By SCOTT WEINGARTEN, MD
COVID-19 exposed our country’s lack of centralized coordination when it comes to managing and preventing disease spread. Today, our public health system relies on flawed data and obsolete technology that fails to accurately track current and suspected cases, risk stratify patients, monitor disease progression or predict future spread. Not only do these blind spots create opportunities for the disease to spread, they also undermine the ability to safely plan for economic recovery.
What may surprise some, though, is the fact that we don’t have to start from scratch in order to build an effective system that stems the spread of COVID-19. In large part, the infrastructure we need is already here.
In 2009, Congress passed the HITECH Act, which allocated roughly $30 billion for providers to purchase electronic health records (EHRs). As a result of this stimulus, EHRs went from relative obscurity to ubiquity, and today about 96 percent of all providers are users of EHRs. Five years later, Congress passed the Protecting Access to Medicare Act (PAMA), which requires healthcare providers to consult with an approved Clinical Decision Support Mechanism (CDSM) in order to receive reimbursement for advanced imaging procedures for Medicare beneficiaries.
The net result of these two laws is that there is now visibility into nearly every patient-provider interaction in the United States at the moment that care is delivered, through more than a dozen CDSMs that have been certified by CMS. Although PAMA was intended for use with imaging, it’s not difficult to add on and repurpose decision support apps to conduct symptom surveillance for COVID, enabling healthcare workers to spot cases more reliably and earlier in the disease progression for prompt action.
The advantages of such an approach are manifold. First, symptom surveillance at the initial point of care is a vastly more viable strategy to stop COVID than the current approach, which is reliant on testing alone. If patients can be spotted when they first approach a primary care physician either through a telemedicine or physician office visit, at the onset of symptoms, quarantining and contract tracing could start earlier, possibly avoiding new cases and enabling more effective contact tracing.
Not only can symptom surveillance be an early indicator, but it’s also important context for interpreting COVID-19 test results. Research shows that false negative test results are produced at least 20 percent of the time and that likelihood grows higher if the test is administered early after infection with the disease. Given this, a New England Journal of Medicine article published in June recommended that symptom surveillance and clinical presentation information be collected and used to interpret test results. For example, if the patient presents with typical symptoms of COVID-19 when COVID-19 is circulating in the community, and the test comes back negative, the likelihood of disease is still high, the patient should be quarantined and contact tracing should be initiated.
Second, symptom surveillance within clinical decision support can also help guide the risk stratification and recommended treatment pathways for specific patients. Based upon clinical presentation and underlying risk factors, some patients may have a low risk of serious illness and a low probability of hospitalization even if they present with concerning symptoms. Conversely, those with mild symptoms may be at greater risk and should be followed by case managers, monitored in the home and reported as a potential future hospitalization case.
Last, because symptom surveillance can be automated, using natural language processing and machine learning to understand physician notes, it can be leveraged to aggregate data nationally to pinpoint new hot spots and make predictions about future spread. This is particularly vital since national guidelines for safe re-openings require data demonstrating a downward trajectory COVID-like symptoms.
The CARES Act allocated a significant amount of funding to modernize surveillance programs to fill in the void left by today’s antiquated surveillance capabilities. Some of this funding could be deployed to create improved syndromic surveillance capabilities. This may be the fastest and most scalable approach to enabling clinical surveillance of patients with suspected and confirmed COVID-19 in communities across the United States.
With the existing work in place to monitor and surveil communities, we have the ability to get ahead of the surges we expect to see in the fall, as well as any future pandemics, bioterrorism and other potential emergent threats to the health of Americans. Because of all of the leg work that was conducted over the last decade, the time it takes to implement this technology and “go live” will be compressed. It is now up the local health departments to work with the organizations they feel best position them for success in this regard.
Dr. Weingarten is consultant to the CEO and Professor of Medicine at Cedars-Sinai; CEO of Stanson Health, a Premier company; and Chief Clinical and Innovation Officer at Premier.