The paradox of crowdsourced medicine

Defined as “the process of seeking a problem's solution from a wide community, often online,” crowdsourcing is common in almost every sector (Sanghavi 1). However, many patients may be unaware that they can also crowdsource their healthcare decisions. Referred to as “a second opinion writ large,” crowdsourcing medical diagnoses is now possible through many different online platforms including CrowdMed and the more artificial intelligence (AI)-driven HumanDx (Arnold 1). The way medical crowdsourcing works is a little more complicated than asking for fine dining tips in Tokyo or even asking the general public for clues to solving a crime. With crowdsourced medicine using the CrowdMed model, doctors, nurses, and other healthcare workers essentially compete for whoever offers the most accurate diagnosis, and receive financial compensation for accurate hits. Compensation is higher for difficult to diagnose problems. The HumanDx platform is different, available only to physicians at the moment and uses AI instead of human input. Regardless of the model being used, crowdsourcing medicine presents a host of ethical and legal problems. When these legal and ethical kinks are ironed out of the system, crowdsourced medicine should become fully integrated into the global healthcare system. Crowdsourcing medicine prevents some of the problems that currently plague the profit-driven and paternalistic healthcare system, allowing all patients to receive an evidence-based, culturally sensitive, intelligent assessment of their needs.

Crowdsourcing is controversial because it has the potential to radically transform the relationship between patients and healthcare providers. For example, crowdsourcing potentially threatens the position of authority many doctors depend on to maintain their professional status. As Sanghavi points out, doctors for millennia have “jealously guarded their secrets,” at least in the European model of medicine (1). Yet all medical diagnoses are to a degree already being crowdsourced, as teams of healthcare professionals often collaborate (Arnold 1). Collaboration among medical professionals may be more common in some countries or healthcare settings than in others, making the adaptation to a crowdsourcing model of diagnosis easier for some than others to swallow. The current medical model has veered towards paternalism, in spite of ethical principles guarding patient autonomy. Crowdsourcing empowers patients, allowing them to receive accurate medical diagnoses anonymously and from a team of medical professionals. Especially in light of the already impersonal nature of the modern medical model, crowdsourcing does not supplant the role that doctors and nurses play. In fact, many patients “feel as if their doctors aren’t actually listening to them,” which is why they turn to crowdsourcing in the first place (Sruthi, on “Blind Spot,” 1).

Whereas using the Internet for self-diagnosis can be problematic, leading to misdiagnosis and hypochondria, crowdsourcing medicine actually has the potential to reduce medical error overall. Crowdsourcing can divert patients away from unreliable sources of information towards a well-informed community. Although research substantiates the potential for crowdsourcing to improve the accuracy of medical diagnoses, credibility is one of the main reasons doctors and patients fear crowdsourcing medicine. One of the reasons why crowdsourcing medicine has become controversial is that with CrowdMed, a medical license or even a medical background is not required to be one of the “medical detectives” offering diagnoses (Arnold 1). The lack of background checking on CrowdMed makes the system seem unlikely to work well. However, the reasoning behind the support for amateur medical sleuths is surprisingly simple: “CrowdMed is a performance-based system,” not a system based on someone’s title or position of authority (Arnold 1). A title and a license do not necessarily guarantee an accurate diagnosis of a problem. After all, many retired healthcare workers can be sleuths but not legally practice medicine. Likewise, many former patients with direct experience with rare diseases can offer accurate diagnoses their doctors might have missed. The CrowdMed system awards accuracy of the participants’ diagnoses, not the caliber of medical school one attended.

Moreover, crowdsourcing has built-in safeguards to protect patients from misinformation and ensure that patients use the information they receive wisely, legally, and ethically. The technology is always advancing, and CrowdMed is only one company that specializes in crowdsourcing medicine. Other companies might follow suit to provide a more competitive market that enhances the quality of diagnoses. Even now, the CrowdMed system guards against amateur diagnoses by using a ranking system that weights licensed physicians higher as medical detectives than their inexperienced counterparts. The only way an inexperienced person would rank higher than a physician was if that person ended up being right more often than the doctor. The system is therefore meritocratic. A more frequently accurate amateur could potentially come to outrank a less accurate physician, which is why crowdsourcing may work especially well for patients with difficult-to-diagnose problems, problems their doctors ignore or overlook, or problems the doctor is unwilling to defer to a colleague. As another safeguard in the system, the CrowdMed model also requires that all patients’ cases are moderated by a licensed physician. With a licensed physician at the helm of each case discussion, pseudoscience does not dominate the discourse (Couch 1). Complementary and herbal medicine might be mentioned, but only in conjunction with evidence-based practice. Finally, Sruthi points out, “for the most part, the best answers rise to the top,” (“Blind Spot” 1). Some doctors might be wrong, some nurses might be wrong, but crowdsourcing ultimately ensures that the more people participate, the greater the likelihood of a successful diagnosis.

Doctors are not infallible; having a crowdsourced opinion is helpful in promoting positive patient outcomes. As Sanghavi notes, one doctor can make a wrong judgment due to any number of factors including lack of familiarity with similar cases or symptoms. Crowdsourcing also prevents cultural biases in diagnoses and treatment options. Some doctors and nurses might not know how to communicate with patients from different backgrounds due to language or cultural barriers; crowdsourcing medicine has the potential to offer patients unbiased assessments. Issues like gender and race can be taken into account for diagnostic reasons, but might not come in the way of doctor-patient communications.

Research shows that the accuracy of a diagnosis dramatically increases the more people assess the case: “if you look at the aggregate choices from tens of thousands of other doctors around the world, the plurality invariably hits the mark,” (Sanghavi 1). Arnold agrees: “a large group of people tends to be smarter and more accurate than any single expert,” (1). In fact, doctors have been slyly crowdsourcing within their own community, such as by maintaining medical blogs that ask for peer input (Sanghavi 1). Professional organizations and the government have effectively crowdsourced to create public health and public service publications that outline diagnoses and treatment for some diseases (Sanghavi 1). In fact, crowdsourcing may also be more effective at treatment interventions. Recent research has shown “groups of doctors outperform individuals not only in diagnosing problems but also in treating them,” (Sanghavi 1). Plurality breeds accuracy.

Crowdsourcing medicine is the first step, after which the patient has the ability to take action armed with the knowledge received online. One of the benefits of crowdsourcing medicine is that it can lead a patient towards appropriate treatments. Patients with problems that remain undiagnosed are especially helped by crowdsourcing medicine. Jared Heyman, the founder of CrowdMed, notes how “ill-equipped our medical system is when it comes to solving tough cases,” (Arnold 1). Crowdsourcing will not be necessary or cost-effective for diagnosing the common cold, but for peculiar or lingering problems, it can certainly offer solace, comfort, and guidance. Given a diagnosis online, a patient still needs to receive treatment and further testing in person. Although CrowdMed does not offer data related to patient outcomes, crowdsourcing medicine overall has the potential to improve patient outcomes. Finally, crowdsourced medicine allows enhanced access to international databases of published research, as other countries might have completed studies that have not yet been published in English or in American medical journals. Crowdsourcing medicine has the potential to transform healthcare worldwide, possibly opening up opportunities for a truly globalized and integrated system relying on advanced technological tools.

Crowdsourced medicine has its limitations. The process cannot be used for emergencies. Crowdsourcing medicine is expensive, too, and cost-prohibitive for most patients given its lack of being covered yet by insurance. Until crowdsourcing becomes fully integrated into the healthcare system, only the wealthy can afford it. Some healthcare workers decry the potential privacy issues, but crowdsourcing does not necessarily impede privacy any more than standard healthcare options. Likewise, the potential for wrong diagnosis or medical error is not necessarily going to be higher with crowdsourced medicine. On the contrary, “crowdsourcing works best when expertise is widely distributed,” (Sanghavi 1). Ultimately, crowdsourcing medicine may be the most promising means of receiving a quick, evidence-based, culturally competent diagnosis and assessment.







Works Cited

Arnold, Carrie. “Can the Crowd Solve Medical Mysteries?” NOVA Next. 20 Aug, 2014. Retrieved online: http://www.pbs.org/wgbh/nova/next/body/crowdsourcing-medical-diagnoses/
“Blind Spot.” Retrieved online: https://gimletmedia.com/episode/42-blindspot/
CrowdMed. Website: https://www.crowdmed.com/
Couch, Christina. “Crowdsourced Medicine Is Transforming the Diagnosis of Rare Disorders.” NBC News. 6 Mar, 2017. Retrieved online: https://www.nbcnews.com/storyline/the-big-questions/how-crowdsourcing-transforming-diagnosis-rare-disorders-n728306
Sanghavi, Darshak. “The Doctors Will See You Now.” Slate. Oct 6, 2010. Retrieved online: http://www.slate.com/articles/health_and_science/medical_examiner/2010/10/the_doctors_will_see_you_now.html