Click to learn more about author Ben Sharma
Early successes from Data Management projects deliver speed, agility and scale.
As the current pandemic continues to spread, and the worldwide number of cases crosses the four million mark, public, private, and government organizations are using big data and analytics in new ways to help with prevention and treatment. From using location data to manage exposure tracking, to using supply chain data to manage testing resources, to aggregating trial data in the race to develop a vaccine, virtually every COVID-19 approach is harnessing big data analytics to solve a broad array of testing, treatment, and containment challenges.
Because the data involved is often massive, housed in a variety of companies, and changing by the minute, the success of each effort requires data that is easily ingested into an analytic platform, scalable storage and compute environments, and an accelerated data pipeline capable of processing near real-time information. In addition, these data operations must maintain a level of data security and compliance in order to meet HIPAA and other regulatory requirements.
The past several weeks have shown the increasingly critical need for companies taking on ambitious COVID-19 efforts to optimize their data supply chain. The following are several examples where Data Management and operations optimization increases the efficiency, governance, and rapid scalability of current pandemic efforts.
Open Research Data Set: Data is driving the race for a cure. Initiatives like the COVID-19 Open Research Dataset (CORD-19) aim to create central repositories for a wide range of projects. Clinicians, researchers, and other healthcare professionals can contribute, query, and analyze data. Collaboration, unifying observations, and validating findings will accelerate pandemic cure timelines.
Vaccine Partnerships: Even as countries succeed in flattening the curve, virus resurgence will happen, and a vaccine remains the best eradication solution. The traditional vaccine development timeline, even in a best-case scenario, is too slow, and big data projects are discovering inefficiencies that will compress the clinical trial timeline. Unique, often global, partnerships, such as the one between GSK and Innovax, accelerate meaningful analytics using real-time data across larger data sets.
Predictive Analytics for Hospital Bed Capacity: While the world’s largest technology companies have proven well-equipped to create repositories for research and data, the logistics and manufacturing industries are less agile. Test kit production and distribution continues to be a challenge with the dependence on the global supply chain being a primary factor. One area where Data Management and analytics are succeeding is predicting the number of hospital beds needed by region and timeline. For states dealing with a high case volume, such as New York, having advanced data for hospital demand results in better supply distribution, staffing decisions, and, ultimately, fewer fatalities.
Telemedicine Becomes Primary: Most states have a stay-in-place order in effect, discouraging people from visiting clinics, hospitals, and treatment centers unless critically needed, and the shortage of COVID-19 test kits has meant that many individuals aren’t able to be tested. These factors have fueled an exponential rise in telemedicine. While this sudden shift in patient treatment creates a data operations challenge for an industry accustomed to in-person, on-site records retrieval and entry, it also creates the opportunity for modernizing treatment with information and process efficiencies.
Patient Data Registry: Several healthcare partnerships have been quickly formed to pool patient data with the goal of developing precise recommendations for the prevention of COVID-19. LabCorp and Ciox Health announced a joint registry to better diagnose and treat the virus while also aiding future pandemic prevention efforts. Datavant and the American Heart Association offer additional examples of companies creating anonymized data pools for the purpose of understanding COVID-19’s effect on patients with particular conditions. Each of these efforts ultimately seeks to lower infection rates and reduce fatalities; the use of streamlined, real-time data operations increases the Data Quality while ensuring patient data security.
AI to Identify High-Risks: COVID-19 has a wide range of severity, and certain populations are at higher risk than others. Many factors contribute, including population density, pre-existing conditions, and age, to name a few. This mix of factors creates an ideal use case for AI and machine learning to look across massive data sets and determine the precise conditions by region that lead to greater risk. Arming government and healthcare organizations with these precise, localized analyses will increase the effectiveness of preventative measures.
Looking Towards the Future
The most important outcomes of COVID-19 Data Management and analytics have yet to be developed. Companies, such as Amazon’s AWS and Microsoft’s Azure, are making as much data as possible available for future analytics through a coordinated, broadly-contributed data lake. Companies that leverage these data repositories, ingest data sets into their own platforms, and apply the data to their BI and analytic tools will be best positioned to successfully accelerate their COVID-19 eradication, treatment, and prevention projects. Modern DataOps has the potential to not only solve many of the challenges facing COVID-19 today but also will establish the healthcare infrastructure needed to prepare for the next global pandemic.