Make thick my blood,
Stop up th’access and passage to remorse,
That no compunctious visitings of nature
Shake my fell purpose
—MACBETH, v.v.43–46
With near-parodic gothicism, Friday the thirteenth was plagued by gray clouds, frigid rain, sleet and snow in succession—a foul day, even by the low standards of January in Detroit. As night fell, Martha Milete finished dinner and padded into her bedroom.
After a moment, she remembers, “I heard a thump.”
It was probably nothing, Milete thought, maybe a dog outside, but she sighed softly, rose to open the bedroom door, and found herself staring at two men brandishing large handguns. Wordlessly, one masked intruder held a gun against her temple as the duo shoved her into her dining room and thrust her onto her knees, snarling at her to keep her hands behind their head. As one thug batted drawers open and pawed through sideboards and chests, the other barked, “Give me all the money, all you’ve got.”
“But I had no money, and I told them so. ‘I have no money. There’s nothing in the house.’ ”
“Yes, you do,” the first invader said slowly and distinctly, his gun steadily trained on her.
“I looked in his eyes: they were dead, and I knew. I knew. I begged him, ‘Don’t! Please don’t! Don’t shoot me.’ I was still begging him when I was blown backward by the force of the gun’s blast. I couldn’t breathe and felt a searing pain in my eye. They ran though the front door, and I was alone. I couldn’t see much, but I was able to call my son and say, ‘I’ve been shot’ before everything went black.”
Milete had just become a victim of urban violence, one of 100,000 Americans shot in 2006: a statistic.
“My son called 911: he saved my life,” she recalls. The ambulance screamed to her home and picked her up, and while she was en route to the hospital in the hurtling vehicle, she woke up. “I was conscious while an EMT gave me blood,1 but then I passed out, and I don’t remember anything after that until I was in the hospital.” Minutes after her son’s frantic 911 calls, Milete lay in Detroit’s Sinai-Grace Hospital, where she underwent six hours of surgery for the gunshot wound to her chest, the force of which had also injured her retina, requiring specialized ocular surgery. For days, she hovered between a vestigial consciousness and oblivion. “I would come to in the hospital and hear a little conversation, and then I’d black out again. For days, two people—a guy and a girl—kept drawing blood from me: at first I thought it was part of the hospital routine. When I became a little more alert I realized that the people who were drawing blood from me several times a day were not part of the team caring for me. They wore insignia that read ‘Wayne State University,’ and all they wanted was my blood, which they took twice a day. I remember getting real upset one day. I shouted, ‘You’re not taking one more ounce of blood from me. Stop and leave me alone.’
“Then my daughter Cathy told me why, that it was part of the medical experiment that I was in. She explained that when I was bleeding heavily, lying between life and death, they had given me an experimental liquid instead of blood, the artificial blood. It was called PolyHeme.”
Did no one from the hospital or research team tell her that she had been enrolled in the PolyHeme study? “No, my daughter told me,” insists Milete. “The medical staff never told me. They never asked me anything. They were still studying me as part of an experiment. There was a coordinator at the hospital, she didn’t say much. She kept telling nurses to ‘take more blood, take more blood’: she was in cahoots with the research people.
“I could not believe it! I told them that I wanted no part of it. It was wrong. Although I survived, it was wrong. How can they make you a guinea pig without asking your permission?”
Between 2003 and 2006, 720 trauma victims at thirty-two U.S. medical centers were “enrolled” in a research study to determine the efficacy of PolyHeme, a patented blood substitute manufactured by Northfield Laboratories. Its maker claimed that PolyHeme could safely and efficiently replace the standard saline solution and blood that keeps heart-attack, car-crash, gunshot, and other accident victims alive. None of these victims was asked for his or her consent, yet this PolyHeme study broke no law. This is because since 1996, federal regulation 21 CFR 50.24 has allowed research with uninformed and unwitting trauma victims to be conducted in emergency settings such as aboard ambulances and medical-evacuation helicopters.
I wished to discuss the case of Milete, Northfield’s use of the 50.24 exception, and other events concerning the testing of PolyHeme with the company. Spokesperson Sophie Twaddell promised to call me back, then failed to do so over eight weeks of unreturned phone calls. She finally responded to an email with a refusal to allow me to speak with any Northfield representative or even to respond to emailed questions.
This media silence in response to probing questions was not unusual for Northfield, and its omertà even extended to medical communication around clinical questions. Instead of releasing data in peer-reviewed medical journals, researchers communicated study data through press releases and apparently issued gag orders to researchers. Of the ten Poly-Heme researchers I asked for interviews, only one, Dr. Andrew Bernard of the University of Kentucky at Lexington, agreed to speak with me, which he did openly and frankly. The other medical investigators cited the Health Insurance Portability and Accountability Act of 1996, or HIPAA, regulations that govern disclosures that might violate patient privacy. Or they claimed that releasing requested information would threaten Northfield’s proprietary information. Or they simply refused to speak to the inquisitive.
How did our nation come to determine that informed consent, and even consent of any kind, is disposable? Unsatisfied with the occasional waivers of informed consent that were issued on a case-by-case basis, in May 1994 a congressional subcommittee charged the FDA and the Department of Health and Human Services with developing consistent guidelines that would allow institutional review boards, or IRBs, to green-light such research. Two years later, FDA Regulation 21 CFR 50.24, which detailed the procedure for exempting research from informed consent, became law.
Today many researchers defend this ethical sea change by invoking the need to provide quicker, better, evidence-based responses to traumas.
But laws on the books already enabled health-care providers to treat trauma victims who were unable to give consent, and some states’ Good Samaritan laws go so far as to protect bystanders who treat seriously ill people, even should the injured worsen or die afterward. So let’s be clear: The exception is not necessary to treat people who are too sick to give consent. It is necessary only to study them. The trauma victim requires intervention, but the “50.24” exception law allows experimental treatments that, unlike the federally approved treatments that constitute the standard of care, are not known to be safe.
“Research is not treatment,” the late Jay Katz, MD, reminded us in 1997, in the wake of the 50.24 approval. Katz, who was Elizabeth K. Dollard professor emeritus of law, medicine, and psychiatry at Yale University, added: “And whenever clear distinctions are not made between the two, the waiver of informed consent becomes problematic because some human subjects are being recruited to serve the ends of others.”
Daniel Nelson, PhD, director of the Office of Human Research Ethics at the University of North Carolina at Chapel Hill, offers reassurance on the basis of the presumed rarity of such exceptions. “You know in 1996, when the rules changed, we discussed the exception proposal at length, we agonized over it, over the language and protections. When we finalized the legislation, we braced ourselves for an onslaught of proposals to use the exception, but they never came. It’s very rarely used.”
However, from the beginning other experts reacted with deep misgivings regarding research with the 50.24 statute. “It’s a fateful step,” proclaimed Katz, who, just a few months after the statute was adopted, wrote in the Hastings Center Report of his “many problems with the way the legislation was drafted.”2 Chief among these were “the vast and vaguely defined discretion granted to IRBs in administering these fateful regulations” and his fear that the affected communities would be duped regarding the true nature of the research into which they were conscripted. “The informed-consent dialogue in research with competent patient-subjects,” he wrote, “must be stripped of the ‘therapeutic illusion,’ which misleads patient-subjects into believing that they are receiving the most advanced and beneficial treatments available, when instead they are being asked to serve the interests of science.”
That same year Adil Shamoo, PhD, a University of Maryland professor and the editor of Accountability in Research, wrote that there may be a limited role for nonconsensual research but warned, “In ten years we’ll have some abuses and people will start rethinking the rules.”3
In this chapter I focus on the PolyHeme study as an example of the pitfalls and ethical failings of 50.24 research, but it is far from the only recent episode of investigative servitude for patients unwittingly conscripted into the studies that are often conducted in order to make or to save money for corporations.
Wartime military expediency has often escalated the erosion of human rights in research, and recent events have proved no exception. The military fired the first modern legal salvo against informed consent in the shadow of the impending Gulf War and subsequent Middle Eastern hostilities—and besides the military, at least one private drug maker, then called BioPort, stood to profit.
The Department of Defense (DOD) sought and secured the FDA’s permission to dispense with informed consent as it forced 2.4 million soldiers to accept injection with an experimental anthrax vaccine, a patented product of BioPort Laboratories, via its 2000–2005 Anthrax Vaccination Immunization Program (AVIP). So, just four decades after the army had overseen the Nuremberg trials of twenty Nazi physicians on charges of conducting experiments upon the powerless without their consent, the DOD opted to experiment on its own soldiers without their consent.4
This odyssey into research without consent proved a medical and legal disaster that eroded many soldiers’ trust in medical research. Soldiers suffered miscarriages and were maimed, blinded, and killed, all of which they blame on experimental anthrax vaccines. The Washington Post raised questions about the safety and quality of the vaccine and alerted the public that the factory in which it was manufactured had been the subject of repeated FDA evaluations, which found substandard hygienic and production conditions.5
By a conservative estimate, 2,500 soldiers refused the experimental vaccines, and as a result many were court-martialed, jailed, or forced to leave the service with less-than-honorable discharges. They literally had no recourse, because the Feres Doctrine stipulates that soldiers on active duty cannot sue the U.S. government for personal injuries experienced in the performance of their duties, and their families cannot sue for wrongful death.6
One of these soldiers was Jamekia Barber, a private first class in the Seventh Infantry Division stationed at Fort Carson, Colorado. She and her husband were alarmed by reports of devastating side effects, including miscarriages in soldiers who had taken “the shot.”
“We did research and found other people who were disabled because of the shot. Another man, an African American, had 20/20 vision before he took the shots: he lost 80 percent of the vision in one eye and 40 percent in the other eye. After the third shot he feared he would die, but the army says it was a reaction to Tylenol.” A lawsuit was not an option for this soldier, because of the Feres Doctrine.7
“We were taken to a hospital on post and given printouts saying that there were no adverse reactions and that the various drugs had been approved by the FDA in 1999 or January 2000. But a pregnant girl in my command had taken two anthrax shots, and when she went to the doctor one day he said, ‘The baby is gone.’ I didn’t want to take the shot. I could not agree to that, because having children was important to us.” Barber’s concerns about the vaccine and pregnancy are supported by the admonitions in the vaccine’s product insert, including, “Studies have not been performed to ascertain whether Anthrax Vaccine absorbed has carcinogenic action, or any effect on fertility.”
But the army did not need Barber’s permission. Barber tried to resolve the issue through legal means and requested a transfer to a unit where she would not need to submit to the injections, but she says her commanding officer blocked it, intending to “make an example” of her, and that he encouraged others to harangue her into compliance.
“I was bothered by the blatant disrespect of the men around me, who were pushing me as they shouted at me to take the shot.” She relates being physically assaulted, followed by confinement to a barracks until the day she jumped out of the second-story window because “I learned that I was being detained in a building where a gang rape had taken place on the same floor just two weeks earlier.”
In the end, demoralized and suffering from PTSD, Jamekia accepted the proffered “Chapter 10” resignation from the army in lieu of a court-martial because, she says, she was assured that she would not receive anything less than an honorable discharge. But on May 11, 2000, a less-than-honorable “administrative discharge” was exactly what she received. She unsuccessfully appealed the decision in 2003, by which time her husband had also been released with an administrative discharge.
Later that year, Judge Emmet G. Sullivan of the United States District Court in Washington, D.C., ruled to end the forced experimentation. The FDA responded by rapidly elevating the anthrax vaccine from a questionable investigational drug to an approved therapeutic, which allowed the DOD to sidestep the intent of the law and force the medications on soldiers as part of fitness-for-battle measures.8
This move returned U.S. soldiers to a state of investigative servitude—“investigative” because the data collection and evaluation of the anthrax vaccine risks, including death, continued. In rapidly approving the vaccine, the FDA had violated not only the intent of Sullivan’s ruling but also its own regulations by failing to hold the required public hearings.
In 2004, half a dozen unnamed soldiers filed a class-action suit protesting the vaccinations.9 Judge Sullivan, again presiding, finally drove a stake through the heart of this pro bellum experimentation with a decision that read in part, “The women and men of our armed forces put their lives on the line every day to preserve and safeguard the freedoms that all Americans cherish and enjoy. Absent an informed consent or presidential waiver, the United States cannot demand that members of the armed forces also serve as guinea pigs for experimental drugs.”10
Between 2004 and 2011, more legal cases were brought by soldiers who had been forced into the vaccination program while the DOD attempted, on several occasions and with limited success, to restore mandatory vaccinations. The FDA issued a string of actions against the vaccine manufacturer, which were triggered by quality issues such as failed potency tests and unapproved changes in manufacturing as well as the soldiers’ injury lawsuits.11 The vaccinations are currently voluntary, but will they remain so? Currently, Emergent BioSolutions, now the parent company of Bioport, has committed to preparing 1.45 million doses of anthrax vaccine by 2011.12
After Sullivan’s decisions ended forced research on soldiers, Barber renewed her appeal and won an honorable discharge. But the price of her vindication was high: she and her husband divorced. Although she has remarried and she and her ex-husband remain friends, Barber attributes the breakdown of their marriage to the strain of fighting the DOD. “Sometimes I wonder what our lives would have been like without ‘the shot.’ But I don’t allow myself to dwell on it.”
There have been many more recent instances of U.S. citizens conscripted into medical research without their knowledge or consent. In a 2001 Maryland appeals court decision, Judge Cathell condemned a Kennedy Krieger Institute study of lead levels in children whose parents were urged by its staff to rent homes that the institute, which serves children with pediatric developmental disabilities, knew to be tainted with lead. The study had been approved by the IRB of Johns Hopkins University, with which the KKI is associated, but its design was unethical because the KKI researchers told families with young children that they would help them find lead-free housing, then referred these families to housing that they knew to be imbued with lead. The motivation was financial: the KKI was trying to determine the cheapest way to reduce the homes’ lead, and the bodies of the children were used to titrate the resultant lead exposure. Predictably, some children’s lead levels rose, and they were visited by a Pandora’s box of ills, including mental retardation. The court found the KKI guilty of using black Baltimore children like “canaries in a coal mine” in a veiled experiment to test lead levels.
In 1995, black and Hispanic children in Los Angeles were given experimental measles vaccines without their parents’ knowledge, and that year the Medical University of South Carolina was accused of illegal human experimentation when it enrolled pregnant women in North Carolina, most of them black, in a drug-treatment research study without their knowledge, and then reported them to police as drug abusers. A year later, three New York City research institutions gave six- to eleven-year-old black boys the cardiotoxic drug fenfluramine as part of research into genetically mediated violence.
As early as 1996, a report by the Cleveland Plain-Dealer determined that 4,154 FDA inspections of new drugs that have been conducted since 1977 uncovered a dramatically broad abandonment of informed consent. The article noted, “More than half the researchers were cited by FDA inspectors for failing to clearly disclose the experimental nature of their work.”13
Many of these nonconsensual studies, whether permitted by an FDA waiver as the anthrax vaccine was, licensed by the 50.24 statute as with the PolyHeme trauma trial, or pursued in violation of the law, like the South Carolina study, were conducted for financial gain. Bio-Port, which manufactured the patented anthrax vaccine that the DOD was testing on soldiers against their will, hoped to sell it widely—but could not do so without FDA approval via the heavily populated clinical trials that the waiver of informed consent guaranteed. As we’ve seen, the KKI’s agenda was also financial, as it sought to determine the cheapest effective route to lead abatement. Money changed hands in the form of generous funding for the ethically indefensible South Carolina drug-treatment and New York City fenfluramine studies. And the very first trial approved under the 50.24 exception was that of HemAssist, a blood substitute that was a patented product of Baxter International of Deerfield, Illinois.14 One hundred uninformed subjects were enrolled in the study, but it was halted when reviewers realized that the mortality rates were unacceptably high.
Given this blighted history, why was PolyHeme research thought urgent enough to merit this draconian testing scheme? One reason is that Northfield laboratories had only PolyHeme, its sole patented product, to sell. Time was running out on its window to complete clinical trials, and dispensing with informed consent was the fastest way to recruit the hundreds of subjects needed to complete them.
Such scenarios raise the question of how extensive a role such corporate financial goals play in promulgating research without consent.
True medical urgency exists with trauma victims because their lives depend upon hemoglobin, the scarlet protein that enables red blood cells to ferry oxygen throughout the body. Whether rent by steering wheels, blown aneurysms, or gunshots, the torn bodies of trauma victims crave blood. When an ambulance arrives at a trauma scene, whether it be a car crash or Martha Milete’s gunshot assault, replacing the blood ebbing from the victim’s body becomes a priority. Too much lost blood, unreplaced for too long, will starve tissues and organs, including the brain, of oxygen, and culminate in unconsciousness, organ damage, and finally death.
A blood transfusion is impractical within an ambulance, however; because blood can take up to an hour to type and match, it must be refrigerated, and storing the necessary types requires too much space. Fortunately, a cheap, plentiful, and portable blood substitute has been proven effective and safe: salt water. Salt water in a 0.9 percentage solution (called “normal” because it mimics the saline concentration in our bodies) can be quickly and easily infused into trauma victims. Saline expands the volume of blood, staving off shock in plenty of time to get urban victims to the hospital, where replacement blood is available. On city streets, saline is the safe, proven, pre-hospital standard of care.
But not outside cities. Saline has a critical limitation for rural residents and for battlefield casualties who may be hours from blood and hospital care: saline has no hemoglobin, leaving the tissues and brains of battlefield victims oxygen-starved and at risk for a cascade of vascular disasters leading to stroke, heart attack, and death.
Acutely aware of the danger to injured soldiers, the Department of Defense has been experimenting for decades with myriad blood substitutes, auditioning everything from the oxygen-saturated perfluorocarbons that inspired the divers’ breathable liquid in the film The Abyss to freeze-dried “blood” powder: just add water. This obsession of military physicians continues a long medical crusade: In the seventeenth century, Christopher Wren infused wine into a dog’s bloodstream, and a century later, American gynecologist Gaillard Thomas flushed milk into the veins of pallid, blood-starved postsurgical patients. Substitutes derived from human blood were first tested in 1933, when William Ruthrauff Amberson of the University of Tennessee boldly infused hemolyzed red blood, only to be rewarded with a stemmed flow of urine, bradycardia (a dangerously slowed heartbeat), and skyrocketing blood pressure, followed by ugly deaths from kidney failure. A host of other contemporary blood substitutes have proved equally disappointing.
By the 1990s, as wars and rumors of wars fed fears that U.S. battlefield traumas would escalate, the DOD and trauma surgeons elsewhere turned their attention to a species of blood substitutes called hemoglobin-based oxygen carriers (HBOCs). PolyHeme was one. Milk and wine may seem arcane blood-substitute candidates, but PolyHeme is made from no less gothic a substance—expired human blood, donated for transfusion but long past its shelf life.
PolyHeme contains hemoglobin. But it shares a devastating medical limitation with other HBOCs: Once sprung from the prison of its blood-cell membrane, an unrestrained hemoglobin molecule becomes a rogue agent tiny enough to indiscriminately penetrate the walls of veins and arteries. There, free hemoglobin molecules trigger inflammation, causing the muscular cell membranes to seize and contract, which can block a blood vessel to the heart, triggering a heart attack. Studies of the HBOC HemAssist, for example, had to be shut down in 1998 when nearly half of the fifty-two trauma patients infused with it died,15 compared to only 17 percent who received standard therapy.
Each HBOC manufacturer in succession has claimed to have generated its own unique, patented process to transform free hemoglobin into a safe bearer of oxygen. So far, none has been shown to work to the FDA’s satisfaction, and the biotech boneyard is littered with “definitively” modified HBOCs.
In PolyHeme, Northfield Laboratories claims to have found the holy grail of hemoglobin safe conduct: polymerization. Northfield’s answer is to harness quadrupled arrays of hemoglobin, called tetramers, to one another in chains called polymers. This rearrangement, Northfield claims, will neutralize free hemoglobin’s errant ways—and eliminate the risk of heart attacks and deaths.
If this worked, PolyHeme would allow medical institutions to do without much of the real blood they must buy, test, type, refrigerate, and store for the 3.5 million Americans who undergo blood transfusions every year. FDA approval would also allow Northfield to capitalize on PolyHeme’s patent to corner the roughly $3 billion blood-substitute market.16
To this end, Northfield’s CEO, Stephen Gould, MD, armed with an initial $1 million from investors, ultimately raised $194 million after taking Northfield public in 1994. He then turned his attention to winning FDA approval.
It is easy to understand that for-profit companies such as Northfield could be willing, even eager, to give personal autonomy and informed consent short shrift if they stand between them and the FDA approval they need. However, it is harder for the public to accept that money can compromise the independence of medical researchers and academics as well.
Northfield has thrown a lot of money around, beginning with the approximately $10,000 per patient it paid hospital research programs. The University of California at San Diego was paid $10,840 for each patient who completed the study, discounted steeply for an early withdrawal: the university was guaranteed approximately $73,000. Some researchers gleaned additional bonuses, such as the $556 per capita handling fee received by Scripps Mercy Hospitals. According to a New York Times account, Northfield paid the University of Texas Health Science Center at Houston $336,000 to test PolyHeme, and the University of Kentucky Medical Center garnered $132,468, funds the hospitals say merely covered operating costs. When pressed about this, University of Kentucky surgeon Andrew C. Bernard responded, “This is not a profit-making endeavor—it is a scientific one.”
Actually it is both. This model has become the norm within the last quarter century, as pharmaceutical corporations and biotechnology firms in partnership with universities have muscled the government aside to become the chief patrons of research.
After conducting several small “test-run” clinical trials, Northfield commissioned a larger 1998 multicenter hospital trial, the Acute Normovolemic Hemodilution investigation, nicknamed the ANH study. Northfield convinced the hospitals and the FDA to allow it to investigate whether infusions of PolyHeme would safely allow the use of less donated blood during surgery. In this experiment, PolyHeme was infused into subjects awaiting surgery at institutions such as the University of Kentucky to repair an abdominal aortic aneurysm, a ballooned, weakened area in the artery wall that could burst at any time, causing massive, frequently fatal, hemorrhaging.
Two measures would have signaled success in the ANH study: the ability to use less, or no, donated blood during surgery, and more important, no higher a rate of adverse events, complications, or deaths when using PolyHeme than with blood. However, its Data Safety Monitoring Committee, an independent panel of experts that scrutinizes the safety and efficacy of an ongoing clinical trial, detected differences in the health status of the PolyHeme and control groups and asked for an in-depth analysis. This revealed that 54 percent of the patients who received PolyHeme suffered serious adverse events, ranging from heart-rhythm disturbances to pneumonia to heart attacks, but only 28 percent of patients who received blood suffered such problems. The results generated serious concerns about PolyHeme’s safety, concerns that were realized when ten of the eighty-one PolyHeme subjects died, in contrast to only four of those who had received blood. The difference was statistically significant, meaning that it was unlikely that it could have arisen by chance.
Northfield responded to these safety alarms by quietly closing the trial in early 2001. Its CEO, Stephen Gould, duly reported the excess morbidity and mortality to the FDA as was legally required, but he attributed it to putative disparities in the physical condition of the subjects. He also claimed that the doctors had bungled the study by infusing subjects who received PolyHeme with excess fluids.
In the wake of the ANH study, time was running out for PolyHeme. Its patent window was good for only about twenty years, and the longer it took Northfield to procure FDA approval, the greater the chance a competitor might emerge and the less time the company might have to enjoy any exclusive profits from its sale. A successful clinical trial was necessary for FDA approval, but the informed-consent process required that Northfield disclose to potential subjects the elevated rates of heart attack and death that dogged recipients of PolyHeme in the ANH trial. The revelation of these adverse effects would have made recruiting the necessary hundreds of subjects extremely difficult.
But Dr. Gould was an invitee to FDA meetings on the subject of HBOCs, where he presented the virtues of PolyHeme, including detailed technical arguments that fluid mismanagement and the profoundly ill condition of some subjects were to blame for the ANH medical disasters, not PolyHeme. These presentations apparently bolstered his arguments on behalf of PolyHeme enough for Northfield to convince the FDA to allow it to exploit the 50.24 exception and begin enrolling unwitting trauma patients in PolyHeme studies.
Although the CFR 50.24 law builds in special requirements and protections in an attempt to compensate for the loss of informed consent, the PolyHeme trial violated many of these and was mishandled so dramatically that the targeted communities and eventual subjects were left vulnerable.
These rules stipulate that the medical condition and time constraints must preclude eliciting informed consent from the subjects or their legally authorized representatives; that the subjects must require treatment for a life-threatening condition; and that known treatments must be “unproven or unsatisfactory.” In addition, the substance being tested must have prior study data suggesting that it provides “a direct benefit” to the subject. The regulations also mandate “community consultations” to inform people living in the targeted research areas—“communities” in CFR parlance—about the nature of the study that provides an opportunity for feedback from the affected residents.
But each rule was followed either in a desultory manner or not at all. The federal requirement that subjects’ medical condition must preclude the ability to give informed consent is usually taken to mean that the patient is unconscious or not lucid. Because trauma victims spiraling down into hemorrhagic shock tend to fall unconscious before help arrives, those who defend these studies claim there is no way to secure their consent. But investigators present no empirical evidence for this, and they provided no means for securing consent from those patients who retain or regain consciousness. “Has anyone actually done the research to establish that this is infeasible?” asked Peter Lurie, MD, of the watchdog group Public Citizen. “The company will complain that it is too difficult; this doesn’t make it impossible. Most reasonable people would want to be informed.” Martha Milete, for example, was conscious and remembers receiving PolyHeme in the ambulance, yet no researcher ever asked her permission or even notified her that she was a subject.
Despite the requirement that an experimental product such as PolyHeme can be administered under the 50.24 exception only if other approved treatments are unavailable, unproven, or unsatisfactory, researchers continued to administer PolyHeme for up to twelve hours after patients arrived in the hospital, where blood, the approved standard of care, was available. During this period subjects were unnecessarily, and according to 50.24 regulations, illegally, exposed to the risks of the experimental substitute because the desire to prove the product’s efficacy trumped the patients’ best medical interests. Nancy King, a lawyer and professor of social medicine of the University of North Carolina, pointed out that “the waiver of consent is permitted only when available treatments are unproven or unsatisfactory, and in a hospital, blood is neither.” Glenn McGee, PhD, director of the Bioethics Education Network (BENE), agreed and was largely responsible for Albany Medical School withdrawing from the PolyHeme study, as did other sites, including Boston University.
King had a personal as well as a professional interest, because “I live in the area from which the study drew subjects and could have become a subject.” She also discovered that the Duke University study in Durham, North Carolina, violated her state’s Patients’ Bill of Rights, which guarantees informed consent to medical-research subjects. She and her colleagues publicized these and other concerns in the American Journal of Bioethics, and the North Carolina study was suspended. Unfortunately the state medical board chose to waive the bill’s guarantee of informed consent in March 2005, and the study at the Duke site resumed.
Despite a legal requirement that the experimental treatment must offer a direct benefit to the unwitting subjects if approved, urban trauma patients could expect no such benefit even if PolyHeme were to work. Urban trauma victims typically reach emergency rooms within twelve to twenty minutes, so they can expect no benefit from the purported ability of PolyHeme to keep them from oxygen deprivation for hours. Moreover, the earlier studies, such as the ANH study, suggested that PolyHeme produced not a direct benefit but a direct hazard to patients.
The deaths and adverse events characterizing prior HBOC studies over more than a decade were equally alarming, according to an April 28, 2008, study titled “Cell-Free Hemoglobin-Based Blood Substitutes and the Risk of Myocardial Infarction and Death,” which was published in the Journal of the American Medical Association (JAMA).17 This meticulous meta-analysis observed sixteen blood-substitute studies that evaluated five different HBOCs with 3,700 subjects. Charles Natanson, MD, and his coauthors found that people who received HBOC transfusions suffered a 30 percent higher risk of death and fifty-nine heart attacks, as opposed to only sixteen of those who received no blood substitute. In short, all these blood substitutes caused higher rates of death and disability than did blood. All the paper’s data prior to the Poly-Heme trauma study were available to both the FDA and to Northfield and should have given them pause.
Equally questionable was the notion of “community consultation,” in which the affected community was to be informed of the study’s details and to offer feedback whose purpose was nebulous. This “consultation” was supposed to compensate for informed consent and was originally called, in an example of semantic duplicity, “community consent.” The usually precise regulatory language of the Code of Federal Regulations suddenly gave way to vaguely defined goals when discussing the nature of this community notification. For example, the purpose of the meetings is ostensibly to solicit the will of the affected community group, but the protocol language left it unclear whether attendees’ feedback could result in any change in the conduct of the trial or could result in its closure. In any event, most investigators did not seem open to such possibilities, because they presented the study to meeting attendees as a fait accompli, and researchers’ scribbled marginalia characterized attendees who asked hard questions as “hostile.”
In practice, community consultation was bowdlerized cheerleading for the study, carefully scripted and tightly directed, featuring hagiographic profiles of PolyHeme’s principal investigators. Presenters deployed uncritical PowerPoint presentations using a template that did not vary from site to site. Q & A PowerPoint slides offered rosy therapeutic promises and unsubstantiated safety reassurances, such as:
Q: Is PolyHeme safe?
A: In clinical trials to date, PolyHeme® has demonstrated no “clinically relevant” adverse effects. That is, they didn’t impact the patient’s safety or recovery.18
Proponents employed artful semantic twists to make forcing subjects into research seem ethically palatable. Research subjects were called “patients,” suggesting a nonexistent doctor-patient relationship, and PolyHeme was relentlessly referred to as a “treatment” and often as “safe.” In perhaps the most dazzling deployment of jargon, some PolyHeme researchers invoked the ethical term “equipoise,” which refers to the fact that in order for comparative research to be ethical, the investigator is supposed to inhabit a state of uncertainty regarding the relative merits of the substances being compared in the study. (If he knew one treatment to be superior, he would be obligated to offer only that treatment to the subject.) But the higher heart-attack and death risks for PolyHeme recipients of the ANH study, and similar injuries that plagued subjects in several earlier HBOC studies—all of which investigators were aware—called into serious question the vaunted “uncertainty” as to whether PolyHeme or blood was the safest treatment.
Worst of all, the few, sparsely attended community consultations failed dismally to notify affected communities of the study’s existence and their own vulnerability. Ross McKinney Jr., vice dean for research at Duke University School of Medicine, estimates that its community consultations “reached about 450 people,” which he acknowledged was only a “tiny fraction” of Durham County’s 267,000 residents. News reports in cities such as San Antonio and Denver19 televised what happened when a reporter posted at a busy downtown intersection at lunchtime asked every passerby, “Have you heard of the PolyHeme study?” Not one person had. While on the faculty of Albany Medical School, McGee conducted a telephone survey of ten thousand city dwellers and found that “essentially no one knew anything about the trial. Those who were presented with the possibility that they might get the substance were quite adamant that they would not want to be involuntarily enrolled—if at all.” (The CFR requirements do not quantify how many community members must be notified.)
When directed by some IRBs and regulators to provide a way for people to opt out of the study, Northfield settled upon a bright blue plastic bracelet inscribed with “I decline the Northfield PolyHeme study” in black block letters to alert emergency personnel. The catch, of course, is that no one could request a bracelet unless she had heard of the study.
The PolyHeme study proceeded, and in May 2006, undeterred by the excess heart attacks and deaths in the ANH study, Northfield began building a larger facility for spinning the dross of expired human blood into PolyHeme gold.
Meanwhile, most accounts note that only twenty-three test sites in twenty cities remained enrolled by the end of the trial. The centers that withdrew include those in Albany, Boston, Cleveland, and Johnson City Medical Center in Tennessee. At least six of the defectors, including those in Albany, Boston, and Durham (which later resumed the study), cited ethical discomfort with the study’s conduct and protocols.
But no center asked the important ethical question of whether minority-group members were more likely than others to be enrolled.
Reverend Charles Williams, president of the National Council for Community Empowerment of Detroit, thinks this was the case. “We African Americans have been treated like guinea pigs,” he declared. “We have suffered a history of research abuse and this is yet another instance.” The council’s website20 criticized the PolyHeme study as exploitative, and Williams held protests in early 2008 to air these concerns.
A few investigative journalists, notably Matt Potter of the San Diego Reader, also alleged that biomedical redlining directed the testing of Poly-Heme to disproportionately black and Hispanic neighborhoods. Potter forced the release of original memos and research proposals through the California Public Records Act that revealed how only the San Diego ambulances that consistently troll three of the city’s black and Hispanic neighborhoods were selected to distribute PolyHeme. “The experiment is targeted at several neighborhoods south of I-8, where many poor and minority residents are unlikely to have heard of the study,”21 he wrote. Detailed demographic data of each site may not exist. “Federal law normally requires that hospitals and researchers collect racial data on their subjects, but the law for private companies is unclear,” points out Heather Butts, JD, of Columbia University’s institutional review board. Columbia did not participate in the study.
Yet U.S. Census data buttress a damning racial-bias case against the PolyHeme trial. Only 10 percent of the 3,141 counties in the United States have “majority-minority” populations where minority-group members make up more than 50 percent of the populace. But according to U.S. Census data, 34 percent of the municipalities where ambulances carried PolyHeme22 were majority-minority. Richmond, Virginia, for example, is 57 percent black; Memphis is 61 percent black; Macon, Georgia, is 62 percent black; and Detroit is 84 percent black. Racial disparities also characterize some of the few rural PolyHeme sites, such as the village of Maywood, Illinois, which is 83 percent black. Of the twenty cities that ultimately completed the trial, thirteen—65 percent—had black populations considerably higher than the national average of 12.9 percent, and some had disproportionately high Hispanic populations as well.
In 2007, the Detroit Free Press reported that blacks and Hispanics constituted fifteen of the city’s sixteen PolyHeme subjects who were brought by ambulance to Detroit Receiving and Sinai-Grace hospitals.23 Both of the Detroit PolyHeme subjects who died were black.
In Indianapolis, which is only 25 percent black, Wishard Health Services scientists report that 59 percent of those who received Poly-Heme and 71 percent of the study control group were black, suggesting that even cities that are not majority-minority employed recruitment schemes that targeted people of color.
Another racial filter feeds the disparity. Although the scenarios in media discussions and on the Northfield website focused on trauma victims who had suffered motor-vehicle accidents or heart attacks, the trauma caused by urban gun violence contributed heavily to the pool of research subjects of color like Martha Milete. “The silence in regard to this,” observes Karla Holloway, PhD, a professor of English and law at Duke University who criticized the PolyHeme trial’s subject recruitment in The American Journal of Bioethics, “suggests a failure to acknowledge the statistical evidence that undoubtedly pointed them [researchers] to these population centers.”
National trauma-study data disseminated by staff at Wishard Hospital in Indianapolis validate this racial overrepresentation by revealing that 35 percent of the U.S. subjects who received PolyHeme were black and that 33 percent of study controls were black.24 This is 2.76 times the national percentage of African Americans.
In May 2009 the FDA denied PolyHeme approval, finding it not only devoid of clinical benefit but also unsafe. The April 2009 letter in which the FDA rejected PolyHeme’s Biologic License application concluded, “The safety data of all controlled studies reveal that the administration of PolyHeme places the patients at a higher risk of significant adverse events.… Based on the totality of the data in the application, FDA has determined that the data submitted do not support the proposed indication.”25 The subjects who received PolyHeme suffered three times the rate of heart attacks of those who received saline.
After the FDA denied Northfield’s ambitions for PolyHeme, stock share plummeted 51 percent, and it ceased construction of the new complex it had begun building in May 2006 to manufacture the blood substitute. On June 1, 2009, the firm filed for bankruptcy, and by July, Northfield’s stock was valued at only five cents a share and the company had lost its stock-exchange seat. By then, many of Northfield’s stockholders, convinced that the firm had lied to them about the deaths and heart attacks in the earlier ANH trial, had mounted a class-action suit against the company. But should they win, there may be little in the way of assets to disburse.
PolyHeme subjects may find justice equally elusive. “I was angry about being used without my permission,” Martha Milete says, “but I didn’t know that PolyHeme was harming people until I read it in the newspaper. I called a lawyer right away, but he wouldn’t help me. He said, ‘You didn’t die, so you have no case.’ ”
The death knell to Northfield’s hopes seemed almost anticlimactic, for by this time its failures were legion: the headlong rush to strip urbanites of their right to say “no” to medical research; the failure effectively to poll the public, to notify communities, or to adhere to even the diluted standards of the informed-consent exception; the withholding of information from stockholders, from the more exacting members of the press, and even from the very researchers who conducted the trials; and the credible accusations of racial targeting—all these must give us pause.
But we must add to these the shrouding of the excess deaths and heart attacks in the ANH hospital trial—and JAMA’s April 2008 excavation of the long, consistent history of HBOC deaths and deficiencies, which belatedly revealed that PolyHeme had promised nothing better from the start.
On the heels of Northfield’s demise, I learned that although PolyHeme may be history, our future includes a far more extensive assault on informed consent. A roll call of new medical innovations is even now being tested through the 21 CFR 50.24 exception. This is the $50 million Resuscitation Outcomes Consortium (ROC), a broader, more ambitious, and potentially more profitable network of eleven regional sites that are conducting nonconsensual studies, again in emergency scenarios. The University of Washington Clinical Trial Center coordinates the network, and the study chair, responsible for the overall scientific leadership of the research, is Myron L. Weisfeldt, MD, chair of medicine at Johns Hopkins University. Each research site is fed by at least one medical center, and some of these centers conduct several investigations.
Reading ROC’s brochures and research proposals evokes déjà vu. As in the PolyHeme study, victims of trauma—people injured in car accidents, shootings, and cardiac arrest—will be “enrolled,” and the studies will be conducted by paramedics and EMTs in ambulances and other emergency settings. As in the PolyHeme study, the likely unconsciousness of victims is invoked, without empirical evidence, to demonstrate the futility of obtaining consent from victims or their families.
The ongoing ROC plans to investigate approaches to neurological damage in six thousand trauma victims and to evaluate resuscitation methods on fifteen thousand people who have suffered sudden cardiac arrest. At least eleven Level I (top-tier) trauma centers in the United States and Canada host these studies, including hospital emergency departments in Seattle, Portland, San Diego, Dallas, Birmingham, Iowa, Milwaukee, Pittsburgh, Ottawa, Toronto, and Vancouver. Each Level I trauma center tests at least one arm, and in many cases several arms, of the trial. Thus a specific site may conduct arms of several individual studies, such as tests of saline solutions and tests of CPR assistive devices.
As in the PolyHeme study, the ROC is testing ways to reverse or to prevent shock that is triggered by traumatic injuries. And, as in the PolyHeme study, the profitability of corporate monopolies is at stake. Among its research protocols, for example, is a study to test a patented valve device made to assist in cardiopulmonary resuscitation, or CPR.
Unlike the PolyHeme study, the ROC studies enroll adolescents as young as fifteen, whose medical fates can now be determined by a computerized coin toss rather than by their parents.
Daniel Nelson, PhD, had reassured me in 2008 that 50.24 research was “very rarely used,” but North Carolina lawyer Nancy King mused, “The emergency-consent exception is supposed to carve out a very narrow window, but that narrow window seems to be expanding.” In mid-2009, ROC study chair Weisfeldt of Johns Hopkins told me, “The ROC has already enrolled seven thousand subjects,” eclipsing the 720 subjects in the PolyHeme study. The ROC study’s published goal is the enrollment of twenty-one thousand people by 2012.26
Like the PolyHeme trials, the ongoing ROC studies have lost both scientific and ethical footing at times, but, like Northfield’s Dr. Gould, ROC investigators seek to cast these in the most positive light possible. One ROC experiment, for example, injected highly concentrated salt water into trauma victims’ blood vessels. Brain injuries are usually treated by giving normal saline solution, but this multicenter ROC study tests concentrations of salt that are dramatically higher than the saline concentration in our bodies. Some arms of the study also randomly add dextran 26, which consists of daisy-chained sugar molecules that are intended to reduce blood clotting. The dangers of such a highly concentrated saltwater solution, including high blood pressure and seizures, have long been known, and dextran creates further risks with its potential to trigger allergic reactions. These concerns were validated on August 25, 2008, when the study’s Data and Safety Monitoring Board (DSMB) suspended research because of concerns about patient safety.
During our discussion, however, Weisfeldt portrayed this suspension not as a safety issue but as a decision made because of “futility. There was so little difference between the study’s two arms that there was no reasonable prospect that it would show a benefit.” Only when pressed did he admit that the DSMB had suspended the study.
Weisfeldt insisted, “Seventy percent of people approve of the [ROC] study, according to a telephone survey.” But once news coverage of the ROC familiarized some area residents with the study’s intentions, the researchers, despite their claim of a mandate, were besieged by nervous potential subjects who wanted no part of it. In Portland, Oregon, seven hundred residents demanded and received opt-out bracelets. Researchers in the Seattle arm of the study were “overwhelmed” by the demands for bracelets: by June 2007, the high demand had rendered them unavailable for more than a year.27
But perhaps the wary needn’t have bothered: ROC publications warn, “Even if you opt out, there is no guarantee.… You might still be enrolled in the study.” When asked to clarify, Weisfeldt stopped short of an assurance: “In most cities, if someone objects, they are given a bracelet. Every effort is made to recognize the bracelet if he or she is a candidate. We do our best to honor that.”
Why, I asked Weisfeldt, is it ethical to conduct research on people who have not given their consent? “It is the only way scientifically to learn what the best treatment is, by doing clinical trials. The [ROC] study is completely legal and is approved by IRBs.” But does that make it right? “We live in a society that goes by the law.” I explained that I was not questioning the legality of the study, but its ethics: Is it right to conduct research with no provision for consent? Weisfeldt paused, then said, “I’d better not answer that question.”28
Withholding the right to consent, whether via 50.24 research, presumed consent, or frank deception, speeds research studies that would be difficult or perhaps impossible to recruit if potential subjects were actually informed of the risks they faced. This can be convenient and profitable for the pharmaceutical and biotechnology companies that embrace it. But why do academic researchers who conduct the studies for pharmaceutical companies put aside their sacred ethical obligation to inform patient-subjects? There is probably more than one answer to this question, but when a researcher or medical institution is dependent upon the largesse of a corporation, they risk behaving as if they were obedient employees rather than independent investigators.
An exchange of press releases by Northfield Laboratories29 and Johns Hopkins University, whose researchers were being paid—well—to test PolyHeme, illustrates this tension by revealing the company’s attempt to control a Johns Hopkins researcher’s statements and behavior.
In early 2006, when Northfield was confronted about its failure to disclose the excess heart attacks and deaths among PolyHeme recipients in its ANH study, the company responded with a press release announcing that principal investigator Dr. Edward Norris of Johns Hopkins University would hold a press conference at a major medical meeting, during which he would explain that no deception existed because he had indeed been informed of the deaths by Northfield. The release assured readers that Norris would also deliver Northfield’s party line on the subject—that researcher error, not PolyHeme itself, had caused the heart attacks and deaths.
However, Dr. Norris was apparently less tractable than Northfield thought, because on March 23, 2006, Johns Hopkins countered with a press release of its own:
Contrary to a statement made as part of a press release by Northfield Laboratories of Evanston, Ill., on Feb. 22, 2006, a Johns Hopkins Medicine faculty member, Edward Norris, M.D., is not presenting information about a clinical trial of the company’s blood substitute PolyHeme at the annual meeting of the Network for the Advancement of Transfusion Alternatives (NATA) in April 2006; was not given access to full study results from Northfield; and does not and cannot substantiate Northfield’s claim that PolyHeme was unlikely to have been the cause of 10 heart attacks and 2 deaths in patients receiving the blood substitute as part of a clinical trial that ended in 2000.
Johns Hopkins Medicine has asked Northfield to retract its February 22 release and to reissue a corrected version.30
Unfortunately, not all clinical researchers display such independence: They have motives to comply with a corporation’s directives. Researchers want to recruit subjects and complete their research in the timely manner required by federal regulations. They also presumably want to validate effective treatments and preserve lives and often promulgate consequentialist arguments for cutting ethical corners—that the laudable ends justify the dubious means. For example, no one questions that a blood substitute would be a stellar advance if it worked safely, or that research to definitively establish the superiority of one resuscitation technique over another would be a good thing. So researchers often focus on these speculative benefits, not the certain abuse of impressing subjects into research.
Moreover, once they have carried out the research study, academic researchers have sometimes found themselves cut out of the process of data evaluation and publication, with no power to influence the interpretation of the study results. Northfield’s policy of withholding information from the more critical members of the press and ethics community even extended to some of the doctors who conducted its research for them.
For example, despite repeated pleas from PolyHeme researchers Drs. Ronald M. Fairman and Albert Cheung at the University of Pennsylvania and from T. J. Gan, MD, of Duke University, Northfield did not provide data about the entire study to participating researchers and failed to publish the data, leaving investigators with only the limited information they could glean from their own subjects. Dr. Cheung told the Wall Street Journal that Northfield’s Gould agreed to meet with doctors of the twenty-one hospitals that had conducted the ANH study in Philadelphia, then canceled the meeting at the last minute.31
And as we have seen, the hefty payments to medical institutions by pharmaceutical and biotechnology companies that are eager to fund promising research provide a financial incentive as well.
But the nonconsensual subject recruitment that has undergirded PolyHeme, the ROC, and many other recent investigations is an appalling ethical reversal. Worse, this forced participation in medical research has been adopted with an utter lack of transparency, so it is even harder to understand the tolerance of many medical ethicists for this broadening abandonment of informed consent.
Some undoubtedly share the consequentialist focus of researchers who fixate on the prospective benefits of a successful blood substitute or other desirable innovation. But naiveté or baser motives may drive some ethicists to defend research without consent.
Others have mulled the question before I: Are some corporations buying the ethics they need? Does pharmaceutical and medical-company funding influence the views of some ethicists, as it does some of the medical professionals and institutions discussed in Chapter 4? It is heresy to suggest that some ethicists are influenced by drug and device makers’ money, even though they may not be conscious of their seduction. However, Carl Elliott, an ethicist at the University of Minnesota, suggests that we ask
… why bioethicists at the University of Toronto take funding from GlaxoSmithKline, Pfizer, and Merck to write editorials on bringing biotechnology to the developing world. Or why the University of Chicago’s MacLean Center for Clinical Medical Ethics cosponsored a recent conference with Pfizer, Merck, and PhRMA, the pharmaceutical industry trade organization, on inequities in American health care. Or why bioethicists at the University of Pennsylvania take money from Pfizer to write an article explaining why physicians should not accept gifts from companies like Pfizer. We may take industry money, bioethicists argue, but we’re not industry stooges. We’re doing God’s work.32
Elliott’s insight is critically important: pharmaceutical companies sometimes give ethicists money not because they expect unalloyed support, but as a component of their “third-party strategy.” Independent bioethicists are not expected to sell the companies’ products—which in this case are an ethical defense of its drug, manufacturing, or marketing actions—directly. This is because the ethicist’s independence is what makes her opinion valuable, lending it power and making her an opinion leader. Instead she indirectly rationalizes, supports, or defends the company’s actions in grand rounds, radio programs, scientific presentations, journal articles, or even in discussions with colleagues. Perhaps she merely raises questions about the condemnation of actions such as Northfield’s no-consent studies or the ROC’s testing of a proprietary device on unwitting subjects: this subtle support is valuable to the corporation, too.
Even if the ethicist devotes much of her writing to assailing a company’s practices, any isolated statement of support or approval can be taken out of context and publicized by a company that may also publicize its financial support of the ethicist. Some of these handpicked ethical “opinion leaders” are undoubtedly convinced of their own impartiality despite the drug makers’ checks in their back pockets, and that is all to the good as far as the industry is concerned.
Even revered bioethicists’ groups such as the American Society for Bioethics and Humanities and the American Society of Law, Medicine and Ethics endorse Pharma’s payments to ethicists, displaying the same naiveté shown by publishers and readers who accept medical-journal review articles written by “objective” scientists in the pay of Pharma. In Chapter 4, we read how Frederick K. Goodwin, MD, defended himself against accusations of purchased bias by arguing that he took money from so many pharmaceutical companies that his opinions favored no one firm. Similarly, ethicists are sometimes encouraged to take money from several pharmaceutical companies to deflect accusations of bias.
The financial benefits that some ethicists enjoy have seduced them into defending, or at least failing to criticize, the erosion of informed consent and other coercive policies that benefit drug makers’ bottom line. I hasten to add that I do not impute an impure motive to all ethicists who defend dispensing with informed consent: fair and reasonable scholars can disagree. I do suggest that some may find their ethical vision clouded by pharmaceutical-firm money.
Ethicists are only one of the many groups vulnerable to seduction by Pharma’s money. We have seen how physicians are particularly vulnerable because drug and device makers pay fully half of the $1.4 billion spent annually on continuing medical education for U.S. physicians. So, increasingly, are medical-advocacy organizations as drug makers fund patient support and advocacy groups—as well as some prominent bioethics centers.
And what of Martha Milete, the story of whose assault opened this chapter? How has she fared, and what does she think of the increasing tendency to bypass the consent of research subjects in emergencies?
“I had suffered so much trauma,” she recalls with a trembling voice. “My whole body was open from my breast all the way down my torso and they couldn’t close me up for a long time. I returned to the hospital for several major surgeries, and the impact of the gunshot tore my retina, so I had to go in for eye surgery. My vision is pretty good now, I can see, but little waves make it seem as if something is always going across my field of vision.
“My body is disfigured now, I have incisions that cross my whole torso, but I am grateful to be alive and my clothing can cover my scars—imagine if I had been shot in the face! I’m disabled: there are a myriad of things wrong with me, and I can’t do half the things I used to do; I just have to accept that.”
Perhaps the psychological scars are the worst. Milete still cannot sleep well, haunted by nightmares and fear despite the dog she bought for protection. Her assailants were never apprehended.
“But I’m walking. I’m able to live in my own home. I’m just glad to be alive and independent. But when I think of the research they did on me, I lose my peace of mind: How can they do that? How can they use you in an experiment without telling you what they want to do?”