Clinicians, scientists, and researchers are learning more about SARS-CoV-2 (COVID-19) each day. The process of scientific discovery is playing out in the public eye at a velocity not seen in recent memory. What could be viewed by the average consumer as confusing and hard to keep up with, is, by many in healthcare, viewed as innovation at its best. In the months since March 2020, the understanding of COVID-19 has evolved dramatically. We have watched a real-time shift in our understanding of how the virus is transmitted, who the most vulnerable populations are, and which treatments are most likely to work.
With so many patients being treated in acute care settings across the country, we—as a country—are better positioned than ever before to capitalize on de-identified, HIPAA-compliant patient data to inform scientists, researchers, and public health authorities with real-world evidence (RWE). As the number of infected patients and deaths continue to rise globally, clinicians around the world are struggling to fully understand all of the implications of the COVID-19 pandemic.
Having sufficient data is the first step to evaluating the impact of COVID-19 and designing adequate local and national public health responses for the management and containment of this pandemic. Here are just five areas in which data is already making an impact in helping providers, healthcare systems, and life sciences companies better deal with the pandemic.
Discovering New Therapeutic Options
Since the start of the outbreak, we have seen major life sciences companies leverage RWE to understand the impact of COVID-19 on pressing market dynamics. For instance, many pharmaceutical companies are vying for data that will help them evaluate the clinical and economic outcomes from the use of certain therapeutics. There could be drugs that work to treat COVID-19 specifically, or on a yet-to-be identified co-morbidity. Without constant data mining, these linkages could remain undiscovered.
For example, the research team at Premier Applied Sciences was conducting a comparative effectiveness study to assess an intervention around transitions of care for a specific disease area. One of the principal researchers decided to comb through the data to see if there were any COVID-19 patients inadvertently captured. Not only did they find these patients, but they also found that the product they were studying impacted a COVID-19 co-morbidity so significantly that it trended towards a positive impact on COVID-19 itself. As a result, a larger, specific analysis has been undertaken that will hopefully result in a way to blunt the impact of COVID-19. This was a previously unrecognized risk factor for severity of disease for COVID-19 that is now being brought to light.
Finding the Right Partners for Clinical Trials
While these types of findings are exciting and generate optimism around patient outcomes, a regulatory process must still be adhered to in order for new products to be introduced to the general public. The rigorous clinical trials that life sciences companies are accustomed to conducting in order to gain FDA approval are met with the challenge of finding health systems that can accommodate prospective studies during a disruption such as COVID-19. Whether it is a new clinical trial that a manufacturer is trying to initiate or an ongoing one with months or years to go, organizations are finding it difficult to identify and maintain study site and patient populations.
One solution for this is pre-qualifying study sites that could commit interest and resources to COVID-19 clinical trials. For example, we found 22 systems that represent over 100 hospitals with geographic and virus timeline diversity, and reflect a provider network prepared to expedite study startup and research activities to assist with COVID-19 resolution.
Helping Medical Device Companies Adjust
Hospitals across the country have spent the last six months struggling with the economic burden this pandemic has caused. Historically, “elective” procedures have been a source of revenue for these institutions, and with the fear of needlessly spreading COVID-19, any healthcare procedures not considered “essential” have experienced a significant dip. This has also had an enormous impact on the medtech companies that provide the goods and services most used in these procedures.
While more recently we have seen a gradual return to normal in regards to elective procedures, the ups and downs of the pandemic have not helped medical device companies in their quest to predict demand and adjust supply accordingly. Up to the minute data is an important asset for these organizations to give them a window into when specific surgeries are coming back online. Without this type of data, and with no historical reference, it would truly be a guessing game for these companies to determine their response to the pandemic.
Predicting the Need for Patient Care
Another area critical to the healthcare community is the ability to leverage artificial intelligence models to help clinicians better diagnose and triage patients. Data-driven, predictive models that provide guidance on anticipated patient admissions and inpatient length of stay are a reality today. But specific areas of analytics have proven most valuable to healthcare systems in order to do these predictions.
One example is the aggregation of symptom data in the ambulatory setting and from telehealth visits. Patients typically contact an urgent care or their primary care provider with the earliest potential symptoms of COVID-19, which can provide early and valuable information.
Technology that can flag potential COVID-19 patient cases directly in the electronic health record (EHR), at the point of care, are critical to recognizing case surges before they hit the hospital. Natural language processing contextualizes free text in a medical record and interprets the patient information as it looks for notes that include comments such as “trouble breathing” and “loss of taste.” As this information is connected with a patient’s COVID-19 test results, modeling of disease progression is enabled, and insights are gleaned around the likelihood of future hospitalizations. This allows hospitals to more accurately allocate or even de-allocate critical healthcare resources.
Forming Partnerships for a More Complete View
An emerging trend from the COVID-19 pandemic is the need for partnerships between life sciences organizations and providers. Technology and data are crucial for the global acceleration of treatments and improved patient outcomes. No singular data set or entity has a complete view of a patient journey, but by pooling together data and insights we get a more comprehensive window that can help make things more predictable.
This concept was recently demonstrated when researchers used de-identified hospital utilization data matched with a life sciences company’s distribution data to determine potential drug shortages and overallocations. Finding information like this early ensures life-saving medications are manufactured at appropriate quantities and prevents a potential healthcare calamity.
It is essential for the research community to continue to catalogue and share information that is learned from data during this pandemic. It is clear that a lot is still unclear with regards to COVID-19. Many of the most critical discoveries can come from the intentional and only partially directed data mining. The clinical management of COVID-19 will be remembered, in part, for how data was used to advance the response for patients in real time.