, something different happens. Perhaps people have lost their immunity. Perhaps the bug has evolved into a more sinister form. Whatever the cause, during these episodes, when the dry air comes and Neisseria escapes in tiny, coughed-out droplets, the guest turns wild. Neisseria speed through the bloodstream of its new hosts—mostly children—finally reaching the meninges, those crucial membranes covering the brain and spinal cord, where it triggers a deadly inflammation.
Vomiting, high fever, and mental confusion follow. Around 80 percent of those sickened with Neisseria’s meningitis die, some within hours. During epidemics, tens of thousands might perish in a matter of months.1
One of the largest recorded outbreaks of epidemic meningitis occurred in 1996. The first reports trickled out of Nigeria in January. The next month, the numbers shot up. By March, the eruption was in full force, and emergency response teams from the Nigerian ministry of health along with the World Health Organization, UNICEF, and Médecins Sans Frontières rushed to Nigeria to dispense eight million doses of vaccine for those still well and powerful antibiotics to the ill. Another ten million vaccines were on the way.2 Among those Nigerian children lucky enough to receive a rapid IV infusion of antibiotics, the chance of death would plummet to around one in ten.3
It didn’t take long for news of the outbreak to reach Pfizer headquarters in chilly southeastern Connecticut, where it fell upon pricked ears. The company was racking up proof that its experimental broad-spectrum antibiotic Trovan worked on everything from gonorrhea to bronchitis and pneumonia, a potential $1 billion blockbuster in the making.4 It seemed a golden opportunity to prove Trovan’s mettle had fallen into the company’s lap. In the United States, only about three thousand people contracted meningococcal meningitis every year.5 In Nigeria, Pfizer could test Trovan on hundreds of untreated patients in a few weeks. If the FDA could be convinced that the drug worked for meningitis in these Nigerian kids, they could effectively break open the entire pediatric market. Pfizer’s Scott Hopkins, MD, quickly drafted a protocol for a quickie experiment.
The news of the brewing plan alarmed Pfizer’s infectious disease specialist, Juan Walterspiel. Walterspiel knew of his employer’s eagerness to deliver the new bug fighter in convenient oral form rather than the usual injections proffered by competitors; to see how an oral formulation worked, the trial would administer Trovan in a tablet or a spoonful of syrup to some of the Nigerian children. But sending a drug into the body via the mouth is a circuitous route compared to shooting it straight into a vein, and Walterspiel knew that a rampaging Neisseria meningitidis can kill people fast. That’s partly why the WHO recommends the use of injected antibiotics to counter meningitis outbreaks.6 Plus, these kids in Nigeria weren’t just facing one pathogenic microbe: they were malnourished and reeling from concurrent epidemics of measles and cholera. What if the bug overwhelmed them before the pills and syrup could be absorbed?7 Their blood would be on the company’s hands.8
But, according to allegations contained in a later lawsuit, neither Hopkins nor the rest of his six-person team took Walterspiel’s concerns seriously.9 In April 1996, filled with faith in Trovan’s prowess, Hopkins and his team boarded a chartered DC-9 headed for the outbreak’s epicenter: the dusty northern Nigerian city of Kano. “We had to move quickly,” Betsy Raymond, a Pfizer spokesperson, explained to the Washington Post reporter who reported on the trial in 2000.10
The facilities in Kano did not impress Hopkins. At the Kano Infectious Disease Hospital, aid doctors conducted mass vaccinations and treated the sick with shots of chloramphenicol, the cheap, off-patent Neisseria-killer recommended by the WHO.11 Chloramphenicol had been developed in the 1940s and had fallen out of favor in the United States in the early 1980s, when it was found to cause fatal aplastic anemia in one out of thirty thousand patients.12 “I wouldn’t give my dog [chloramphenicol],” Hopkins said to the Post in 2000.
The Pfizer scientist set to work taking over a ward of the crumbling cinder-block facility for his Trovan test. A nurse at a triage desk would direct the stream of ailing patients stumbling into the hospital either to the aid workers or to the Pfizer test clinic, where Hopkins would offer what he considered to be superior treatment options. Around one hundred lucky children would get Pfizer’s Trovan and another one hundred Roche’s Rocephin, a brand-name meningitis drug approved in 1993.13
Like many drug companies taking their trials overseas, Pfizer hadn’t fussed with filing an application with the FDA before jetting over to Kano. Pfizer’s researchers knew that in terms of patient protections, all the company needed to do was follow the mostly philosophical edicts of the Declaration of Helsinki and the FDA would likely happily accept the forthcoming data. This wasn’t a high-profile experiment funded with public money, but a proprietary experiment that few outside the FDA and Pfizer would probably ever review in detail. The only concrete requirements entailed acquiring approval from a local ethics committee and, of course, written evidence of the subjects’ informed consent.
* * *
Underneath the layers of oversight fashioned to protect human subjects from harm in medical experiments lies a hard inner core, forged when the horrors of coerced human experimentation were finally laid bare after World War II. It is the subject’s own informed and voluntary consent—the oldest and most universally accepted ethical standard in research. “Ultimately,” wrote NBAC researchers in a typical formulation of the standard’s centrality, “research can go forward only if participants understand what the research entails.”14
Few researchers, however, ever bother to verify whether their subjects do, in fact, understand. It wouldn’t be difficult. Medical researchers routinely double-check, duplicate, and re-analyze nearly every other aspect of clinical trials by means of a profusion of journal articles, conferences, workshops, and lectures. “Relentless scrutiny of details” could be said to be the research industry’s motto. But in the area of informed consent, an atypical atmosphere of “don’t ask, don’t tell” prevails.
One survey found that while over half of researchers admitted that it would be a good idea to verify their subjects’ understanding of experiments, a scant 16 percent had ever done so in their own trials.15 Partly that’s because when researchers have attempted to confirm the integrity of informed consent, the bedrock has revealed itself to be made of very soft shale indeed.
In standard practice, to “consent” a subject, the investigator holds a single meeting with a prospective participant, explains the trial and the consent form, answers a few questions, and passes over the form for a signature. It is a brief, legalistic exchange that satisfies sponsors, oversight committees, and regulatory authorities, protecting all concerned from liability. But the evidence suggests it does little to enlighten test subjects about the risks of experimentation. This comprehension gap is even wider in the case of Western-sponsored trials conducted in developing countries.
Indrek Kelder, an accountant from Estonia, wasn’t even sure what drug he’d been given in the 1998 industry trial he participated in. “Maybe it was written between all this mumbo jumbo,” he offered to the Washington Post. His fellow trial subject, Estonia’s Irme Petrimae, was similarly confused. “They told me something about skin disease,” he allowed. “All these forms, and we didn’t get any copies. I didn’t like it . . . but it was too late to change our minds.”16 Despite having signed informed consent forms, no fewer than thirty of thirty-three Thai volunteers enrolled in one trial of an experimental HIV vaccine later said they shared the mistaken belief that “the vaccine they received was effective,” according to a 1997 study published in the Journal of the Medical Associ
ation of Thailand.17 In Haiti in 2002, when researchers turned around and quizzed subjects on the basic outlines of the HIV transmission experiment they had just agreed to, 80 percent flunked.18 In a 1998 study in Brazil all subjects in a trial of an experimental contraceptive enrolled under the erroneous impression that the experimental intervention “would be good for them.”19
Disturbing signs that test subjects’ participation in trials is not particularly voluntary have also emerged. In a 1998 analysis by South African epidemiologists, more than eight out of ten South African women inducted into an HIV prevention trial, all of whom had already been through an informed consent process, professed to feeling coerced into remaining in the trial.20 “Two little bottles of blood was taken from us without asking our permission,” complained an HIV-positive South African woman who enrolled in a trial after being approached by a recruiter on her street.21 A similar proportion of Bangladeshi women in a 1998 trial on iron supplements revealed that they had no idea they were free to abandon the experiment.22
The best evidence of voluntary, informed consent is when some subjects drop out or refuse to participate in a trial, researchers say. “That’s a good sign,” one noted to NBAC researchers. “They are able to refuse, they are free to refuse [to enroll].”23 And yet, in NBAC’s 2001 survey of researchers active in developing countries, 45 percent reported that their low-literacy subjects never refused to participate.24
“I want the money,” explained Thai taxi driver Wivat Chotchatmala, a drug user who enrolled in one of VaxGen’s HIV vaccine trials.25 “We continued to visit the distribution center” to participate in a U.S.-government sponsored trial of a new antimalarial drug, said Kenyan villager Lucas Oyombe Otieno, “so as to enjoy the free meals.” Otieno and his fellow subjects had already stopped taking the study drug because of its unpleasant side effects. “We would collect our drugs, pretend to swallow them, but we would throw them away later,” he told the East African Standard in 2004. “When they offered to provide us food, many of us accepted to be used in the study,” added Andrew Okal Seda, who enrolled in a similar antimalarial drug study in Kenya. “Around that time, there was widespread famine in the area.”26
And the truth is that, contrary to the blithe pronouncements of the NBAC and others that research can only proceed when subjects are informed and participate consensually, for some CROs, uninformed, coerced subjects don’t impede medical research at all. On the contrary, unknowing, docile subjects make trials easier and faster.
“Russian subjects don’t miss appointments, they take all the required pills, they fill in the questionnaires and the diaries, and only very rarely do they withdraw their consent. . . . Russian subjects do what their doctors tell them to do. What a phenomenon!” a promotional article in the trade journal Applied Clinical Trials gushed.27 CROs routinely tout the “compliance” of their experimental subjects on offer, especially those in poor countries where many patients are lucky if they can get a doctor to pay attention to them at all. James Loudon, an executive from a CRO active in China, encouraged drug companies to experiment on patients there, not because the patients are empowered and informed, but because “the Chinese are not that fully emancipated as in the U.S. They are more willing to be guinea pigs,” he said to CenterWatch in 2002. “It’s not like Japan, where it’s difficult to get patients to do trials, or they refuse at the last minute.”28
It isn’t technically difficult to ensure that even the most unknowing, premodern, and illiterate subjects are informed and consensual partners in high-tech medical research. Scientific notions ingrained in Western society may be foreign to others, but they are not beyond comprehension. “You can do it, if you put in enough resources,” says Anne-Valerie Kaninda, an aid doctor who treated meningitis-stricken children in the same Kano hospital where Pfizer conducted its Trovan test. “People are not as stupid as they are portrayed. I understand that it might be difficult to talk about informed consent when you are coming in with Western concepts of medicine, and there you have a mother with a sick child who doesn’t know what a microbe is. But still, if you put in enough resources, with a translator you can still explain, ‘You have a choice, either you go the regular way, or you come with us; we are testing.’ . . . It means that whoever is conducting the trial has to put resources into that, with a lot of well trained people.”29
Several simple methods have been shown to bridge the gulf between Western investigators and patients from developing countries. Researchers can take advantage of the fact that some patients in developing countries are more capable and inquisitive than their Western counterparts by virtue of having survived in more difficult, resource-constrained environments.30 After subjects sign the forms, researchers can quiz them. Those with low scores can be counseled further, or dropped from the trial. Study staff can hold periodic meetings with subjects to answer questions as they come up and ensure their participation remains voluntary.31 Researchers can hire local people to explain the study, by using pictures and diagrams, and references to local practices in agriculture or community life. They can launch educational programs about the experiments in the larger community. They can approach community leaders to help establish legitimacy first.32
If after all this, informed consent is still not achieved, then the prudent response might be similar to that taken when a piece of crucial equipment fails. The study, rendered untenable, is abandoned. Such practices might stymie some trials, but “not all knowledge is accessible,” says bioethicist Jonathan Moreno. “It is one of the trade-offs between realizing there is a moral difference between people and lab rats.”33
But in many cases, that is not what happens.
Instead, researchers and regulators profess defeat—and press on anyway. They try and they try, but getting impoverished, illiterate villagers to understand and consent to their experiments is simply very difficult, if not impossible, some insist. M. Upvall and S. Hashwani of Aga Khan University in Pakistan compared informed consent procedures in Pakistan and Swaziland for a 2001 International Nursing Review paper. According to their study, researchers were often forced to penetrate layer after layer of tribal hierarchy and corrupted bureaucracy in order to obtain informed consent. Sometimes they had to ask the village elders or the husbands of women participants first, or employ police escorts, or have tea and snacks, “regardless of the time it took.” Plus, they had to struggle with the fact that some subjects don’t have telephones, or permanent addresses, and may even be afraid to sign their names. Such findings on the barriers to informed consent “demonstrate the inadequacy and complexity of applying western-based concepts of informed consent to developing countries,” Upvall and Hashwani wrote.34
It is a pointless exercise, some researchers insisted to NBAC analysts in an anonymous 2001 survey. In some African languages “there is no word for ‘research’ or ‘science,’” one says. “There is no concept of an experiment or placebos.”35 “Informed consent is a joke,” said another. “It is not possible to claim a person who has never heard of a bacteria or virus is informed about what a vaccine or drug is doing or how their participation fits into any such study.”36 Perhaps informed consent is just too Western a notion, they say. After all, “this concept of the consent of the individual, rights, and individual decision-making doesn’t exist” in many authoritarian developing countries, one said. “People do what they are told, and people are told what to do.”37 In India, according to a 1996 Journal of the Royal Society of Medicine paper, physicians consider informed consent “an unnecessary ritual.”38 In Latin America, one doctor told the Washington Post in 2000, physicians were “against” informed consent. “Patients want doctors who say, ‘Do this because I say so.’”39
Some industry researchers are even more forthright. They must mislead subjects because “if all potential adverse effects are listed in the informed consent, the patient will be scared away,” as John C.M. Lee, a Bristol-Myers Squibb researcher working in China wrote. “One needs some flexibility in implementing” requ
irements for informed consent. Due to educational and cultural background, a complete written informed consent is very difficult to achieve. It is a very long-term goal that China follow the international standard in this area.40
Even regulators profess impotence. “We have requirements for informed consent,” says the FDA’s Robert Temple, but “I’m not going to tell you it is always well done. . . . Should there be more real-time interviews? Some ethicists talk about this. It is an area that needs attention. But we don’t do anything about it immediately. We rarely go to a study that is ongoing.”41
Not a single subject signed a form indicating his or her consensual participation in Pfizer’s Trovan experiment in Nigeria. The children were too young to sign, the company said in its defense, and the parents didn’t speak English, the language the form was in. But in addition, no witnesses attested to any verbal consent, if it was given at all.42 According to Elaine Kusel, a lawyer who would later represent the Nigerian families involved, it wasn’t. The subjects “had no idea they were part of any clinical trial.”43
Despite Pfizer’s apparent inability to acquire written evidence of consent, the researchers didn’t throw their hands in the air and jet back home. They moved forward with their experiment anyway.
Researchers are often convinced they are helping their subjects, whether the patients know it or not. In the end, Pfizer lost the same proportion of children in their Trovan group as their Rocephin group, around 6 percent.44 So, even with subjects later saying they were tricked into participating, the attitude of other Western researchers is cavalier. “I don’t see the harm in it,” said UCLA meningitis expert Larry Baraff of the study. “It sounds like they all did well.”45
