COVID-19: Q&A on health systems using local modeling to plan for the uncertain year ahead

April 17, 2020

HRI talked with PwC principal Vaughn Kauffman about the role predictive modeling can play to help healthcare systems plan for their local regions in the uncertain times of COVID-19.

PwC Health Research Institute (HRI)

Predictive modeling can provide healthcare systems deep insights about the patient pool in their own ZIP codes, as well as the severity level they can expect for COVID-19 cases. How important is it for health systems to use this information to plan for their own regions? 

Vaughn Kauffman, PwC principal

Even though this pandemic is a global crisis, it is very much a regional issue, with a series of waves within a wave. The predictive nature of this modeling can help health systems in a region to see when those waves are projected to hit peak, with specificity, at a ZIP code level.

Whether it is a health plan or a health system, information about the broader family dynamics within a member’s home or neighborhood may not be well understood. We have created a synthetic population that can provide insights for healthcare organizations at the household level and, as such, help to better understand the risk of spread. Our datasets and models are complementary with what the traditional health systems have right now.

HRI: How can this model help health systems plan for the other side of their peak for COVID-19 cases?

Vaughn Kauffman: This type of capability really should be embedded into how health systems are thinking about the next six to 12 months. Until a vaccine becomes available, we’re going to live in an environment where this virus will still be around, and health systems are going to have to figure out different ways to reach out to patients, who will be hesitant in coming back.

The only certainty the virus has created is that there will be more uncertainty coming. We’ve been investing years in building these prospective analytics and machine learning models to help bring the future forward today. The COVID pandemic shows the need for this prospective understanding and need for health systems to use other data than they are typically using today—it’s not just about retrospective analysis based on claims or clinical data.

Health systems not only play a key role in treating the virus but they will also play an important role in helping to keep the virus suppressed in communities until a vaccine is available. An important way to understand how to suppress the spread of the virus is to have a broad understanding of the impacts of population density and other socioeconomic factors at a household level, such as comorbidities and other risk factors. For us, it’s really important that this type of modeling is part of health systems’ planning and preparation going forward. 

The only certainty the virus has created is that there will be more uncertainty coming. We’ve been investing years in building these prospective analytics and machine learning models to help bring the future forward today. The COVID pandemic shows the need for this prospective understanding and need for health systems to use other data than they are typically using today—it’s not just about retrospective analysis based on claims or clinical data.

HRI: This model gives healthcare systems a better understanding of their own local communities. Looking longer term, how can that insight be used by health providers to convince the community that it’s OK to come back to their facilities for care when the time is right?

Vaughn Kauffman: At the heart of our predictive modeling capabilities is this notion of understanding behaviors, and as health systems look to reconnect with patients, part of it will be about regaining their confidence. I still trust my healthcare provider, but my confidence may be wavering around whether I really want to go into that environment knowing that COVID-19 exposure may still be a risk. How health systems represent that safety factor is going to be important to gaining the confidence of the individual patient. Health systems are going to have to be more proactive in their consumer outreach, which is something they traditionally have not had to do.

HRI: Will it be important to use this information to help regions understand where they stand in the pandemic waves?

Vaughn Kauffman: The ripple effect of all this is not only about health systems themselves; everything around the value chain is impacted. How successful a hospital is at getting those patients back who need care will directly affect the manufacturing requirements downstream, how funds flow and what gets reimbursed. Hopefully these types of predictive insights can have a significant impact across the entire ecosystem to help better plan for this type of situation in the future. If health systems have better foresight, they can create better plans.

Read our research

Contact us

Vaughn Kauffman

Principal, Health Industries, PwC US

Follow us