P36 Forecasting Trends in Syphilis Using Prediction Markets

Wednesday, March 10, 2010
Pre-Function Lobby & Grand Ballroom D2/E (M4) (Omni Hotel)
Thomas Peterman, MD, MSc, Epidemiology and Surveillance Branch, Centers for Disease Control and Prevention, Atlanta, GA, Fred Bloom, PhD, Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta, GA, Phillip Polgreen, MD, MPH, Department of Internal Medicine, Carver College of Medicine, University of Iowa, Iowa City, IA and Forrest Nelson, PhD, University of Iowa, Iowa City, IA

Background: Prediction markets have been used to predict elections, business trends, and influenza activity.

Objectives: To describe the implementation of a market to predict syphilis trends, recruit traders, solicit ideas for future prediction markets.

Methods: In July, 2009, the first prediction markets for syphilis were opened.  Anyone could apply to be a trader.  Traders were given $100 to buy and sell contracts.  One market predicts the quarterly total number of primary and secondary syphilis cases in the United States.  In the third quarter of 2008, there were 3216 cases reported.  For 2009, contracts were available for the predicted totals of: ≤3350, 3351-3450, 3451-3550, 3551-3650, and ≥3651 cases.  At the end of the quarter, traders earn $1 for each share they own of the contract that correctly predicted the total.  Other shares are worthless.  While the market is open, the share price reflects the market’s prediction for the total at the end.  A second market is predicting quarterly total cases of primary and secondary syphilis cases for women.

Results: By September 2, 2009, there were 22 traders.  For quarterly total number of cases, ≤3350 last sold at 1 cent per share; 3351-3450 last sold at 3 cents per share; 3451-3550 last sold at 57cents pre share; and both 3551-3650 and ≥3651 last sold at 32 cents per share.  Most are predicting a modest (235-334 case) increase in total primary and secondary syphilis for the quarter compared to 2008.  For women, traders are predicting no increase compared to the third quarter of 2008.  Share prices are fluctuating widely, probably due to small numbers of active traders.

Conclusions: Syphilis prediction markets have started.  More traders are needed.  Sign up at http://iehm.uiowa.edu.

Implications for Programs, Policy, and/or Research:If we can establish a market to predict the number of cases, perhaps we can establish a market to help predict successful interventions.

See more of: Poster Session 2
See more of: Oral and Poster