The evolving COVID-19 pandemic is forcing healthcare systems to seek new ways to meet the needs of affected patients with greater speed, flexibility and efficiency than ever before. While cutting-edge data analysis techniques are available that can provide healthcare professionals with the tools they need to quickly answer urgent medical questions, they depend on large-scale access to real data. Hospitals and health systems play an important role in providing the data pool needed to realize their full potential in this health data revolution.
Thanks to advances in machine learning, we can now analyze millions of medical cases in minutes. Five years ago, such speed was unthinkable. Hospital systems across the country, including MedStar Health where I work, are beginning to responsibly share large repositories of anonymized patient data, fueling this faster, more representative array of medical research. We are committed to using aggregated, anonymized patient data in the public interest to provide quick and accurate answers to our most pressing medical questions.
Currently, 20 health systems across the country are pooling their resources to apply information gained from interactions with doctors, laboratories and medical equipment. But we need to engage more hospitals to further uncover the promise of more accurate data to improve patient care and find faster treatments. The benefits of combining large, generalizable datasets are endless. Here are three to keep in mind.
First, we now have the ability to find trends and connect information from seemingly disparate cases using advanced analytics, artificial intelligence (AI), and machine learning. Connecting these critical points provides greater diagnostic accuracy or new insights into how a treatment or clinical practice works in the real world. For example, recent research have shown that artificially assessed ultrasound images of the heart can predict mortality in COVID-19 patients, even when the same image, as interpreted by a human medical expert, cannot.
Second, scalable data platforms offer clinicians and researchers a speed advantage. If we have enough aggregated data at hand, we can reduce the time it takes to address emerging and urgent health issues as they arise. Consider halting deployment of the COVID-19 vaccine earlier this year due to concerns about rare side effects. With the right dataset across multiple health systems, we could analyze the anonymized medical records of all those vaccinated in less than a day and identify possible causes quickly and efficiently.
Finally, large datasets allow us to incorporate anonymized patient data from different communities, geographic regions, and races. Clinical trials are notorious for not having enough participants from under-represented communities. Medical research from real-world datasets is far more likely to effectively represent our communities on a wide national scale than ever before. Research based on data from different communities can accelerate our understanding of how social determinants can affect health.
Researchers like me are genuinely excited about advances in health data science and the role that academic health systems can play in improving the health of everyone. Machine learning has shifted the focus of our research away from “how do we get the data we need?” to “what questions can we ask about this data today?” That’s why MedStar recently joined along with other healthcare systems as part of Truveta, the company helping to bring this medical data revolution to life. Participating health systems ensure proper and ethical use of data.
Data in healthcare is too fragmented, incomplete, or scattered across disparate systems due to limitations in how electronic health records are shared and work together. As healthcare professionals, we understand that the public is concerned about how their personal medical data is processed. However, healthcare systems have a long history and a good track record of being responsible custodians of this data, backed up by laws that carry severe penalties if violated. We are uniquely placed to ensure that patient data is protected and used for legitimate academic purposes only.
Together, we can apply this approach to implementing data science to not only optimize our approach to meet the current needs of our patients, but also change the future of medicine. Healthcare executives should start conversations with their chief technologists and data scientists and explore the possibilities of how their systems can partner with other organizations to uncover the promise of this evolving health data revolution.