Skip Navigation
 
CHS blue logo square
 
 
Home > Our Work

Disease Prediction

Use your knowledge to predict public health outcomes.

Sign up today to become a #HealthSecurityForecaster.

Researchers at the Johns Hopkins Center for Health Security are challenging YOU to make predictions about future outbreaks and other health security related events for the Collective Intelligence for Disease Prediction project. Your predictions will provide public health leaders with forecasting data to help inform their decisions on preparedness and response policies and interventions.

The 5 highest-rated forecasters, over the course of the project, will receive a monetary prize.

It’s easy: register to participate, thoughtfully predict answers each week, have a chance to win a prize, and make a difference in the Health Security field.

 

Collective Intelligence for Disease Prediction: Frequently Asked Questions

Who is asking these questions?
The Johns Hopkins Center for Health Security (the Center) developed this set of forecasting questions for its Collective Intelligence for Disease Prediction project, which involves gathering, analyzing, and disseminating collective intelligence from public health experts and expert predictors about the outcomes of potentially significant public health events.


What benefits are there to participating in this forecasting work?
The goal of the Center's project is to provide information for decision makers that can supplement current traditional surveillance and modeling efforts and improve response to infectious disease and other public health emergencies. By participating you will be contributing to this effort.

If you make accurate predictions and participate consistently on the platform you may become one of our top predictors, making you eligible to receive a gift card, have a donation made in your name, and/or have the opportunity to be highlighted in the Center for Health Security's digital communications.


Who can participate?
This collective forecasting platform is aimed toward public health experts, doctors, epidemiologists, modelers, risk experts, vector control officials, individuals with on-the-ground understanding of conditions surrounding disease outbreaks, and others who are interested in outbreaks. Anyone can participate, but we ask that participants understand the serious nature of predicting public health outcomes.


What information from participants will be included in project materials?
All data collected will be used for research purposes and in ways that will not reveal any individually identifiable information. Any reports, publications or presentations related to this project will report data in aggregate or by demographic cohort only. In addition,

  1. Any personal information that could identify you will be removed or changed before files are shared with other researchers or results are made public;
  2. Any predictions you make are anonymous and identified publicly only by the username you select


What are the prizes and how will they be determined?
Prizes will be awarded at two points during the contest: at 6 months and 12 months. The five highest-rated forecasters at each point will receive a choice of AMEX or VISA gift card, or a donation in their name to a charity of their choice.

First place prize is $599, 2nd place prize is $400, 3rd place prize is $250, 4th place prize is $150, and 5th place prize is $100.

An additional 20 participants will win $50 each in a lottery drawing. The lottery system is also performance-based with lottery tickets awarded based on ranking. Thus, those who perform better are more likely to win a prize.

Note: Individuals currently affiliated with the Johns Hopkins Center for Health Security are ineligible for any of the cash awards. Students and personnel from other parts of the University are eligible.


How is my forecasting performance scored?

Your performance is measured relative to everyone else's (the crowd) performance. Two factors are important for your score - timeliness and accuracy. Every day, the platform records your latest forecast on every question. When a question is resolved, we can compute the accuracy of each of your daily forecasts on that question, then compare it to the median accuracy of other participants. You score points when your forecasts are more accurate than those of most other participants, otherwise you lose points. The better or worse your relative accuracy is, the more you win or lose points. You score 0 when you are right in the middle of the pack.

The bottom line is that you are scored for both the accuracy and timeliness, so being right before others increases your score. A full explanation of the scoring method is included at the bottom the page MY DASHBOARDS>MY PERFORMANCE.


How often should I update my forecasts?
As often as possible. The best forecasters often revise their forecasts to reflect their most up-to-date understanding of the questions. A rule of thumb is to reexamine your forecast on a regular basis and update it whenever new information becomes available to you, or if any of the assumptions underlying your forecast appear to be changing. The platform lets you set Alarms for each question to prompt you to review your forecast.

That said, when you update your forecast several times on any particular day, only your latest forecast will be scored that day, so there's usually no point in updating your forecasts more than once a day.


How do I get scored on days when I don't forecast?
On days before you start forecasting a question, your daily score is 0, as if your accuracy was no better or worse than the median forecaster.

Once you start forecasting, your latest forecast is carried over from one day to the next, until you update it, or until the question is resolved.

 

 

Our Mission

To protect people’s health from epidemics and disasters and ensure that communities are resilient to major challenges.