Living in the Era of Big Data, Algorithms, and Inefficient Health Care: Stories from the Frontlines of Health Data Science
With the unrelenting exponential growth in the volume of health and health-related electronic data, the potential to rapidly and accurately monitor, predict, intervene, and ultimately improve human health has never been more promising. However, care must be taken to uncover and understand the capabilities and limitations of these resources. Electronic data is rife with biases and other data quality issues; machine learning and other classes of algorithms, when improperly applied, can produce spurious findings; and ultimately, humans who consume the fruits of these rapidly emerging amalgamations of “data plus algorithms” may be limited in their capacity to apply these findings. Dr. Grannis will share outcomes and lessons learned from recent projects leveraging “data plus algorithm” using unparalleled data from one of the country’s largest and longest-tenured health information exchanges. He will also highlight important future directions in the health data science field.