Science & Enterprise logo
Science for business people. Enterprise for scientists.

Crowdsourcing Can Aid Health Research, but Guidelines Needed

Raina Merchant
Raina Merchant (University of Pennsylvania)

Researchers from the medical and business schools at University of Pennsylvania in Philadelphia evaluated crowdsourcing as a research tool in health care, and found it has potential to improve quality and lower costs of studies, but ground rules are needed for the technique. Findings from the team led by Penn emergency medicine professor Raina Merchant appear online in the Journal of General Internal Medicine (paid subscription required).

Merchant and colleagues are no stranger to crowdsourcing. Last year, a group at Penn led by Merchant held MyHeartMap Challenge, an eight-week crowdsourcing competition in Philadelphia to locate automated external defibrillators in or near the city. The competition resulted in the identification and photographing of more than 1,400 of these devices in some 800 different buildings. Two individuals each located more than 400 defibrillators and were awarded the top prizes.

In this study, the Penn team — including participants from the university’s medical school, Wharton School of Business, economics department, library services, and a local veterans hospital — sought a better handle on the scope and different ways crowdsourcing is used in health care research. Their review of literature, including searches of online services Yahoo Answers and Quora, unearthed 21 studies that used crowdsourcing techniques in some way.

The Penn group found the 21 studies together enlisted the services of some 136,400 people, performing four kinds of tasks: problem solving, data processing, surveillance/monitoring, and surveying. The crowds took on tasks such as quickly tracking outbreaks of H1N1 flu and classifying different polyps in the colon.

First author and medical student Benjamin Ranard notes that “studies we reviewed showed that the crowd can be very successful, such as  solving novel complex protein structure problems or identifying malaria infected red blood cells with a similar accuracy as a medical professional.”

The studies did not so such a good job, however, in telling about the crowds being sourced. In fact, only one of the 21 studies reported on the age, gender, or race of the participants. Other health-related studies, such as clinical trials, routinely collect and report data such as age, gender, and geographic location of participants.

Crowdsourcing, say the researchers, can be valuable for harnessing human energy for large-scale simultaneous problem solving or reporting data about the environment. However, notes Merchant, “This review points to the need for streamlining the process and implementing more rigorous guidelines for this approach.”

Merchant will have a chance to put some of those ideas into practice. Earlier this month, she was named director of the medical school’s Social Media Lab that’s exploring new communication channels to enhance the school’s ability to understand and improve the health and health care of patients and other populations.

Read more:

*     *     *


Posted

in

,

by