Diagnostic crowdsourcing for rare diseases

The Economist has an excellent article on a new movement to crowdsource (i.e. obtain information from a crowd of people over the Internet) specialized information about rare diseases that may be difficult to obtain from traditional sources (e.g. Wikipedia or WebMD). The key resource is a centralized website (CrowdMed.com, Figure 1) that enables interactions between patients with mysterious symptoms and professional medical experts, as well as amateur medical detectives and other patients:

"But to create a lone website in the hope that a knowledgeable passer-by may shed light on a mysterious illness is the cyber-equivalent of crying in the wilderness. To create one on which anyone and everyone can post a problem, so that anybody who might, for whatever reason, want to help knows where to go, is more like setting up a stall in a market at which buyers and sellers know where to meet. And such a stall, called CrowdMed, now exists."

Rare diseases are often overlooked as a major medical issue. Most people focus on the common disorders like heart disease and cancer that make up the majority of causes of death. By definition, each rare disease afflicts a relatively small number of people (< 200,000). However when taken in aggregate, about 8% of Americans are affected by some rare disease. The following characteristics are associated with rare diseases (MedlinePlus):

• May involve chronic illness, disability, and often premature death
• Often have no treatment or not very effective treatment
• Are frequently not diagnosed correctly
• Are often very complex
• Are often caused by changes in genes

As Point #3 mentions, rare diseases are often misdiagnosed. Because there are so many different ones, a doctor may not be familiar with a specific rare disease. Each affects a small number of patients.  Because they are rare, most are not well-studied by the medical community which tends to focus on the "major" diseases. The need therefore exists for acquiring and disseminating information on these disorders (Economist):

"The need for a “crowdsourced” service like this comes from the number of rare diseases around. The National Institutes of Health, America’s medical agency, recognises 7,000—defined as those that each affect fewer than 200,000 people. A general practitioner cannot possibly recognise all of these. Moreover, it may not be clear to him, even when he knows he cannot help, what sort of specialist the patient should be referred to. Research published in 2013, in the Journal of Rare Disorders, says about 8% of Americans—some 25m people—are affected by rare diseases, and that it takes an average of 7½ years to get a diagnosis. Even in Britain, with all the resources of the country’s National Health Service at a GP’s disposal, rare-disease diagnosis takes an average of 5½ years. Also, doctors often get it wrong. A survey of eight rare diseases in Europe found that around 40% of patients received an erroneous diagnosis at first. This is something that can lead to life-threatening complications."

How does CrowdMed work?

"CrowdMed, though, brings numerous pairs of eyeballs, each with different knowledge behind them, to every problem. Patients submit their cases and may offer a reward of a few hundred dollars to lubricate the process. The volunteer diagnosticians are students, retired doctors, nurses and even laymen and women who enjoy pitting their wits against a good medical mystery. Besides the cash, successful volunteers also get the kudos of rising in the website’s ranking system—and that ranking system is, in turn, used to filter the feedback given to patients, to try to avoid mistakes."

The article concludes by mentioning that diagnostic crowdsourcing will face competition in the future from artificial intelligence (AI) diagnostic systems such as IBM Watson:

"Diagnostic crowdsourcing will not have the field to itself for long, however. It faces competition from machine-based systems such as Watson, a computer built by IBM that digests large bodies of data and draws inferences from them. It will be interesting to see whether the collective wisdom of practitioners and enthusiastic amateurs prevails over an algorithmic synthesis of the world’s medical literature. Whichever emerges victorious, though, it will be better than just entering symptoms into a search engine—and it will not just be cyber-chondriacs who benefit."

I don't believe that the two approaches are mutually exclusive; they can co-exist side-by-side. One would expect Watson to perform better on those rare diseases in which there are more data available. For the more obscure disorders, the "wisdom of the crowds" aggregated in a website like CrowdMed could prove to be a valuable resource.

Figure 1. Homepage for CrowdMed (crowdmed.com).