Our friend Marty Tenenbaum at Cancer Commons recently published an important article in STAT, documenting how researchers can develop treatments for new patients by looking at previous cases. In response to COVID-19, the healthcare industry is scrambling to find effective treatments that might combat the virus, yet as Marty points out: “instead of frontline doctors deciding to try experimental drugs on the fly, they could tap into the collective experiences of thousands of doctors around the world to learn what has worked and what hasn’t for similar patients.”
What he’s describing is a public database of patient records that show treatment outcomes across a wide spectrum of cases. With such a database, doctors could use AI technology to mine through these records and “learn how treatments have worked for similar patients in the past.” In short, rather than subject a patient to numerous trials and potential treatment plans that may or may not prove effective, physicians can use patient data to hone in on what has already shown to work.
As Marty also adds: “reckless use of a drug can cause devastating side effects, including death.” So why aren’t doctors everywhere using a giant database of patient records in order to limit the amount of testing necessary, and instead focus on what’s already shown to be effective? Because that database doesn’t exist yet. But, hopefully, with COVID-19 pushing healthcare into the digital age, we might be getting closer to making this dream a reality.
When it comes to cancer, patients in advanced states have often exhausted their therapeutic options, so the data is scarce as to what works and what doesn’t. But the real reason more research into previous therapies isn’t happening is due to the lack of interconnectivity between existing databases and the unwillingness of our medical institutions to share patient information. As we at Ciitizen know all too well, matching patients to clinical trials for new drug therapies can take months if not years. Yet, as Marty concludes, if we had an interconnected database of patient records, “the answers these trials are seeking would, in many cases, already be known.”
In another recent article from STAT, Matthew Herper links our inability to conduct clinical research with any real speed to thousands of COVID-19 deaths, stating that “our inability to start and run clinical trials faster — whether in normal times or in a pandemic — is a legacy of our decision not to develop the technologies and approaches that would make doing so easier. Simply put, more people will die from COVID-19 because we cannot study drugs more quickly.” Like Marty, Herper points out that our government’s touting of remdesivir and chloroquine has gained some traction, but more data would help us know the potential consequences.
And why don’t we know more about patient histories using both drugs? Because the databases that hold those records are not interconnected. Medical researcher Raj Mehta tells Herper that the data is there, but the partnerships to use the data have not materialized. “What we don’t have is anything useful to produce evidence-based medicine,” he states, “but if we could trade all the data silos for a fully integrated, nationwide, RCT platform in EHRs, we would all do it in a heartbeat.”
A recent article in the Lancet echoes these points, stating that “patient-level COVID-19 data is unfortunately not publicly available” and that in our interconnected world it’s a shame that no database exists, given we clearly have the technology to create one. The lack of data concerning potential COVID-19 treatments is showcasing our health system’s greatest inefficiency on a national scale and drawing more attention than ever to our inability to share patient information quickly, efficiently, and effectively. But due to the growing pandemic, there’s a growing demand for major players to jump in and help remedy the situation.
As an example, Google recently announced it would make COVID-19 datasets freely available to researchers in an attempt to remove current barriers and provide access to crucial information quickly and easily. The Regenstrief Institute also announced that it would be conducting a survey about COVID-19 symptoms and share that data openly to “help state, federal and international health agencies monitor trends and anticipate outbreaks.” A recent article in Health IT Analytics summarizes how COVID-19 could ultimately refine AI and data analytics in healthcare, writing that “the stage is set for major industry players to come together” and “overcome healthcare’s long-standing data challenges.”
Through data partnerships that help us better understand COVID-19 and the efficacy of potential treatments, the stage will indeed be set for similar data sharing more in line with what Marty at Cancer Commons and we here at Ciitizen have been talking about for years: an interconnected patient database that allows researchers to inform current treatments of cancer patients by analyzing previous cases.
With all the devastation that COVID-19 has brought upon our planet, we can hope that—at the very least—it will help force the shortcomings of our healthcare system’s siloed data infrastructure into the light.