Advances in computing technology, big data analytics, and electronic medical records have dramatically progressed over the past decade. Both the will and the necessary data now exist to treat the entire US health care system as a true learning system. However, like so many hollow claims of huge impact, this potential will never be realized without nuanced attention to potential obstacles and necessary actions for using our rapidly accumulating big data to actually improve health.¹

The learning health system model has emerged as a means to identify quality gaps and evaluate interventions using data in real time. Precision medicine and implementation science, when combined, can produce a learning health system capable of high-quality care and high-impact research.² The Institute of Medicine has described a vision where 鈥渟cience, informatics, incentives, and culture are aligned for continuous improvement and innovation鈥�and new knowledge is captured as an integral by-product of the care experience.鈥�³ A recent National Academies of Sciences, Engineering, and Medicine report echoed this approach for the military medical system.⁴ Select health systems have thus been able to master the rapid acquisition of big data with ongoing analytics, allowing for real-time learning.