Carla K. Johnson, AP Medical Writer
Women often find it somewhat mysterious to do a breast exam. It also can be mysterious to medical students, says a Chicago researcher who has a $1.8 million federal grant to design the first physical test to measure how well future doctors examine breasts by touch and find possible cancers.
Using plastic models of breasts embedded with data-capturing sensors and simulated tumors, Dr. Carla Pugh of Northwestern University Feinberg School of Medicine is zeroing in on how to measure a medical student's ability to tell the difference between a cancerous lump and a benign cyst.
Medical students' hands-on skills in breast palpation are not tested in their licensing exam. Pugh hopes her research will change that.
She says doctors do not want to talk about it, but every doctor knows a colleague who has missed a cancer during a breast exam, or they have missed one themselves. Pugh, 45, is a surgeon with a doctorate in education and a patent in sensor technology. Her team includes experts in mechanical engineering, computer science, statistics and medical education.
"What if there were a test and you had to meet a minimum level of proficiency before you could do breast exams and say you're competent?" she asks. That might lead to better exams and more accurate diagnoses, she says.
In 2009, the U.S. Preventive Services Task Force recommended against teaching women how to do breast self-exams, finding little evidence they reduce breast cancer death rates. The task force did not give doctors' breast exams a ringing endorsement either, saying the techniques used are not standardized enough but can detect a large proportion of cancers if done well. Many groups, including the American Cancer Society, still recommend that women get regular breast exams from their doctors.
Testing med students' breast exam skills with a sophisticated simulator would be an important advance, says Dr. James Gordon, who directs the Gilbert Program in Medical Simulation at Harvard Medical School. He is not involved in Pugh's research.
Simulators used in medical education have been evolving rapidly over the past decade, Gordon says. The newest models are fitted with sensors that can give students feedback whether or not they are doing an exam properly.
These tools may ultimately improve patient safety by giving medical students repeated learning experiences with a wider variety and number of cases than they might happen to encounter in real patients, although working with real human beings will continue to be at the core of medical education, Gordon says.
Familiar with another Pugh device, a pelvic exam simulator, Gordon says Pugh's work is at the forefront of the field.
"She's at the cutting edge and does some of best work in the world in creating and validating simulators that students and medical professionals could use," Gordon says.
No clowning is tolerated in Pugh's lab. The current prototype, a disembodied plastic breast hooked by wires to a computer, modestly wears a blue cloth covering when "she" is not being examined.
"Because it is a human being," Pugh explains when asked about the covering. "She's sitting there on the table with her breast exposed, so when we're not examining her we cover her. It's a habit that we've formed."
Breast exam skills are not tested in the current U.S. medical licensing exam, says Dr. Stephen Clyman of the National Board of Medical Examiners, a co-sponsor of the test that is the entryway to the medical profession.
Clyman directs the board's Center for Innovation, which investigates how new technology might be used in future licensing exams. He is a consultant on Pugh's breast exam simulator research.
Deciding how to score a breast exam using a simulator in the high-stakes licensing exam would be challenging, Clyman says. "Is it just a matter of finding a lump?" he asks. "Or are you concerned about whether somebody palpates by accepted protocol when you know that many doctors don't follow the accepted protocol?"
Pugh's project will attempt to answer those questions by recording the movements, pressure and patterns of experienced doctors as they palpate simulator breasts with various masses and cysts hidden inside.
The mock tumors must feel authentic to a physician's fingers. Pugh has built tumors from beans glued together, glass beads and hardened clay. Lentils embedded in rubber mimic the feel of fibrocystic breast changes, a noncancerous condition.
More than once she has been at a supermarket, restaurant or clothing store, spotted an object and thought, "Oh, I could use that!"
"Any material I encounter on a daily basis is fair game to help me build a patient," she says.
The next step is testing the mock-up on experienced doctors. A meeting of cancer doctors in Chicago in June is the next test run. Foam rubber alone was not convincing doctors during the last trial that it was a fibroadenoma, another noncancerous condition, so Pugh directs research assistant Jon Salud to make a change.
"This is what I want next, Jon," she tells him. "I want lentils in this."