PA Cure Twitter Study

Validating a tool to monitor and predict high morbidity health conditions and real-time dynamic health events among Pennsylvanians

/sites/default/files/Twitter.png

With the recent explosion in the volume and types of data collected in medicine, there is the opportunity for a paradigm shift from a health care system that reacts to disease exacerbations as they happen to one that is proactive in targeting disease and clinical events before they occur.

The aim of the project is to integrate social media data with statewide data to build and validate a tool to monitor and predict high morbidity health conditions and real-time dynamic health events among Pennsylvanians.

Our investigation has two primary aims.

The first aim is to assess whether Twitter provides valid estimates of the prevalence of major health conditions that drive morbidity and mortality across Pennsylvania and its communities.

The second aim is to investigate the use of Twitter data to predict dynamic health events, such as food-borne illness and influenza. We will develop computational models utilizing Twitter data for accurate prediction of dynamic public health events for Pennsylvania and its communities.

Through this work, we are using new tools to map health across the state of Pennsylvania.

This work is funded by the PA Department of Health.  Principal Investigators are Dan Polsky and Kevin Volpp.  Additional collaborators include Amol Navathe, Mitesh Patel, and Peter Groeneveld.