Pop quiz for parents:

Researchers want your 9-year-old child to give up a day of summer vacation to participate in a medical experiment. They need to find out whether a new drug, tested only in adults, is safe for children.

The drug will not benefit your already healthy child, but the researchers assure you that, in accordance with federal regulations, the risks will be no greater than the ones kids encounter in daily life. What do you say?

By that measure alone, parents may be wise to say "No," according to one of the first studies to quantify the risks kids routinely face.

As it turns out, the day-to-day risks of being a kid are considerably higher than many people appreciate, these experts assert. And therein, they conclude, lies a problem with federal rules on research involving children. By assuring that risks will be no more than "routine," they convey a false ring of security.

"Everyone's got an intuition about the risk of everyday life," said Ezekiel Emanuel, chairman of the department of clinical bioethics at the National Institutes of Health. "We thought, 'Wouldn't it be great if we could quantify it?' And once you begin to do that, you realize that everyday life is not benign."

The new analysis, led by Emanuel and NIH co-worker David Wendler, is the latest contribution to a bubbling controversy on whether federal research protections for children are adequate. That debate has grown in recent years as it has become clear that the vast majority of medications prescribed for kids have never been tested for safety or efficacy in youngsters -- a risky situation in itself, since children can respond to drugs very differently than adults.

The federal government is making a major push to have more medications tested in children, offering pharmaceutical companies valuable patent extensions, for example, on drugs that the companies test in kids. But that push has led a number of experts to scrutinize the current standards for such tests.

Of special concern are studies involving healthy children, since they offer no potential benefits to offset the risks. Federal rules allow such experiments with parental permission only if a review board deems the risks "minimal" or, in some cases, a "minor increase over minimal."

"Minimal risk," according to regulations, means that the anticipated harm or discomfort will be no greater than that "ordinarily encountered in daily life or during the performance of routine physical or psychological . . . tests."

That standard has proved squishy. Kids living in crime-ridden neighborhoods face terrible risks every day. If the regulations are interpreted by that standard, those children would be wonderful prospects for researchers wanting to conduct dangerous studies. After all, those kids regularly risk getting shot or stabbed.

To avoid that kind of exploitation, the regulations have been interpreted by the Institute of Medicine and others to mean "risks ordinarily encountered by average, healthy, normal children." But here a second problem arises: Just how big are those risks?

To answer that question, Emanuel and co-workers searched high and low for statistics on the risks of daily living. They got some from the federal Centers for Disease Control and Prevention. Others came from sports organizations. Many others came from individual researchers who had quietly spent their lives compiling statistics about obscure activities that they thought nobody would ever care about.

To get a sense of the risk inherent in a standard medical test, for example -- a standard that the regulations say should be used for comparison purposes -- the team tried to track the fates of people who had taken a glucose tolerance test, commonly given to suspected diabetics. After many dead ends, they found a doctor who had tracked injuries from 14,000 such tests but had never published his work.

Similarly, while looking for data on psychological stress in kids, they came across a study comparing the stress of being young and fat to that of being young and diagnosed with cancer.

"We said, 'Wow, he must have some data,' " Emanuel said. Sure enough, the researcher had stress-related information on 6,000 kids.

The product of all this work, published in last week's Journal of the American Medical Association, provides a comprehensive look at the risks in a typical kid's day that might include a ride in a car (the major risk of death for children), time on a playground or in a sporting event (the major cause of injury), and other routine activities such as bathing or swimming.

Their conclusion: In the course of an average day, kids routinely face risks as high as 1 in 250 of an injury requiring hospitalization or a visit to the emergency room. For young people 15 to 19, the cumulative risk of dying in an accident is as high as one in 100,000 each day.

Those risks may not sound huge, but studies have shown that research review boards routinely reject experiments whose risks are substantially lower than those. That disparity reveals a disconnect between experts' intuitive sense of acceptable risk and the actual risks that regulations allow.

Experts agree that the continued participation of children in clinical trials is very important. Without them, such lifesavers as the polio and measles vaccines might never have been approved, and doctors in the 1950s might never have learned that their use of supplemental oxygen was causing blindness in babies born prematurely.

In their article, Emanuel and colleagues propose a few alternatives -- none of which, they emphasized, reflect the views of the NIH.

One of them, the "de minimus" standard, would require that research risks not exceed the negligible risks people tolerate every day (such as when walking to work), rather than using a standard that includes such dangerous extracurricular activities as playing football.

An alternative is to use the model of charitable giving, to achieve a standard of allowable risk that falls between de minimus and the current standard. Just as people accept certain unnecessary risks because of the benefits they bring (such as the convenience of a car or the fun of sports), people also take risks to help others (such as helping build homes for the homeless). If people more consciously appreciated the societal benefits of medical volunteerism, some argue, a higher level of risk might be justified.

The terrain is an ethical minefield, experts warn. And most agree that changes in the current system must await the results of further research.

"No one's presented an alternative that a good philosophy professor couldn't tear to shreds in about 30 minutes," said Terrence F. Ackerman, chairman of ethics at the University of Tennessee College of Medicine in Memphis.

But after decades of inattention, Wendler said, at least the topic is beginning to be addressed.

"We want safe and effective medicine for kids, but we don't want them exposed to risk, so we've closed our eyes," Wendler said. "This begins a discussion of what the appropriate levels of risk are."