By Megan Molteni

In Gerald Grant’s line of work, there isn’t such a thing as an “average” patient. As a chief of pediatric neurosurgery at Stanford University Medical Center, the children that come into his operating room are unique, each requiring a complex surgical procedure tailored to the architecture of a young brain.

But that doesn’t mean he can’t learn from what other people have done. Grant’s always searching for similar cases to give his patients the best possible shot. And more and more he’s finding those answers not in prestigious, paywalled publications like the Journal of Neurosurgery (of whose editorial board he is a member), but on the freely available pages of an upstart publishing platform modeled after, of all things, Turbotax.

The Cureus Journal of Medical Science (that’s pronounced “curious”) is the brainchild of one of Grant’s colleagues at Stanford, a fellow neurosurgeon named John Adler. He’s on a mission to build the world’s most comprehensive library of medical case studies. Cureus is the first and only peer-reviewed publication with step-by-step article templates for authors—which dramatically speeds up publication times. (Just like your tax software!) If case studies are published in weeks instead of months, that means millions more medical lessons to learn from.

The science of medicine is all about power in numbers. Big cohorts, long-term trials, and lots of money ensure that most treatments will work for most people with a disease. But the practice of medicine is all about individuals. And case reports—detailed accounts of an individual patient’s symptoms, diagnosis, and treatment response—are, by definition, outliers. So there’s a healthy (and long-standing) debate about where in the hierarchy of evidence they should sit.

Practicing physicians like Grant and Adler tend to argue for the educational value of case reports. What seems like a one-off might actually fall into a pattern—but how will anyone know if no one writes it down? “Most case reports are undocumented beyond just two surgeons talking over a scrub sink,” says Adler. “Not enough of those stories get told.” Biomedical researchers and subscription-based journal editors tend to spurn case reports, in no small part because of how rarely they get cited. Paywalled journals, like dead tree newspapers, only have so much space to print articles. And they want to make every column count.

But the digitization of peer-reviewed publishing is changing that. Since 2011, the number of journals that focus on case reports has tripled. “Thirty years ago we didn’t have ways to do it,” says Adler. “But now we have control over the floodgates.” Most case report journals are open access—meaning the articles aren’t behind a paywall. Instead, the authors pay a publishing fee, usually a few thousand dollars, to cover editors salaries and other overhead expenses.

This model is easily and often exploited by “predatory” publishers, companies that solicit authors directly, collect their fees, and then don’t follow through with promises of proper peer-review and article indexing. According to Katherine Akers, a biomedical research specialist at Wayne State University and editor in chief of the Journal of the Medical Library Association, approximately half of the publishers of medical case report journals engage in predatory practices. That’s why she tends to view any new publications, including Cureus, with a healthy dose of skepticism. “For the most part, this one actually looks okay,” she says, noting that Cureus is totally fee-free and indexed in PubMed—the database most used by biomedical researchers to find interesting papers relevant to their interests.

But there is one red flag.

For most reputable biomedical journals, the process of reviewing an article takes about three hours. Cureus boasts that with its easy-to-use form, reviews take no more than an hour to complete. “That’s really fast,” says Akers. “Usually that’s a warning sign that these articles aren’t being looked at that rigorously.” Cureus says its review process only verifies the basic scientific credibility of a report. And Grant agreed that the bar to peer review might indeed be a little bit lower on Cureus than elsewhere. But when it comes to case reports, he says that might actually be OK.

“In our world, we’re missing a lot of the science because so many journals don’t see case reports as publishable,” says Grant. “But these rare one-offs could become really interesting if they were all reported, instead of just passed around by word of mouth. I don’t think it dilutes the literature at all.”

Adler, with his goal of publishing tens of millions of articles a year on Cureus, is obviously ambivalent on the dilution question. But that’s because his team built another tool for telling the good from the bad. Once a paper has been published, any of the platform’s 10,000-some users can leave comments and rate the paper’s quality and clinical significance on a scale of one to ten. The idea, Adler says, is not unlike estimating the number of marbles in a jar. If you ask a few people to guess, you get wildly different numbers. But ask enough people you eventually wind up with an average that’s close to the real answer.

With enough data, Cureus could be more than just a publishing platform: It could become a prediction engine. Because case reports are mostly about rare, isolated events, it can take years, even decades to find patterns. Adler envisions his crowd-sourced metric as a way to arrive at that answer sooner.

But to do that, he needs a lot more articles, and a lot more data. Since launching in December of 2012, Cureus has published about 1,600 articles, and it currently publishes about 25 per week. At that rate, it will take more than 100 years to hit 1 million reports. And it’s hard to get doctors to spend as much time rating reports as they do liking Facebook posts and Twitter threads. Right now less than 60 percent of Cureus articles have been rated more than once.

On the other hand, articles don’t need upvotes to be useful in operating rooms. A few months ago, a family brought their 13-year-old daughter in to see Grant. She was suffering from cerebral palsy; Her muscles were in a near constant state of contraction because of a missed connection between sensory nerves in the spinal cord and the brain. Grant realized the best course of action was probably a procedure called a selective dorsal rhizotomy, which would require him to separate out the nerves that worked from the ones that were misfiring, and then cut out only the dysfunctional ones. But because it’s so invasive (and expensive), he wanted to make sure the effects would last long after the operation.

So he logged on to Cureus. Doctors at the Washington University School of Medicine in St. Louis, he found, had performed the same procedure on 94 patients between the years 1989 and 1999. And they followed up with them 20 years later. Nearly 90 percent of the patients said they’d recommend the procedure; they could move better, they had less pain, and the effects were indeed long lasting. “That kind of data is hard to get into a high impact journal,” says Grant. “But in the clinic, this is the kind of stuff my patients are asking me. These are practical questions. And these reports are able to address them.”

Grant performed the surgery. And so far, his patient is doing very well.