How RPA Can Help Get COVID-19 Vaccines to High-Risk Patients First

Ram Sathia, VP of Intelligent Automation at PK

While most of the public’s attention is on horse racing for an approved COVID-19 vaccine, another major hurdle is just around the corner: the distribution of hundreds of millions of vaccine doses. In today’s highly complex and disjointed health data landscape, technologies like AI, machine learning, and robotic process automation (RPA) are essential to ensure that the patients at greatest risk receive the vaccine first.

Why it is incredibly difficult to identify patients at risk

Once a vaccine is approved, it will take months or years to manufacture and distribute enough doses for the US’s 330 million people. Inevitably, hospital systems, general practitioners (PCPs), and provider networks must prioritize administration to high-risk patients and potentially focus on patients with underlying diseases and comorbidities. This requires an unimaginable amount of work by healthcare workers to identify patient cohorts, understand each patient’s individual priority level, and communicate instructions before and after the visit. The required degree of coordination between health systems and the urgent need to get the vaccine to risk groups clearly differentiates the situation from other nationally distributed vaccinations such as the flu.

A major challenge is that there is no infrastructure in place to simplify this process. All the data required for this is stored in different information silos. Many countries have legacy information systems or rely on fax to exchange information, which significantly hinders efforts to identify at-risk patients. In contrast, consider the earthquake risk data available in the United States. You can just open a federal geological map and see if you are in a seismic hazard zone. All information is in one place and can be sorted quickly. However, this is not the case with our healthcare system due to its fragmentation, as well as HIPAA and patient privacy laws.

There are several multidimensional barriers that make it nearly impossible for health care workers employed by providers and government health organizations to manually assemble patient cohorts:

Providers must follow CDC’s guidelines on prioritization factors that may include specific conditions, ethnicities, ages, pregnancies, regions, living situations (e.g. multi-generational homes) and disabilities based on the current guidelines for individuals at increased risk can. Identifying patients with these factors requires intelligent analysis of patient profiles against existing electronic health records (EHR), which are used by a wide variety of providers.

– Some hospital networks use multiple EHR and care management systems with limited ability to exchange and correlate data. These information silos prevent providers from displaying all information related to health data of the patient population.

– Data for supply outside the network that might require prioritization, such as A visit to the emergency room, for example, is often stored in payer data systems and is difficult to access for hospital systems and PCPs. This means that payer data systems must also be analyzed in order to effectively prioritize patients.

– All information must be shared and analyzed in accordance with HIPAA laws, and the mountain of pre-visit and post-visit planning communications and guidance shared with patients must also comply with federal guidelines.

– Patients with certain conditions, such as heart disease, may need additional procedures or tests (such as a blood pressure reading) before the vaccine can be given safely. Guidelines for each patient must be identified and clearly communicated to the care team.

– Vendors may not be able to distribute vaccines to all of their priority patients. Hence, providers need to coordinate care and may need to send patients to third party locations like Walgreens, Costco, etc.

All of these factors create a situation in which it is extremely difficult and time consuming for health care workers to make the vaccine available on a large scale to patients at risk. If the entire process of analyzing, identifying, and administering the vaccine takes just two hours per patient in the US, that’s 660 million hours for healthcare workers. A combination of analytics, AI, and machine learning could be a solution used by healthcare workers and senior physicians to identify patient priority, complemented by CDC standards.

How RPA can automate administration to high-risk patients

The technology is uniquely positioned to enable health workers to get vaccines into the hands of those who need them most faster than they could with humans alone. Robotic process automation (RPA) in the form of digital health workers with artificial intelligence can significantly reduce the time required to prioritize and communicate with high-risk patients. These digital health workers can intelligently analyze patient records and send notifications 24 hours a day, reducing the time spent per patient from hours to minutes.

Imagine a hypothetical situation where the CDC prioritizes certain risk profiles, making patients with diabetes among those likely to receive the vaccine first. In this scenario, RPA offers significant advantages in terms of its ability to:

Analyze EHR and population health data:

Thousands of smart digital healthcare workers could prepare patient data for analysis and then divide patients into different cohorts based on hemoglobin levels. These digital health workers could then intelligently review documents to compare hemoglobin levels to other CDC prioritization factors (such as recent emergency room admission or additional pre-existing or chronic medical conditions), COVID-19 tests, and antibody tests to determine the identify those most at risk and then identify a local provider with appointment availability.

Automate patient engagement, communication and planning:

Once patients with diabetes have been identified and prioritized, communication is essential to quickly planning the people at risk and preparing them for their appointments, including making them feel comfortable and informed. For example, digital health workers could communicate with diabetes patients via the protocol they should follow before and after their appointment – should they eat before the visit, what to expect during their visit and is it safe for them to go to work afterwards to return. It is also very likely that widespread vaccine administration would require a far greater amount of information than other health communications, as one in three Americans claims they would not be vaccinated if a vaccine were available today. On the order of magnitude, communication and scheduling together will potentially take up millions of hours, and all of that time takes healthcare workers away from actual care.

While the timeline for a COVID-19 vaccine to be approved is unclear, now is the time for hospitals to prepare their technology and operations for rollout. By adopting RPA, government health organizations and providers can prepare for success and ensure that the patients who need a vaccine most urgently get it first.

About Ram Sathia

Ram Sathia is Vice President, Intelligent Automation at PK. Ram has nearly 20 years of experience helping customers reduce time to market, improve quality, and increase efficiency through transformative RPA, AI, machine learning, DevOps, and automation.

April 3, 2021