By Tracy Seipel
San Jose Mercury News.
If your doctor can’t determine what’s ailing you, can the collective wisdom of crowds?
What Wikipedia has done for knowledge, a San Francisco company called CrowdMed is betting it can do for medicine.
Send your symptoms and a nominal fee to CrowdMed.com, and dozens of medical professionals, students and average Joes will “crowdsource”, that is, share their knowledge and expertise, to help diagnose what’s wrong with you.
The company isn’t out to replace your family doctor but instead take advantage of the reach of social media to tap into an age-old medical practice: seeking second opinions. Or, in this case, hundreds of them.
“We’re essentially trying to match up the patient with a group of detectives who can help solve their case,” said CrowdMed CEO Jared Heyman, whose company is backed by $2.4 million in venture capital funding. “And the more diverse backgrounds they have, the better our chances.”
The concept, however, is not without its critics. Some physicians say the business model raises a number of serious concerns, from the credibility of those offering advice to the security of sensitive medical information that patients upload onto the website.
Asked about the company, a California Medical Association spokeswoman said their group’s 40,000 members believe that doctors “with the appropriate education, training, access to medical records and a physician-patient relationship are in the best position to diagnose medical conditions.”
Still, some doctors are open to the idea, and others are downright enthusiastic.
“It may be of value for difficult diagnoses,” said Dr. Michael Hogarth, professor of pathology and laboratory medicine at the University of California-Davis Health System. And CrowdMed’s greatest potential, he said, may be its ability to provide a way to “double check” a physician’s work.
At the UC-Berkeley/UC-San Francisco Joint Medical Program, Dr. Amin Azzam, director of the “problem-based learning” curriculum, wants to use CrowdMed “to push the boundaries of how we train medical students.”
Instead of teaching first- and second-year students with “pretend patients,” as is done now, Azzam is proposing adding CrowdMed’s cases to the curriculum. “They might even be more motivated to learn because it’s a real patient, not a fake patient,” he said.
Within 90 days of a consumer putting a case online, CrowdMed’s algorithm generates a list of the most probable diagnoses submitted by its “medical detectives,” along with their explanations.
Patients are asked to give those suggestions to their physicians for consideration and possible treatment. Once it’s confirmed that the suggestions were helpful, the patients are refunded their $50 deposit and the detective who made the correct diagnosis gets his or her reward.
Heyman said 60 percent of the site’s 200 active detectives say they work in or study medicine. Still, he added, some of the best sleuths are not licensed physicians, but rather medical students, alternative medicine providers and patients with firsthand experience with specific medical conditions.
So far, Heyman said, CrowdMed has featured 220 cases on its website. He says it has an 80 percent success rate in coming up with a correct diagnosis.
Patients can post their cases on the site for free, but adding cash rewards motivates more detectives to work cases. Rewards average about $200; CrowdMed takes a 10 percent cut.
Heyman emphasized that the sensitive information patients post anonymously on the site, from medical records to test results, even X-rays, is private, adding that the site is audited for security.
He acknowledges that the year-old company has yet to show a profit, but he believes the process can save consumers time and money spent on multiple doctor visits and testing.
That was certainly true for Monica Seeley, a 46-year-old Ventura, Calif., mother who said she spent five frustrating years visiting doctors with her oldest son, Joseph, now 18, to find out why he has been so fatigued and suffering joint pain.
Over that time, she estimates, the family’s health insurer footed at least $75,000 in medical bills related to his case. But the best diagnosis doctors could come up with was chronic fatigue syndrome.
In January, after hearing about CrowdMed, Seeley posted Joseph’s case online. Early posts suggested Lyme disease, something her son had already been tested for, with negative results. Based on further online chats she had with some medical detectives, including one who identified himself as “AnonymousDoctorTeach,” she had him retested.
A second, more thorough test revealed that Joseph carried Lyme disease markers. And a new doctor prescribed antibiotics that seem to be helping him.
“Crowdsourcing” health problems isn’t new.
In its “Think Like a Doctor” column, The New York Times blog offers readers an opportunity to “solve a medical mystery” with Dr. Lisa Sanders, who was also a medical adviser to the popular TV show “House, M.D.”
For all its promise, CrowdMed’s medical adviser, Dr. Greg Denari, a retired Kaiser Permanente family doctor and adjunct clinical associate professor at Stanford University School of Medicine, acknowledges there is room for improvement.
Still, he said, getting more eyeballs to review medical cases remains the company’s greatest strength.
“The expectation of doctors in general is that they’re supposed to be right 100 percent of the time,” Denari said. “That’s an impossible expectation.”