How Non–COVID-19 Care Deferral May Reduce Employer Health Care Costs

In initial estimates from an actuarial analysis of self-funded employers by Willis Towers Watson, health care benefits were projected to potentially rise by as much as 7% this year from testing and treatment costs related to coronavirus disease 2019 (COVID-19). However, as non—COVID-19-related care continues to be deferred or cancelled, researchers report in an updated analysis that the pandemic could reduce employer health care costs by as much as 4%.

As indicated in a separate survey by Willis Towers Watson, 24% of employers expect no increase or decrease to their health care benefit costs, yet 64% note that COVID-19 will have a moderate to large impact on employee well-being. Notably, the eventual return of care being postponed may cause a subsequent wave of health care costs that can still result in an eventual increase of overall cost of care by 3%.

We spoke with Trevis Parson, chief actuary of Health & Benefits North America at Willis Towers Watson, on what implications the new study findings may have for employers and how the uncertainty surrounding the pandemic may still alter current projections.

Transcript

AJMC^®: Hello, I'm Matthew Gavidia. Today on MJH Life Sciences News Network, The American Journal of Managed Care^® is pleased to welcome Trevis Parson, chief actuary of Health & Benefits North America at Willis Towers Watson. Can you introduce yourself and tell us a little bit about your work?

Parson: Sure thing. Thanks for having me, Matt! My name is Trevis Parson, as you mentioned. I work for Willis Towers Watson. I've been with the company for going on 20 years now and have the privilege to work with some 300 health care actuaries within the company here in North America. I've been in the chief actuary role for the last 5 years and have had the opportunity to work on some pretty interesting assignments, not the least of which is trying to estimate the impact of COVID-19 on employer health plans.

AJMC^®: In the study conducted by Willis Towers Watson last month, projected findings indicated that employer costs could rise by as much as 7% due to COVID-19—related costs. Can you explain how much of an impact care deferral played in initial estimates and what other factors may impact costs for employers?

Parson: In our early study—we've since updated it—but in our early study, we were projecting an increase from 1% to about 7% to self-funded employer health care plan spending in 2020. At the time we did that, Matt, we were working with some of the early data, and we looked at showing scenarios basically reflecting 2 sensitive assumptions. One was the ultimate infection level, how many people would become infected and how sick would they get when they became infected. So we had basically low, medium, and high levels of morbidity [of] how sick people became. And then we assumed ultimate infection rates anywhere from 10% all the way up to 50%.

When we did that, we got the high of 7% and the low of 1% increase. We didn't assume a significant amount of care deferral at that time for that initial study, and so when we went back and updated, we updated for 2 things. One was, we weren't seeing the expected ultimate infection levels emerge that we originally thought might come to pass, and so we reduced our ultimate infection level. We showed results between say 1% and up to 20%, and we did that because we knew that the geographic impact of COVID-19 infections was very uneven. So, New York City, for example, has, as a hotspot, significantly more infections than say Little Rock, Arkansas.

So, we wanted to continue to show that wide range of infection to reflect that geographic unevenness. But we also saw, as data emerged, a significant volume of care deferral, and basically, care deferral as we've defined it is healthy patients, non—COVID-19 infected patients not going to the hospital, not going to [the] doctor's office because either the care isn't available to be provided because providers have been redirected or patients simply don't want to go to the system for care for fear that they will become sick if they do so. When we did that, we said, if care ... we again, similar to the morbidity assumption in our prior study, we did a low, medium, and high scenario for level of care deferral, and our new estimates are now showing anywhere from a 4% increase at the high ultimate infection level of 20% and low care deferral to about a negative 4% change for low infection rates and high care deferral.

When you add all that together, care deferral all on its own might be worth in that 5% to 6% range. If care were eliminated for 3 months this year, which it's not, but if it were eliminated, that would be a 25% reduction right there. If you assume, say half of it was reduced and not all of it, now you're down to about 12%, and if you assume half of that comes back before the end of the year, now you're down to about 6%. So, that was sort of our top-end estimate of care deferral. We did it a little bit more scientifically than that, but that's sort of the high-level math that helps get to the reasonableness of our answer.

AJMC^®: Can you discuss how the current analysis, “Beyond the Wave: Potential Impact of COVID-19 On Deferral of Healthcare,” was conducted? And what was the most surprising finding for you?

Parson: Well, we essentially did the ups and downs that I mentioned before, where we put the ultimate infection levels, how sick people became and how fast infection spread, and then made sure that we also did some of the downs, which were also supply limits. You can't necessarily supply all the care that may be demanded in a time when everybody's infected with a virus, right? There are only so many beds, for example. We also know, we've talked about care deferral before, so our study included all of those variables, and then some. Notably, for any particular population, as I mentioned to you a minute ago, where employees happen to be located might significantly impact the cost to your plan. If everybody's in New York City, as opposed to Bismarck, you might have a different answer. So that was essentially the methodological underpinnings of our analysis, and we mirrored that in the Beyond the Wave component where we tested lower ultimate infection levels and then a higher level of care deferral.

The most surprising result really has been the variability. Actuary would prefer to put a far tighter range around their estimates of costs, and in normal times, you can do that. But in times like this, where you have very, very little history—and no history that really compares to a situation like this upon which to study and draw conclusion—you get forced to putting wide ranges around things. So, we've ended up with an answer here, in our second paper, basically saying it might be up a little, it might be down a little, and depending upon where you happen to be and the nature of your demographic profile, you could find yourself on either side of the line. So, it's not where we typically like to draw answers, but that's where we've had to land given the massive amount of uncertainty related to the impact of COVID-19.

AJMC^®: As medical care for non—COVID-19 infected patients has declined during the pandemic, do you see a potential increase in use as social distancing measures ease nationwide?

Parson: Well, whether distancing measures ease or not, there is a certain amount of care in this initial way that simply hasn't hit the system. As you and I talked a moment ago, there is a significant volume of care, maybe 30%, 40%, 50% here in these most recent months, say March, April, maybe May here, that isn't hitting the system. Some of that care is going to return at some point in time, not all of it. For example, if you think about a dental cleaning, you might go get your teeth cleaned every 6 months. If you miss one of those, you're not going to go twice the next time; you'll just skip that one and move on. There will be some care like that that simply will get skipped and never return.

There will be other types of care where if you skip it now, you might be able to get by for a little while. Maybe it's a joint replacement, for example. You might be able to continue to limp along on that knee, but eventually, you're going to need to get that knee replaced and so that care is going to come back. There are other types of care like that that one of my clinical friends could opine further on, but certain cancer treatments, for example, maybe you postpone a cancer treatment this time—that's going to have to come back at a future date.

So, you'll want to be able to effectively monitor that and make sure that the ultimate level of care that returns is known and baked into your budget as best you can. It will vary depending upon the type of care, and so it'll be good for employers to understand what type of utilization patterns their populations are exhibiting, particularly self-funded employers, so that they can keep a good eye on that and understand, "Hey, if this care comes back, how much do we think that's going to be worth?" and be able to say, "If at the time it comes back, we're in an environment of perhaps a subsequent wave, how will the system be able to handle all that?" So, there are lots of unknowns, which simply require employers to pay really careful attention.

AJMC^®: Has the transition to telehealth assisted in reducing the potential reduction in employer health care costs?

Parson: Well, we've certainly seen a significant uptick in the volume of telehealth in the middle of this initial wave. Some estimates have been 20 times higher than previous levels, and I think that there are some advantages to that, right, to the extent that it can be more efficient, care can be delivered more timely—you don't have to wait less to get care. And to the extent that it happens to be a little bit cheaper, I think employers would be all too happy to provide that type of experience to their employees. Ideally, this is one of the things, Matt, that becomes perhaps permanent post COVID-19. Maybe it's one of the very few good things to come with this, right, if people learn to engage our health care system in a better way, as I said before. And maybe there are fewer emergency room visits, like, "Hey, maybe I didn't need to go to the emergency room for that, and I'm not going now because I'm afraid to go there for getting sick. Maybe I'll change my mind as to how I engage the emergency department going forward." Similarly, with telehealth, maybe it's a matter of just clicking a couple of clicks on the Web or otherwise picking up the phone to get your care. Ideally, that would be something that we carry through on into the future post COVID-19.

AJMC^®: As indicated in the report, even at high infection rates and low care deferral, costs will increase far less than what was previously anticipated. What serves as the largest takeaway for employers from these findings?

Parson: Well, it's still very uncertain, and data is emerging. As I mentioned a moment ago, we don't have strong stable data upon which to base these predictions, and so, as I mentioned, we'd love to put a tighter range around our results. We had a pretty wide range the first time, 1% to 7%, and now we're 4% to —4%, and so we've even had to accommodate our own new information to improve our estimates. Ideally, we continue to do that going forward. Employers are definitely looking for answers, and we want to be there to help them find those answers. It’s just very uncertain.

So, what that requires, I think, employers to do is to be ready, and you have to have an effective measurement plan in place, and you have to feed that measurement plan with information that's current as possible and have some definite conclusions from what you would draw from that data to drive informed decision making. That's what I think employers should do throughout this period. There'll be little that they can do immediately, from a health care plan perspective, in most cases or that they'll want to do immediately, but an effective measurement plan that will inform them as to how they might best make decisions moving forward.

Maybe, again, another positive of this pandemic is that employers take the opportunity to really examine their health care plan in light of the future, which might see greater consolidation amongst providers, different utilization patterns amongst their employees. How will benefits need to look in an environment like that? Employers might have some good opportunities to make some effective changes down the road as a result of all of this, but they won't know which changes to make without an effective measurement.

AJMC^®: What other COVID-19—related research is under consideration for Willis Towers Watson?

Parson: Well, that's a great question. So, we haven't stopped. We've now released our second paper, as you've mentioned, and the next place we want to go really is further into the future if we can. I've already indicated my dislike for the range of estimates that we've provided so far in terms of its size, but we need to continue to try and look down the road to see what the impact of this will be for employers, and one of the things we want to look at more strongly is 2021.

Employers will be looking to set budgets, designs, and contributions for employees for the 2021 calendar year, and they'll be looking to us for answers. So, what we want to be able to do is develop some potential scenarios as to what 2021 might look like. We want to be able to quantify those, put them into context, and work with employers to see, based on their particular situation, which scenario might fit them best and then customize that to their particular profile to get a better handle on what 2021 looks like.

As I mentioned a second ago, Matt, there's still a significant amount of uncertainty here. I don't like the fact that we don't have as much data as I would normally have in order to help employers solve these problems, but we're working very hard to try and think of what the possibilities might be, take the information that we do know, put it to best use, and take what we know from the employer—their demographic profile, where they happen to be, their benefit level, etc—and channel that into a custom piece of work that helps them make good informed decisions as to what their benefits look like going forward post COVID-19.

AJMC^®: Thanks, Trevis!

Parson: Thank you, Matt! It was a pleasure.

AJMC^®: To learn more, visit our website at ajmc.com. I'm Matthew Gavidia. Thanks for joining us.