Web startup invites everyone to solve medical mysteries

The 54-year-old female patient was in good health, “very athletic and fit,” until September 2013, at which time she participated in a tennis match in 100-degree-plus heat. The patient experienced a tachycardia episode, with a heart rate of 185 bpm, and was taken by ambulance to the hospital. At the hospital, it was determined that her potassium level was 2.1 mEq/L, down from 4.5 mEq/L when measured during a physical exam 2 weeks earlier.

Since the initial episode, the patient continued to have ongoing episodes of tachycardia, controlled with a beta-blocker, as well as random BP spikes as high as 170 mm Hg/120 mm Hg. Through the use of a calcium-channel blocker, the spikes now go no higher than 150 mm Hg/105 mm Hg.

Another major symptom for this patient is extreme fatigue. She notes that she now requires a great deal of sleep and has chest pains with physical exertion. The patient reports having “premature heart beats” a few times a day. Several physicians and countless imaging tests later, she still has no diagnosis.

“The puzzle is how I can be perfectly healthy and, after one night of dehydration, end up completely exhausted with a range of what appear to be cardiac symptoms with no blockages,” the patient wrote.

What do you think this patient suffers from?

How much are you willing to bet on it?

At the Web startup CrowdMed, you actually can place a bet on this case, as well as various others, and win a cash or points award as a “Medical Detective.”

Not a doctor? Not a problem.

“No one would tell you that you can’t play sports if you’re not a professional athlete,” the CrowdMed website states. “While we love to have Medical Detectives with a medical background, we welcome everyone to join CrowdMed. We value intellectual diversity and think that even non-experts have useful wisdom and personal experience to share.”

‘Wisdom of crowds’

According to its website, CrowdMed was founded by tech veteran Jared Heyman, whose teenaged sister had a debilitating medical condition that eluded a diagnosis. After a unique chance to consult with an interdisciplinary team at the NIH, Heyman’s sister was diagnosed with fragile X-associated primary ovarian insufficiency, a rare genetic mutation. After undergoing a simple treatment with a hormone patch, her symptoms were resolved within a month.

Although most patients with puzzling cases do not get the benefit of a diagnosis straight from the NIH, Heyman decided they should be able to benefit from a collaborative diagnostic process. The concept of CrowdMed was based on a theory of the “wisdom of crowds,” which states that if properly harnessed, a large collective of non-experts can demonstrate intelligence superior to that of a single expert.

The concept also builds upon crowdsourcing, a sort of outsourcing that utilizes an online community to provide content or services.

At CrowdMed, a case is submitted by an anonymous patient, and Medical Detectives place bets as to the most likely diagnosis. Their bets consist of either cash or points, which can be accumulated or donated to Watsi, a nonprofit that funds medical treatments for patients around the world.

A case remains on the site for 90 days, after which time the patient and his or her physician decide on the most likely diagnosis. The winnings are divided between all Medical Detectives who bet on that diagnosis.

Currently, 57% of CrowdMed’s 192 active Medical Detectives work in the field of medicine in some capacity, the average age is 34 years and 51% are male.

A time-honored tradition

According to John G. Byrne, MD, chief of the division of cardiac surgery at Brigham and Women's Hospital, the concept of a collaborative process is not without precedent in medicine.

“The notion of getting a lot of people involved is a tried-and-true tradition that dates back decades,” Byrne told Cardiology Today. “It’s been done widely as part of our professional culture. The idea of getting 30 or 40 people in a room to help solve a problem, as opposed to just one person, is not new. We’ve been doing this for decades, with good results.”

William T. Abraham

William T. Abraham, MD, director of the division of cardiovascular medicine and professor of internal medicine, physiology and cell biology at the Ohio State University Wexner Medical Center (OSUWMC), said correct diagnoses do not always correspond with medical credentials or seniority.

“I am a strong believer in the old adage that two heads are better than one, and I think you can take that concept and extend it to multiple heads,” Abraham said. “When I see patients at the clinic or on rounds at the hospital, it’s often along with a cardiology fellow, an internal medicine resident, a medical student, a nurse — a whole multidisciplinary team. And oftentimes, it is perhaps the most junior member of that team who offers the most important insight.”

PAGE BREAK

Occasionally, Abraham said, a patient or the patient’s family may offer information that is essential to arriving at a diagnosis.

“Sometimes, a family member might say, ‘You know, doc, a neighbor of ours had something like this,’ and this whole conversation leads us to an answer,” he said.

A little bit of knowledge

Although the concept of harnessing and aggregating multiple diagnostic possibilities may be intriguing, Abraham said he recognized its potential drawbacks.

“Unfortunately, sometimes a little bit of knowledge can be a dangerous thing,” he said. “When so much information is presented in a completely unfiltered way, it has the potential to lead to misdiagnosis, and it has the potential to cause anxiety where anxiety should not exist.”

Byrne agreed, although he said that the physician oversight provides a bit of a safety net.

“I suppose as long as the patient’s personal physician takes ultimate responsibility for the diagnostic treatment, it’s probably not harmful from a medical point of view,” Byrne said. “However, a few downsides are definitely possible, such as time spent chasing down diagnostic rabbit holes that may ultimately be fruitless. There could be a lot of patient misinformation and anxiety. And then the doctors might feel obligated to order every imaging test that is suggested, even if it doesn’t have medical merit.”

Also problematic, according to both Byrne and Abraham, is the idea of multiple unfiltered opinions. Byrne likened it to discussing a medical issue with a group of friends at a backyard barbecue.

“You might tell a group of friends, ‘I have this lump,’ and there will be one person who’ll say, ‘Maybe it’s a just bone spur,’ but there will also be that person who’ll say, ‘Maybe it’s a sarcoma.’”

This sort of uncontrolled feedback can be upsetting to the person with the health problem, Byrne said, particularly since there is a good chance that everyone is wrong.

“Sometimes, I get in a room full of Harvard professors, and we can’t even figure out a case,” he said.

Additionally, stirring up patient fears through unfiltered opinion sharing can cause real problems, Abraham said.

“One thing I’ve seen is that as patients become more informed, they become aware of not only the more common side effects of a drug, but they go online and see every potential reported side effect, every one-in-a-million effect,” he said. “Or they go on a discussion board where other patients talk about side effects, and sometimes they get scared and stop taking their medication. These are often proven medications that improve medical conditions, but patients get anxious and stop taking them, often to their detriment.”

The ‘Wild West’

In many ways, the Internet seems to have ushered in the age of the amateur — or, more frighteningly, has rendered everyone an expert.

Faisal G. Bakaeen

“You would want a certified pilot to fly your plane, instead of an expert who is into flying,” said Faisal G. Bakaeen, MD, professor of cardiothoracic surgery at Baylor College of Medicine. “Just because our current technology allows us to do it, doesn’t mean it’s the right or best way.”

Abraham said although it seems as though Internet crowdsourcing and opinion sharing is the “new reality,” the public should remember the difference between medical opinion and Internet forum opinion.

“One of the big differences between the Internet and the type of information that medical practitioners base their decisions on is that medical studies and publications are very rigorously vetted for a process of data verification and peer review, so that by the time a paper appears in The New England Journal of Medicine, it has been critically reviewed by a number of specialists in the field,” he said. “Whereas the Internet is sort of the Wild West, the new frontier. Anyone can write anything they want.”

Byrne said the concept of Medical Detectives reminds him of a time in the past when CAT scans were available at shopping malls — an idea that fomented a great deal of anxiety for patients.

“It caused a lot of confusion and fear,” Byrne said. “What’s a signal and what’s just noise? What’s the difference? That’s where the judgment should come in.”

Overall, Abraham said, he thinks multiple opinions can never be harmful. However, they should be presented responsibly.

“Completely unfiltered and uninterrupted information can lead to poor decision-making,” he said. “But I think patients should be well-informed and involved. The era of paternalistic medicine, where the doctor was right and the patient should just listen, is well behind us. And it’s gladly well behind us.” – by Jen Byrne