If the fine folks at Health 2.0 are any indication, contests seem to be the best way to spark innovation these days. CureTogether is throwing another contest into the mix this summer:
Over the past 3 years, CureTogether has gathered millions of patient-reported data points on symptoms and treatments for over 500 conditions. Now it’s time to test on a larger scale how well CureTogether data represents the general population. Do they match up or not?
So we’re running a contest to tap the most brilliant stats minds out there. Challenge our dataset! See whether or not it holds up to existing research studies. Why? You’ll be helping to demonstrate the effectiveness of online platforms for medical discovery, and ultimately helping to reduce global suffering.
There are cash prizes, and the deadline to join the contest is June 29, 2011.The contest will be judged by Dr. Will Dampier from the Center for Integrated Bioinformatics at Drexel University, Prof. Seth Roberts at Tsinghua University and University of California at Berkeley, and Prof. Victoria Stodden from the Department of Statistics at Columbia University.
Examples of discoveries to replicate include disease correlations and comorbidities, treatment effectiveness studies, symptom prevalence studies, gender/age prevalence for different diseases, symptom biomarkers for disease diagnosis or treatment response, and anything else you find interesting or relevant (some ideas can be found in CureTogether’s research findings so far)
NOTE! The goal of the competition is not to reach any predetermined conclusion. It is to measure as well as we can how much our conclusions agree or disagree with the PubMed literature. Whether you find agreement or disagreement or something in between will have no effect on your chances of winning.
Here’s how it works:
– Pick one or more of CureTogether’s top conditions
– Find one or many discoveries on PubMed or GoPubMed to compare to CureTogether data
– Ask me for the anonymized data
– Prepare analyses and/or visualizations that demonstrate and quantify how well CureTogether data matches up with the independent studies (use R, Python,
SQL, SAS, SPSS, Excel, Sweave, or your favorite tool)
– Send me what you find
– Win $3,000 top prize, $1,000 second prize, or one of four $250 third prizes
Good luck everyone!
Alexandra Carmichael is CEO of CureTogether
Categories: Health 2.0