Recent collaborations and partnerships have showcased the potential that data can have in shaping the future of healthcare.
Christopher M. Cornue
President & Founder, sláinte global partners (sgp); Expert, International Society for Quality in Healthcare (ISQua)
There is a collective understanding across healthcare leaders that our industry is in the throes of constant disruption, a drive to become more efficient, and an urgency to be more responsive to the needs of patients and communities. While we’ve seen many incremental efforts to support the evolution of this change over the past several years, we know there is a great deal of work still to do. Core to any of these efforts is a fundamental need to understand and utilize data in comprehensive and informed ways. And there is a lot of data to be used.
Forbes reported in 2021 that a “single patient generated nearly 80 MB of data each year in imaging and EMR data” and further reported that RBC Capital Market projected “by 2024, the compound annual growth rate of data for healthcare will reach 36 percent.” Historically, we’ve naturally collected and used critical data in caring for patients, such as clinical, demographic, and other key data sets; however, these data have traditionally been siloed, transactional, and retrospective. This is changing, and there are now many examples of partnerships and efforts that are leveraging data to inform focused care delivery, new insights, more patient-focused outcomes, and usage across silos. Among the many, here are a few examples of the impact data can and will have in the years ahead.
A team effort
Children’s Mercy Hospital in Kansas City recently launched a program that uses predictive analytics and digital technologies to help patients with Type 1 diabetes cope with their disease via new treatment models and approaches. In partnership with a cloud-based predictive analytics platform, and leveraging patients, clinicians, researchers, Children’s Mercy is able to predict patient outcomes and test focused treatment interventions via digital technologies to ensure the right intervention is applied for the right patient at the right time. This has led to several tested treatment interventions within a short period of time, including a mentorship program, remote patient monitoring to allow for self-reporting and two-way interactions between visits, and the ability for “nudging” to promote better adherence to care plans.
One of the largest efforts to utilize data to inform clinical research and decision-making is a multi-health system collaboration that launched in 2020. Born originally from an idea within Washington-based Providence, Truveta includes 20 health systems (representing more than 16 percent of clinical care in the United States) who are contributing de-identified clinical data from their respective health systems. Leveraging the scale and richness of these data sets, this collaboration is working to promote new insights into care delivery, advance health equity, and improve the health of patients and populations. In late 2021, Truveta released some early insights from this collaboration specific to COVID-19. Specifically, they found that “9-15 percent of fully vaccinated people who have a breakthrough infection end up hospitalized” and “less than 1 percent of vaccinated people experience adverse events.” As future implications are considered, just imagine the impact these data can have in informing new treatment approaches that can advance care for diseases that, to date, have no treatment.
While the use of data will likely impact all conditions and diseases over time, there has been a strong focus on using data and technology to impact how we address cancer. There are many examples of this work, but a recent one highlights the ability for artificial intelligence-based risk models to examine raw patient data to advance more personalized and precise cancer screenings. It’s been long understood that mammography screenings are one of the best ways to detect breast cancer, but the uniqueness of each person — coupled with potential false-positive results and broad guidelines — yield an imperfect approach. Scientists from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) sought to address this and used AI to develop more personalized screenings such as a specific timing in the future for screening based upon past screenings, diagnoses, and age. Through a technology they created and tested on patient datasets at hospitals in the United States, Sweden, and Taiwan, the team has seen recommended screenings at more precise timing than an annual exam, as well as fewer mammograms than would have been previously needed. The impact of this work in the future is potentially profound, from earlier detection of cancer to reduction of cost and the peace of mind that patients will receive. Additionally, the ability to leverage these results and approaches to other conditions and diseases is exciting to consider.
Finally, the past two years have provided the best use case for using reliable data to inform action, caution, planning, and response. The COVID-19 pandemic showcased a host of examples whereby data was leveraged to inform hot spots and trends related to COVID-19 surges, hospitalizations, deaths, and capacity constraints at hospitals. Whether they were led or created by governments, leading healthcare institutions, or academic data institutions, these tools became commonplace for those both inside and outside of healthcare. It’s likely that this extent of reporting will become incorporated into daily life in the years and decades to come and will usher in a new way of informing the care we provide, both in clinical and community-based settings.
It’s important to note the usage of data in any effective way is wrought with many organizational challenges and opportunities, such as staff competency, governance, infrastructure, security, and regulatory and reporting requirements to name just a few. The focus on these skillsets and structures is critically important for success, which is why existing roles such as chief information officers, chief digital officers, and chief medical information officers have evolved and healthcare is now hiring data scientists, data architects, data engineers, data analysts, statisticians, and economists to complement internal knowledge. These resources are driving the desire for more meaningful insights and a shift from descriptive and diagnostic analytics to more predictive and prescriptive analytics, which will inform a better ability to make the needed changes in healthcare going forward.
Although many health systems recognize the importance of these rich data that exist within their organizations, still others have yet to realize the incredible value that can be created from such information. The partnerships and examples described above give examples of what’s possible when the right resources and infrastructure are implemented.
These are only a few of many examples where data in healthcare is being used to advance care delivery, address current challenges, and provide caregivers more relevant and useful insights to help with their day-to-day work. There is so much more to consider beyond these examples. How will machine learning and artificial intelligence shape the future of healthcare? How will clinicians be able to effectively pour through and consider all these data sets in such a way to make more informed decisions? How will new data sets add to the expansive data that allow for more focused treatment options, including precision care? There are developing models, start-ups, and answers to these questions that the length of this article won’t permit; however, these all support the premise that the time for transformative data has arrived, and it will dramatically shape healthcare in the coming years.