My brief life as a laboratory animal began one winter’s day with a call from a researcher at Northwestern Memorial Hospital. For six weeks, she wanted me to allow a phalanx of scientists to track my eating habits, my daily routines, and the ebb and flow of my blood sugar. In return, she told me, I would be eligible for free aerobics classes in a “dazzling” rooftop exercise studio with “breathtaking” skyline views. She also mentioned, glancingly, the 33 blood draws.
I was intrigued. As a health writer, I’ve always considered clinical trial volunteers as the unsung heroes of medicine, and I did want to serve the cause of scientific progress. Less altruistically, I welcomed the opportunity to get the inside track on new research on my own particular maladies. So I agreed to participate, becoming an accidental tourist in the perplexing world of clinical trials-controlled tests of drugs and other therapies.
It’s a world that’s expanding rapidly: An estimated 60,000 to 70,000 trials are under way at any given time in the United States. During the past decade, the average number of volunteers needed for each new drug study has tripled; it takes an average of 4,237 human subjects to bring a new drug to market, and no one knows just how many Americans are acting as guinea pigs in other studies.
It’s also a world where big business intersects with high-stakes medicine, confronting researchers with knotty ethical decisions and unwary volunteers with stark choices about their own health, as I was to learn firsthand.
The Northwestern researcher called because I had developed gestational diabetes in my last pregnancy. Though it’s a temporary problem, usually clearing up immediately after childbirth, it greatly increases the odds that I will someday develop permanent diabetes. In this study, designed to show that even limited exercise improves the body’s ability to process sugar, a small group of high-risk women like me were to test the validity of the standard diabetes-prevention advice: Eat less, exercise more. Half the group was expected to remain on their normal routine of sloth and Fritos, while the other would go on a supervised six-week diet and exercise plan, including the free aerobics. I hoped to luck out and get six free weeks of spa-quality health supervision. Instead, I was slotted into the Fritos group.
The trial began in a cramped hospital room, where a basketful of empty syringes waited by the bed. I was handed a thin sheaf of papers explaining the possible risks–in this case, mainly bruises and punctures from IV lines. After I signed, they laid me out on a narrow mattress to run lines into each arm. Since my thin, curling veins are hard to find, it took several painful tries before the obviously exasperated nurse could insert the needles properly. Finally, she fed a baby bottle of glucose into my arm, which slowly chilled my whole body; it tingled unpleasantly. Over the next two hours, she pulled 22 vials of blood from a main line plugged into my vein.
Afterward, a businesslike researcher measured my body-fat percentage with electrodes attached to my foot and hand. The dial registered an appalling 38 percent fat. Horrified and humiliated, I wanted to run home and embark on an emergency fast, but the researcher was insistent: Any unusual change in my normal eating or exercise patterns would skew the study results. In other words, I had to stay fat for science.
Once the six weeks were up, the researchers started the tedious process of analyzing all their data. I embarked on an equally arduous task, a new health regime based on the groundbreaking principles of eating less fatty foods and burning up more calories. Over a year, I shed 30 pounds.
I never learned whether the diet and exercise plan worked as well for the study’s other subjects; as in many small preliminary trials, the final results were not published. But the experience had served as a loud wake-up call, alerting me to the looming threat of diabetes. So I was initially enthusiastic when, in 1997, I was asked to volunteer for another human experiment at Northwestern.
This was a big one. The Diabetes Prevention Program was to be the first-ever nationwide study on preventing the disease; 27 medical centers–funded by $150 million from the National Institutes of Health–were trying to sign up more than 3,000 patients willing to stick with the trial for up to six years. A computer would randomly assign those volunteers into four groups. The first would get intensive counseling on diet and exercise; the second would take metformin, a drug already used to treat active diabetes; the third would take a new drug, troglitazone; and the fourth group would receive a placebo–a dummy pill containing no active ingredients. The placebo is a necessary part of clinical trials because of the well-established but mystifying fact that the mere act of taking a pill can help a patient improve dramatically.
I agreed to go through a preliminary screening, a set of blood tests that showed I met the study criteria. The next step would be randomization, the computerized coin flip that would assign me to a treatment group. Although enrollment in clinical trials is completely voluntary, I could sense the researchers’ eagerness to sign me up; when I wavered, they called repeatedly to ask whether I’d made up my mind.
The pressure to find volunteers willing to sign up for trials can be tremendous. When a drug company tests new medications, millions of dollars in potential profits are at stake. There can be substantial rewards for the researchers too. Pharmaceutical companies sometimes offer doctors big cash bounties for every patient they enroll in a trial. In one widely reported California case, a doctor received $1,600 for enrolling a patient in a prostate drug trial, even though the patient had no prostate problems.
Most doctors who get paid for trials don’t tell their patients about the drug companies’ fees, and most patients, advocates say, aren’t likely to scrutinize their doctors’ motives. “There is a serious coercion problem,” says Michele Russel Einhorn, director of regulatory affairs at the Department of Health and Human Services’ Office for Human Research Protections. “It’s hard to disagree with your doctors. I’m an attorney, and I’ve been intimidated by my own doctors. I’m sure there’s a huge population out there that’s even more intimidated.”
Legislation was introduced in the U.S. Senate last year that would require doctors to disclose payments they received for referring patients to clinical trials; the measure died in committee. While there are federal guidelines intended to protect patients from potentially unscrupulous researchers and drugmakers, the government’s monitoring of clinical trials is based largely on an honor system.
In my case, I felt confident that the researchers weren’t trying to sign me up just to line their pockets. Yet even though everyone connected with the trial had impressive credentials, I still hesitated. I had been willing to put the cause of science ahead of my own health concerns for six weeks. But six years? Whose body is this, anyway?
To help me weigh the pros and cons, I consulted my own doctor. When I showed her the lab results from the screening, she looked for my blood-glucose levels–the best available indicator of my progression toward diabetes. They weren’t there. “Then what’s the point of this?” she yelped. I explained that the glucose numbers weren’t disclosed because the experiment was double-blind, meaning that neither patient nor doctor could know who was taking active medication. A drastic drop in my blood sugar would almost certainly mean I was taking one of the “real” drugs. I could afford to monitor my blood sugar privately, but that would almost certainly affect my behavior–and the study’s results. Frankly, if my blood sugar kept rising and I knew about it, I would drop out of the study and take immediate action to bring those numbers down.
I began to appreciate the kind of dilemma faced by volunteers who have serious health problems: If I chose to stay blind to my numbers, I would be ignoring my own health for the sake of preserving the integrity of the trial.
Volunteers quickly learn that the needs of science come before the subjects’ individual health. One recent trial of a hepatitis C drug had patients up in arms when researchers refused to disclose their viral load, a measurement of how well the drug is working. Patients wanted to know whether the debilitating side effects they were enduring from the drug were paying off in better health; researchers feared that patients who weren’t benefiting from the drug would drop out, ruining their study. Many of the volunteers couldn’t afford to buy the drug outside the study, so they felt stuck and exploited.
“It’s the most outrageous thing,” fumes Dr. Peter Lurie, a medical researcher for the Health Research Group for the consumer advocacy organization Public Citizen. “The medical system leaves so many people uncovered that they would stoop to tossing a coin in order to get medical care.”
My ever-ethical doctor told me I’d have to make my own decision about joining the trial, although her raised eyebrow gave me a fairly clear reading of her personal opinion. Despite her obvious lack of enthusiasm, I couldn’t make up my mind. If I declined, would I miss the chance to get cutting-edge treatment that might keep me healthy for years? But if I signed up, could one of these medications end up doing more harm than good?
I was also troubled by the idea of taking medication for six years. What if I got pregnant, or an unexpected side effect permanently injured my health? Perplexed, I pored over the glossy handouts explaining the study’s risks and benefits. “No statistically significant adverse events have been noted,” assured the information sheet. I was still uncertain, so I asked the researchers to send me the drugs’ package insert, the detailed fact sheet the FDA requires for every drug. According to the insert, troglitazone seemed very effective in helping adult diabetics. But in tiny print, it mentioned “liver function test abnormalities.” Alarmed, I did a little more research, and found that some health advocates were very worried about the drug’s safety.
It’s not surprising that the researchers’ materials didn’t highlight the rare but frightening side effects associated with the drug; if the informed consent began with “WARNING! This just might kill you!” who would sign up? However, unsuspecting patients may assume the researchers’ need for willing subjects will always be counterbalanced by health advocates who make sure potential volunteers get all the information they need to make a truly informed decision.
It’s hard enough to make a wise decision about joining a trial when you’re well and well-insured, as I was. When you’re sick and desperate, a clinical trial can look like your only shot at decent medical care. The advertisements recruiting trial volunteers often promise free tests and free drugs. One Texas center studying pain relief for menstrual cramps offered participants no-cost physicals, pelvic exams and Pap smears, plus a cash payment for those who completed the trials. CenterWatch, an information services company for the clinical trials industry, maintains a Web site listing trials seeking subjects; volunteers’ stipends can go as high as $1,500.
Even beyond the question of safety, I was troubled by a visceral aversion to the whole notion of “randomization,” using a computer to decide which patient gets what treatment. Because it avoids human bias, scientists consider randomization a vital part of any clinical trial. But the random selection made me feel that I was abdicating control over a major medical decision to a computer program.
That worry is common, as shown by the recruitment problems plaguing one current national study, Surgical Treatments Outcome Project for Dysfunctional Uterine Bleeding (STOP-DUB). The $6.6 million federally funded study, which compares treatments for out-of-control menstrual periods, at first seems almost absurdly anti-patient. The study’s volunteers must agree to let a computer decide whether they keep or lose their uteruses. That may be why the study, which began in 1997, has recruited only 194 of the 250 volunteers needed.
But in listening to Kay Dickersin, the trial’s director, it becomes clear that the study design reflects serious thinking about women’s lives. Dysfunctional uterine bleeding accounts for as many as 40 percent of the 590,000 hysterectomies performed in the U.S. each year. For patients who want to avoid hysterectomies, there are some newer options: laser and heat treatments that stop the bleeding by destroying endometrial tissue, the inner lining of the uterus. But until now there has been no research showing which treatment provides the best outcome. So a woman who chooses a treatment on her own, or lets her doctor make the decision for her, is basically taking a shot in the dark.
Dickersin says she’s sympathetic to patients who feel that they’re letting a computer make one of the biggest health decisions of their lives. But, absent solid research, she says she doesn’t see how that’s different from letting a doctor make the call. “At this point, we truly don’t know what treatment is best,” she says. “For most surgeries, typically, there haven’t been randomized trials. So you’re really just relying on some doctor’s hunch. Why is that better than entering a trial, where a computer decides which hunch surgery you’re going to get?”
I spent a few sleepless nights trying to decide whether I was willing to let a computer prescribe a drug that might prove just as harmful as diabetes. I did feel a real responsibility to help science figure out the best way to treat women like me. But I didn’t know how I could go on for the next six years, not knowing whether I was slowly developing a devastating illness.
Finally, I opted not to volunteer. Later, I learned my fears of unknown side effects were justified; the troglitazone treatment was discontinued in 1998, when one of the study’s volunteers died of liver damage apparently caused by the drug, which has since been pulled from the market.
Even with the benefit of hindsight, I’ll never be sure that I was justified in not participating. Despite that single tragic death, the Diabetes Prevention Program is still going on and is being hailed as a promising, powerful weapon in the fight to prevent diabetes–thanks to the thousands of volunteers who decided to accept the dangers and join up.
By their very nature, clinical trials require their subjects to venture into unknown and possibly dangerous territory. Yet without them, we end up practicing medicine with our eyes shut. The desperate need for trials, and trial volunteers, was underlined by a recent study of bone marrow transplants for breast cancer patients.
Twenty years ago, researchers trying desperately to save the lives of women with advanced breast cancer came up with a new idea: If a limited dose of highly toxic chemotherapy drugs could kill some cancer cells, could a big dose kill them all? The trouble was, the massive chemo doses would kill the patient’s bone marrow. So the doctors decided to harvest healthy marrow before the chemo treatment and transplant it back afterward. It was a scary procedure. In the early days, nearly one out of five patients died from the heavy drug dose alone. Many who survived suffered permanent injury, including heart damage. But the early results were promising: Of the handful of patients who went through the experimental procedure, about 50 percent survived with their cancer apparently in check.
That was enough to convince many doctors and patients to try the new treatment. So there were few takers when researchers tried to recruit breast cancer patients into randomized clinical trials–half getting the high-dose chemo, half getting the usual treatment. Women who saw the bone marrow transplant as their last chance simply weren’t willing to let a computer consign them to a painful and untimely death. The expensive transplants became standard procedure, even though there was no proof that they actually worked. For patients, hope was enough.
It took years for researchers to find adequate numbers of women willing to enroll in the randomized trials. But finally, last May, data from the first five trials were announced, And the results were devastating to those who believed the transplants were lifesavers. Four of the five studies showed no difference in survival between the two groups; the results of the fifth study, which showed some benefit from transplants, were questionable. “Had these procedures been performed within a randomized clinical trial, we would have had the answers some time ago,” the National Breast Cancer Coalition fumed in a news release. Instead, some 30,000 desperate women went through a grueling procedure that, apparently, did nothing to lengthen their lives.
It’s frightening to entrust your health, and possibly your life, to random chance. But I think it’s even more frightening to realize how scanty our medical knowledge really is. It’s hard to imagine yourself, like the STOP-DUB volunteers, donning a hospital gown, lying down in an operating room and allowing a doctor to slice open your abdomen and end your reproductive life, all on the basis of a computer’s whim. But it’s just as painful to imagine making that decision in an information vacuum. For Dickersin, the question finally boils down to trust-trust in the doctor who refers you to the study, in the researchers who designed it, and in the institution that’s looking over their shoulders.
“We have to trust somebody,” Dickersin says. “The bottom line is, how are we going to get the real answers?”
WHO’S STUDYING THE STUDIES?
NEW PRESCRIPTIONS ORDERED UP FOR REVIEW BOARDS
The government requires every clinical trial to be approved by an Institutional Review Board, created to ensure that the study’s design meets basic safety and ethical standards. However, investigations by the Department of Health and Human Services’ inspector general indicate the boards are often unable or unwilling to do that job.
An inspector general’s report in 1998 found that the average IRB at an academic medical center was reviewing 297 proposals a year, about two per working day. “They review too much, too quickly, with too little expertise,” George Grob, deputy inspector general at HHS, told a congressional subcommittee last May.
Although the boards are supposed to include one layperson and one outside scientist to represent patient interests, those rules are often ignored, the report stated. At university-affiliated hospitals, IRBs are mainly made up of scientists from that institution–colleagues of the doctors whose research they’re supposed to scrutinize. The ethical waters get even murkier when private research centers hire for-profit IRBs, paid by drug companies whose studies they’re reviewing. To consumer advocates, it seems obvious that a for-profit IRB would want to avoid a reputation for being overly strict. Another risk is that researchers will go IRB shopping, looking for a board lenient enough to let a faulty study go forward.
After the inspector general’s scathing report on the review boards, HHS took action against several major institutions, including two in Chicago, citing problems with their research oversight process. In 1998, the National Institutes of Health suspended all federally funded human research at Rush-Presbyterian-St. Luke’s Medical Center. The University of Illinois at Chicago was hit with a similar suspension in 1999, and it voluntarily halted all privately funded research as well.
“I think it was a real wake-up call to all IRBs all over the country, that they have to be given better resources and look much more closely at human subjects research,” said Betty Musburger, an attorney specializing in health and infertility law, who was named a community member of a UIC review board in the wake of the suspension.
She said the university has greatly improved its IRB oversight. Before the suspension, the institution had one IRB that met once a month, at which members had to review a daunting pile of research proposals. Now two IRBs each meet twice a week, and all members have undergone rigorous training.
Musburger notes that while the university’s old surveillance system may have been sketchy, it was hardly unique. “What happened at UIC or Rush could have happened at almost any university in the country,” she said.
To give volunteers more protection, the HHS upgraded the Office for Protection From Research Risks, renaming it the Office for Human Research Protections. The new office will focus solely on the 4,000 HHS-funded institutions that conduct research on human subjects; previously, the office was expected to oversee both human and animal research. A new Human Research Protections Advisory Committee is being formed to help the office balance patient protection with research needs.
In May, HHS Secretary Donna Shalala announced new ethics and conflict-of-interest guidelines for researchers and IRBs. She also is asking Congress for the power to fine universities up to $1 million for research violations.
