NEW BALTIMORE, Mich. — For the last three and a half months she was pregnant, Tammie Snyder had a small medical device strapped to her thigh. It pumped a drug called terbutaline through her body to prevent her from going into labor too soon.
On Sept. 17, 2002, Snyder gave birth to two healthy girls. Within days, however, her lungs filled with fluid, her heart began to fail and she was told she might need a heart transplant. She recovered, but she's been told she can never have a baby again. Her heart wouldn't stand the strain.
Terbutaline is an asthma drug, and the Food and Drug Administration hasn't approved its use to prevent premature labor. The FDA has warned doctors that the treatment is "potentially dangerous" and may not be effective. Snyder said her doctor never told her about the warning or that the FDA had approved terbutaline only to treat asthma.
A six-month Knight Ridder investigation has found that patients nationwide are being injured and killed as doctors routinely prescribe drugs in ways the FDA never certified as safe and effective.
Moreover, these unapproved prescriptions are soaring. Over the last year, 115 million such prescriptions were written, nearly double the number of five years ago, an exclusive Knight Ridder analysis of prescriptions for the country's top-selling drugs found.
The practice, called off-label prescribing, often is driven by questionable research, aggressive drug-company marketing and cavalier doctors, and condoned by tepid regulators.
Doctors are giving their patients epilepsy drugs for depression and hot flashes and to help them lose weight. They use antidepressants to treat premature ejaculation and pain, and powerful antipsychotics for insomnia and attention deficit disorder. High blood-pressure pills are prescribed for headaches and anxiety; antibiotics are used to treat viruses.
Some drugs, in fact, are sold mostly for unapproved purposes. Eight out of 10 prescriptions for the epilepsy drug Topamax aren't for epilepsy. Thalidomide, the notorious morning-sickness drug that caused horrible birth defects and ushered in today's FDA drug-safety rules, today is on the market, and 99 percent of its prescriptions are off-label.
Knight Ridder reviewed 15 top-selling classes of drugs and found that some, such as cholesterol medicines, rarely are given as unapproved treatments. But three-quarters of anti-seizure medications are prescribed off-label, as are nearly two-thirds of antipsychotics and about one-quarter of antidepressants, the analysis found.
For patients with rare, intractable or fatal illnesses, off-label prescribing is sometimes appropriate. In other cases, there may be gold-standard studies backing an off-label use. But doctors routinely are choosing unapproved therapies that are questionable at best.
The practice is perfectly legal, widely accepted and defended by doctors and the American Medical Association — and it's taking a toll.
Victims of off-label prescribing whom Knight Ridder interviewed have suffered heart attacks and strokes, had permanent nerve damage or lost their eyesight. Most said they never were told that the FDA hadn't approved their treatments.
Based on the FDA's own data, Knight Ridder estimates that at least 8,000 people became seriously ill last year after taking some of the nation's most popular drugs off-label. The true number is likely to be many times higher.
"Sometimes it may help, sometimes it may do more harm than good and sometimes it may kill people," said Arnold Relman, a former editor of the prestigious New England Journal of Medicine.
Despite the rise in off-label drug use, the FDA has done little to discourage it, and is considering whether to allow drug companies greater leeway in pushing unapproved therapies.
George Murphy's hands, made strong by years of climbing utility poles for Houston Lighting & Power, shake with tremors. His legs, now rigid, shuffle as he pushes his walker through his studio apartment in Deer Park, Texas.
"I wish I didn't have to use this thing," he said as the walker snagged on his recliner while he was showing off his Army dog tags from World War II, a plaque for 40 years of service as a Mason, his Shriner fez and the oil paintings his wife did a few years before she died.
Murphy, now 85, began having the tremors last year after he had a series of stroke-like attacks while taking Risperdal, a powerful antipsychotic drug that the FDA has approved only for treating schizophrenia.
Murphy's family practitioner in Pasadena, Texas, Dr. Dennis Yaworski, prescribed Risperdal for an off-label purpose: "cancer phobia," according to case notes from an office visit on Sept. 9, 2002.
The drug's maker, Johnson & Johnson, has marketed Risperdal heavily to doctors who treat elderly patients.
In 1999 the FDA cited Johnson & Johnson for downplaying the drug's risks to the elderly and making false and misleading claims that it could be used not just to treat schizophrenia, but also "for psychotic symptoms associated with a broad range of disorders."
While doctors are free to prescribe as they wish, the FDA prohibits drug-makers from marketing unapproved treatments.
Despite the FDA's action, Risperdal has become a popular off-label treatment for Alzheimer's disease and dementia. About 670,000 such prescriptions were written last year, up more than 350 percent from 1998, the Knight Ridder analysis found. Sixty-five percent of Risperdal's prescriptions last year were for unapproved treatments, generating $929 million in retail sales.
Murphy had been a familiar figure at his doctor's office, complaining of stomach pains, arthritis and the like, often convinced that any pain was a sign that he was dying of cancer. But he was otherwise fit, his daughters said, enjoying family dinners, a recent trip to an alligator festival and an active social life at his independent living center.
The FDA hasn't approved Risperdal for the treatment of hypochondria. But Murphy's family members said they weren't told this. Yaworski declined to be interviewed.
Within a month of starting the drug, Murphy had the first in a series of stroke-like events, called transient ischemic attacks, according to his family and hospital records. During the worst of them, on Nov. 14, 2002, Murphy suddenly couldn't walk, his speech became slurred and his face drooped. He became easily confused, and doctors at the hospital added dementia to his diagnoses.
Attack followed attack, but Murphy kept taking his Risperdal. He was still afraid of getting cancer, his family said.
Then in April 2003, Johnson & Johnson sent a letter to U.S. doctors warning that Risperdal may be associated with an increase in strokes when prescribed off-label to elderly dementia patients.
The public warning came nearly two years after the drug maker privately alerted the FDA that there was a problem with Risperdal, agency officials said in response to questions from Knight Ridder. It came six months after drug regulators in Canada issued a similar warning and urged doctors in that country to reassess their use of Risperdal to treat dementia.
FDA officials, in a written statement, said it took several rounds of questions to the drug maker before they had enough evidence to have the drug company issue the warning. Johnson & Johnson, based in New Brunswick, N.J., had no comment.
Murphy and his family have sued the drug company, which in court filings denies any wrongdoing. His daughter, Robbie Murphy, said: "Our father has been taken away from us. Basically the last enjoyable times he could have with us are gone."
Dr. Raymond Woosley, the vice president of health sciences at the University of Arizona, said off-label prescribing puts patients at greater risk than when doctors follow a drug's FDA-approved directions.
"I have no doubt about it," said Woosley, who also is the director of one of the national centers for drug research established by the federal government. "The caveat is we can't quantify it."
Few have even tried. One study that did was published in 1999 in Great Britain.
Examining about 1,000 children, researchers found that the number of side effects among those who were taking off-label prescriptions was small, but more frequent than for those taking drugs for approved uses.
"If you give a medicine in the right dose, and with good information on how a patient with that illness will handle it, you are less likely to get an adverse drug reaction than if you are prescribing outside of those boundaries," said one of the study's authors, Imti Choonara, a professor in child health at the University of Nottingham. "Otherwise, there's no point to anybody studying medicine. You might as well say, 'Here's a medicine, take it as you like and come tell me if there is a problem.' "
Doctors don't have to go very far off-label before they can put patients in danger. Sometimes, simply prescribing a drug for longer than it's approved for can cause problems.
In 2001, Glenna Baker, a loan officer from Burke, Va., came down with a debilitating stomach disorder that was suspected to be diabetic gastroparesis. She vomited repeatedly, prompting a specialist to prescribe Reglan.
The FDA has approved the drug to be used for less than three months at a time, but studies have found that it's frequently prescribed improperly and that long-term use exposes patients to unnecessary side effects.
One of the worst is tardive dyskinesia, a condition that causes relentless body tremors and facial tics. Baker, now 55, said she was never told about this, so when she moved into her fourth month on the drug she didn't realize what was happening when she began to twitch every now and then.
In her fifth month on Reglan, her symptoms worsened. Her primary care doctor quickly saw the connection.
"It's the Reglan," she said he told her. "We have to get you off it immediately." Her specialist, hearing the news, called Baker at home, angry with her for not alerting him to the oncoming symptoms.
"Don't scream at me," she recalled saying during the June 2001 phone call. "You didn't even warn me. You didn't tell me."
Today, Baker is out of work; the tremors make holding a job impossible. She can sit only for short spells; her right leg constantly bounces, and she endlessly wrings her hands.
She sued the specialist, Dr. Gabriel Herman of Fairfax, Va., and drug maker Wyeth. Both declined comment. Wyeth, of Madison, N.J., has since sold its interest in the drug. In its court filings, Wyeth noted that the drug's label mentioned both the risk of tardive dyskinesia and that the drug is recommended for short-term therapy. In a court filing, Herman has disputed Baker's lawsuit.
The national prescription data that Knight Ridder examined reveal the startling breadth of off-label prescribing. Virtually every drug has been prescribed that way at some point, and many are regularly.
Dr. Nancy Nielsen, an elected official of the American Medical Association, doesn't think doctors have been "cavalier" about it. "They have been in meetings," she said. "They know it works."
But individual doctors and patients aren't in a good position to gauge the safety or effectiveness of off-label treatments, experts say. Even in the busiest of practices, doctors see too few patients to assess the drugs' range of side effects. They also have no way of knowing whether the drugs are working, if it's a placebo effect or whether the patients simply got better on their own.
Medical history is filled with examples of doctors who were convinced that an off-label therapy was safe and effective, only to be proven disastrously wrong.
Often, they based their certainty on secondhand anecdotes, small published studies or observations from their own practices. Off-label prescribing can continue for years before a thorough clinical trial finds it's ineffective or even dangerous. Often such trials are never done.
In the 1990s, there was Fen-Phen, an unapproved cocktail of two prescription appetite suppressants that was widely prescribed until 1997, when the Mayo Clinic noticed that some Fen-Phen patients were suffering from a rare heart-valve disease.
More recently, there was the rampant off-label prescribing of hormone replacements. Though they were approved for treating specific menopause symptoms, such as hot flashes, doctors put millions of women on the drugs for life. They believed hormones would prevent heart disease, breast cancer and Alzheimer's disease, uses the FDA hadn't approved. They even started women on the drugs years after they had gone through menopause.
A massive government-run study, the Women's Health Initiative, found that hormone replacement therapy actually increases a woman's risk of getting these diseases.
Many doctors don't believe the findings, theorizing that the study's outcome would have been different if the women in it had started hormone therapy earlier and had taken it longer.
Despite decades of off-label prescribing, drug makers universally deny that they push these uses at doctors.
"We don't track what you're calling prescribing for unapproved uses," said Doug Petkus, a spokesman for Wyeth, which makes an antidepressant in the Knight Ridder analysis as well as hormone replacement therapy. "We don't recommend that our products be used off-label."
Once it's on the market, a drug might be prescribed for a dozen or more unapproved conditions.
The FDA approved Topamax in 1996 as a supplemental treatment for epilepsy.
It has several potentially serious side effects. It often causes numbness and tingling in the hands and feet. It can cause depression and kidney stones. It can slow the thinking of many patients, impairing memory and making it difficult to choose words. It can cause vision problems, including a form of glaucoma that can result in blindness.
To stop epileptic seizures, the FDA deemed these risks acceptable.
Knight Ridder's analysis of the last year of prescription data for Topamax found that doctors are giving it to patients for migraine headaches, schizophrenia, bipolar disorder, depression, pain, nerve damage and to help them lose weight.
In 1998, Topamax was prescribed only for epileptic seizures, the data show. Now, 79 percent of Topamax prescriptions are for illnesses and conditions that the drug hasn't been approved to treat, Knight Ridder's analysis found. The firm has asked the FDA to approve its use for migraines.
Carolyn Bartley nearly went blind in June 2000, a week after she began taking Topamax, which her psychiatrist prescribed as a treatment for bipolar, a disorder characterized by bouts of depression and mania.
"Everywhere I looked, it was like a watercolor painting, and somebody had smeared it," said the 44-year-old Annapolis, Md., bookkeeper. Laser surgery reduced the pressure in both eyes and restored her sight. In 2001, the maker of Topamax sent a letter to doctors warning about the kind of sudden glaucoma Bartley experienced.
Bartley's psychiatrist, Dr. Parviz Sahandy, said he hadn't researched the medical literature on Topamax, but that he considered it effective for bipolar disorder about 50 percent of the time.
Those who've read the studies are less certain.
"Topamax is no better than a placebo," said Dr. Joseph Goldberg, a research scientist at Zucker Hillside Hospital on Long Island, N.Y., who published a review of epilepsy drugs as treatments for bipolar disorder earlier this year while at Cornell University. "It's giving an ineffective treatment for a potentially life-threatening illness. It would be like giving Tylenol for pneumonia."
Goldberg said that even studies by the drug's maker, Johnson & Johnson, found it didn't work to treat bipolar disorder. As a result, company spokeswoman Lesley Fishman said, the drug maker didn't seek FDA approval for treating that condition.
Over the last year, doctors wrote 586,000 Topamax prescriptions to treat bipolar disorder, Knight Ridder found.
The terbutaline pump therapy prescribed for Tammie Snyder in suburban Detroit last year is one of many off-label drug treatments doctors use in an attempt to stop preterm labor.
Nearly 500,000 babies are born prematurely each year, but the causes and cures for preterm birth largely elude science. Most of the treatments are based on hope and a desperate desire to try something.
Dr. John Thorp Jr. was part of a research team for the federal government's Agency for Healthcare Research and Quality that reviewed the scientific evidence for terbutaline and a host of other drugs in preventing preterm labor.
Their report, published in June, found that the drugs weren't effective in prolonging pregnancy for a long term and can cause a wide range of harms, including heart-rhythm disorders and heart failure.
"There really is no evidence," Thorp said, noting that early contractions stop without any medical intervention 50 to 70 percent of the time.
Nonetheless, women across the country are taking these drugs that doctors know very little about.
"I think experiment is too good a word," said Thorp, a professor of obstetrics at the University of North Carolina at Chapel Hill. "It implies observation, measurement, alteration _ that you're actually conducting science. Clinical crap-shoot would probably be better."
The best use of terbutaline is a series of three injections to calm contractions for about 48 hours or so, enough time to administer steroids to help the babies' underdeveloped lungs, said Dr. Washington C. Hill, the chairman of obstetrics and gynecology at Sarasota Memorial Hospital in Florida and a member of the board of directors of the national Society for Maternal-Fetal Medicine.
But doctors still send women home with long-term prescriptions for terbutaline pills. If they were candid, Hill said, those doctors would admit: "I know it doesn't work, but it cuts down on the phone calls."
Last year, 63 percent of the more than 392,000 prescriptions for terbutaline pills were for pregnant women, despite pharmaceutical company labels that warn against the asthma drug being used this way.
While many doctors and mothers passionately believe in the treatments, especially in the terbutaline pump, the national company that is the largest provider of the therapy is ambivalent.
Dr. Gary Stanziano, the vice president for medical affairs at Matria Healthcare, based in Marietta, Ga., said he had no opinion about whether the therapy his company sold for about $10,000 a month worked. "There are studies out there that are positive and studies out there that are negative," Stanziano said.
Stanziano and Roberta McCaw, Matria general counsel, said their company is in the business of following doctors' orders, since physicians are the ones who write prescriptions for terbutaline pumps. Matria is a middleman, supplying the drug and pumps, then having its nurses hook up the women to the devices and monitor their progress. Stanziano wouldn't say how many women use Matria's pump therapy each year, but said he hadn't heard of any client who had been seriously harmed.
In 1997, however, the FDA warned U.S. doctors that the terbutaline pump "has not been demonstrated to be effective and is potentially dangerous." Four years earlier, the FDA had warned Tokos Medical Corp., one of the two companies that merged to become Matria, about promoting unapproved preterm-labor therapies despite promises to stop.
"This case could be viewed as a conspiracy to circumvent the FDA approval process," an FDA compliance officer wrote in a memo that year.
Matria officials said they had no information about this, because it happened before the merger. They said Matria doesn't promote the off-label use of terbutaline to doctors.
Snyder said she felt betrayed by her obstetrician, Dr. Federico Mariona, a clinical professor at Wayne State University in Detroit and a leader in the local medical societies. Mariona didn't respond to repeated requests for an interview.
Snyder's medical records from Matria include signed consent forms that say, among other things, that some of the treatments being prescribed by her doctor may involve the use of drugs "outside of their labeling." Only after having congestive heart failure, Snyder said, did she learn what those cryptic words meant.
While her heart has improved enough that she doesn't need a transplant, Snyder, 30, said her doctors had told her she couldn't have more children; pregnancy would be too dangerous for her heart.
Snyder and her husband, Chris, had wanted to have more children. Because they used in vitro fertilization to have their twins, they have other frozen embryos.
Now, those embryos remain in limbo.
"I hate terbutaline. I hate what they did," Chris Snyder said.
(Researcher Tish Wells contributed to this report.)