When Professor of Pediatrics Carole Jenny diagnosed one toddler's brain damage as abuse-related back in 1995, the child had already been to eight visits to other doctors over the previous four months. Yet, somehow, the abuse had been missed until now. Shocked and curious, Jenny and her team from Children's Hospital in Denver pored over medical records on nights and weekends to see how often doctors missed such cases of abuse. Working without funding, they discovered that abuse-related head injuries in babies and toddlers are misdiagnosed almost one-third of the time. The results of their work were published in the February 17 Journal of the American Medical Association in February.

"We're all sort of in collective denial," she says. "Who wants to think somebody would shake their baby?" Jenny and her colleagues studied 173 children with abusive head trauma such as fractures, brain contusions, and hemorrhaging. The team found that doctors at first had missed the abusive head traumas in 31.2 percent of the cases. The correct diagnosis was eventually made, but only after an average of seven days. Twenty-five of the 173 children died. Four of the five deaths, Jenny and her colleagues conclude, could have been prevented if the abuse had been diagnosed earlier.

One reason for the slow diagnoses, Jenny says, is what she calls a "subtle bias." For example, abuse is more often missed in cases involving white children from two-parent homes, even though, she adds, "child abuse is a problem that transcends socioeconomic and cultural boundaries."

There are also disincentives to doctors' recognizing abuse. These include the likelihood of alienating the family and the messy business of having to meet with police and testify in court. "You have to sit in court half the day," Jenny says. "Who's going to see your patients?"

Jenny joined the Brown faculty in 1996 and quickly established ChildSafe, a child-protection program at Hasbro Children's Hospital in Providence. She leads a team of three doctors who study forensic pediatrics and all forms of child abuse. The team, which also includes a Hasbro social worker, meets weekly with representatives of the state attorney general's office and of the Department of Children, Youth, and Families (DCYF) to review the worst abuse cases.

Her work has led Jenny to conclude that child protection requires a three-pronged approach. "You have to have an effective criminal justice system," she says. "You have to have a social service agency that can effectively protect children. And you have to have medical care. If one of these pieces is missing, the stool doesn't stand up." Doctors, she adds, also have to be better trained to recognize abuse. "Here at Brown we do a good job at it," she says. "But most residency programs don't have training. It's expensive. It's new."

Jenny began this grim work when tried to return to medical research after seven years as a full-time mother. At first, she was told that a "housewife" could not become a faculty member; then, in 1984, the University of Washington offered her a child-abuse faculty opening that no one wanted. She accepted the job, hoping to soon switch to another type of research. "I said, 'I can't imagine doing child abuse. It's so depressing.' "

But Jenny was hooked when she realized that "there were so many questions that had to be answered." She moved to the University of Colorado in 1990. "Most of the people my age [fifty-two] in the field have quit," Jenny says. "I can't figure out why I'm not burned out." When pressed, she points to her program at Hasbro. "I've really trained some of the leaders in the field," she explains. "It's just watching them develop and become the next generation of experts."

Jenny hopes that increased awareness of child abuse will lead to more funding to study the problem. "Abused children are not a very good lobbying group," she says. "It's just not a real high priority among federal agencies."