May 23, 2002

A Prescription for Disaster

First appeared in on May 23, 2002

Ten-year-old Shaina Dunkle had been taking the psychiatric drug desipramine (trade name, Norpramin) for her attention-deficit/hyperactivity disorder (ADHD) last year, when she suddenly fell and had a grand mal seizure. She died within minutes in the arms of her mother, who watched helplessly as her daughter's life ended.

Shaina's autopsy revealed no other cause for her death than the desipramine. Her parents, in the small central Pennsylvania town of Smithtown, struggled privately with tremendous guilt and anger until they began to learn more about their daughter's treatment. Now, the outrage they feel about the circumstances of Shaina's death has prompted them to go public with their concerns.

Shaina Dunkle should never have been on this drug. In the late 1980s, child psychiatrists promoted desipramine for the treatment of ADHD. But by the mid-1990s, reports of seven sudden deaths of children taking the medication in appropriate doses for ADHD led most doctors to abandon desipramine for other, safer medications. According to Shaina's parents, desipramine was offered to them as a once-a-day drug that, unlike Ritalin, the most commonly prescribed medication for ADHD, had no abuse potential. However, when used properly Ritalin has not been implicated in sudden death (the one reported exception was a 14-year-old boy who died suddenly after taking the drug for more than 10 years). The Dunkles were never told of the increased risk for this catastrophic side effect related to desipramine.

Desipramine is just one of many drugs that received FDA approval for treatment of an adult condition -- in this case depression -- but was then, and apparently is still now, used in children for a variety of behavioral and performance problems. Once the FDA approves a drug, it can be used with a physician's prescription for any purpose, a practice called off-label prescribing. Many doctors prescribe many medications off label for children, but none do it as frequently as child psychiatrists.

Child psychiatrists have long been viewed as the ultimate authorities in the evaluation and treatment of children's emotional and behavioral problems. Today, however, these doctors appear to be pushing pills exclusive of anything else. In fact, a recent survey of child psychiatry practices by the Yale Child Study Center, published in the Journal of the American Academy of Child and Adolescent Psychiatry, revealed that only one in 10 children who visit a child psychiatrist's office leaves without a psychiatric drug prescription.

Even leaders within the community of child psychiatrists, doctors like Michael Jellinek, the head of Harvard's child psychiatry department, and Peter Jensen, former director of the head of the children's section of the National Institute of Mental Health, have publicly expressed concern that there is not enough evidence of the effectiveness and safety of these medications to support their widespread use for kids.

And they should be concerned. About 4 million children take either Ritalin, the only psychiatric drug adequately researched for use by children, or an equally well-tested equivalent like Adderall, Concerta, or Metadate. About 1.5 million children take adult antidepressants like Prozac for a variety of problems. Hundreds of thousands are on various other adult medications: clonidine, a drug originally approved for adult hypertension, is prescribed to take the edge off difficult behavior caused by ADHD, especially in the late afternoon or evening, when taking a stimulant will keep the kid up all night. Depakote, a drug tested in adults for epilepsy, is the latest prescriptive drug in fashion for treatment of the burgeoning number of children diagnosed with bipolar disorder. Risperdal, a new generation anti-psychotic drug, is prescribed for children primarily to control aggressive behavior.

None of these drugs, except Ritalin and others in its stimulant class, have been tested in more than a few dozen children for more than about two or three months. Even Ritalin, which has been around for decades, has been studied no longer than 18 months for its effectiveness or safety. Until recently, the pharmaceutical industry has had little incentive to test drugs in kids -- the market wasn't large enough and pediatric drug studies pose additional ethical hurdles.
But drug companies became very interested in drug trials involving children about two years ago, when the government agreed to extend exclusive patents on drugs for six months if the companies facilitated testing in children. About two years after the extension, however, the FDA, still not satisfied with the amount of pediatric testing being done, introduced an administrative mandate that required drug companies to test previously approved drugs frequently used in children. The move brought three thinly veiled citizen suits funded by the drug industry that challenged the FDA's authority. When the FDA backed down under this pressure, child advocacy groups -- including the American Academy of Pediatrics and the American Academy of Child and Adolescent Psychiatrists -- expressed outrage. The FDA finally reinstituted the mandate, with the apparent public approval of the Bush administration.

Now Sens. Hillary Clinton, (D-N.Y.), Christopher Dodd (D-Conn.), and Mike DeWine (R-Ohio) are sponsoring legislation to make the testing mandatory for drugs that are regularly prescribed for children. But this proposal doesn't go far enough. At the moment, doctors and their patients learn only in a haphazard fashion about the side effects that develop with a drug's long-term use. There are no requirements for what is called post-marketing surveillance -- for any drug -- despite repeated calls for such a procedure from medical and patient advocacy groups.
Drug companies aggressively oppose this kind of thorough follow-up on drugs, not only because it is expensive, but because they don't really want to find out whether their drugs continue to work over time or if long-term side effects develop. Currently, that kind of research is a job for the country's trial lawyers. But this de facto system of monitoring the effects of drugs requires many casualties before an adverse outcome is discovered or established in the medical and popular literature.

At the heart of this debate is the recent surge in the use of psychiatric drugs to treat children. So many children are taking these medications that the need for more testing and follow-up is suddenly crucial. Unfortunately, the alternative -- effective non-drug interventions for the treatment of children's behavioral issues -- has been overlooked.

Specific psychotherapeutic approaches to the treatment of behavioral issues in children -- ADHD, anxiety and obsessive compulsive disorder -- have been shown to eliminate or decrease the need for drugs. But the pharmaceutical industry, with its control of psychiatric research funding, influence in the media, and direct advertising to patients, has no interest in the promotion of these approaches. The problem? Psychosocial treatments like family therapy and special education do not generate stock dividends or equity, and they are less available than drugs.

If other interventions were more widely available and aggressively promoted, it is possible that Shaina Dunkle's apparently mild ADHD symptoms, diagnosed by her school psychologist, could have been managed without medication. Yes, some children who do receive non-drug help can still benefit from taking a psychiatric medication. But Shaina's parents say they never would have given desipramine to their daughter had they known more about it. They trusted the doctor; they believed the drug to be safe.

The extensive prescription of these medications for children, without adequate testing for safety and effectiveness in children constitutes a hidden time bomb that could explode with still more casualties. Catastrophic side effects, like the one that killed Shaina Dunkle, may be rare, but they become predictable when we treat so many children with so many drugs. We need to know more, much more. And until we do, it is imperative that we give adequate consideration to treatments that do not involve drugs whose full effects we don't completely understand.


2099 Mt. Diablo Blvd., Suite 208
Walnut Creek, CA 94596