– With that I’m gonna introduce our speaker for today, Dr. Jeff Colgrin. He is an assistant professor here at the University of
Michigan in Internal Medicine and at the Center for Clinical Management Research at the VA. And he is a former clinical scholar, not here at University of Michigan but from the University of Pennsylvania’s RWJ Clinical Scholars program. And he is an expert in
behavioral economics as it’s applied to medical interventions. And he uses the principles
of behavioral economics to develop scalable interventions to help patients and clinicians
make better decisions. But he’s gonna talk to us today about a sort of a new area of his research that’s focusing on price transparency and how that impacts
patient decision making. So I’m looking forward to learning from it and will let you take it from there. – Okay.
(paper shuffling) Alright, well thank you, Caroline and thank you all for being here today. And so I’m thrilled to
be able to talk to you about helping patients and
how deductible health plans get the care they need at
a price they can afford. So I’ve got three objectives
for the talk today and I will try to save about 10 minutes at the end for questions and then, you know, happy to continue to have dialogue with you
afterwards if you’re able to stay. The first objective is
to examine the growth of enrollment in
high-deductible health plans, help you better understand
what that looks like currently across the U.S., understand some of the motivations and some of the factors driving
the growth of enrollment in high-deductible health plans. We’ll then spend a little bit a time setting the stage reviewing what we know about the challenges that are faced by patients in high-deductible plans. We know a lot at this point about what patients in those plans face and some of the problems that they create that I’m trying to address in my work. And then I’ll close by
spending most of the time talking about strategies
that we might be able to use to help patients in high-deductible plans get the care they need at
a price they can afford. So how many of you would say you are very familiar with
high-deductible health plans? I’d say, some hands. How many of you have heard
of high-deductible plans? So let’s go with a lower bar. Okay, so most of you have
at least heard of it before. I just want to provide a
level set at the get-go in terms of helping you understand what those plans are and
what we’re talking about. So high-deductible plans are
private health insurance plans that have large deductibles. Now what a large deductible
would be for one person may not be the same thing as a large deductible for somewhere else. That line is actually drawn by
the Internal Revenue Service, believe it or not. And where that sets at currently is that if one has individual coverage, so you’re the only person in the plan, the threshold for that
deductible is $1,350. Now, if you’re in a family
plan, that threshold is $2,700. So kinda weird numbers. High-deductible plans started in 2003. They were actually created as part of the Medicare Modernization Act. That is the same piece
of federal legislation that created the Medicare Part D plan, the Medicare Advantage program, also created something called
health savings accounts at that time. And so it wasn’t the legislation creating high-deductible plans, it was the legislation creating
health savings accounts, or tax-advantaged savings accounts that one can only have when you also have a private health insurance plan that is a high-deductible plan. So it was the HSAs starting then and then in order for you
to be able to get an HSA or health savings account you had to have a high-deductible plan. So just rewinding to the political context of when that started, this was 2003 so this was during the
George W. Bush administration and this was part of
the Ownership Society. It was a big political
initiative at that time. And it was the idea that more people should have health savings accounts; They should be able to
have tax advantage means for saving for their healthcare. But at the same time, there were policymakers that were trying to encourage patients to act more like consumers in healthcare and try to be more judicious in their use of the healthcare system and more cost conscious
in their decision making. So they kind of paired the two together and that’s how you ended up with high-deductible health plans. Now those plans are sometimes now linked with a health savings
account but not always. So those plans when they are offered and not required to be linked with an HSA, but in order for one to have an HSA you need to have that kind of plan. So, these are data from the
National Health Interview Survey which probably provides
the best national data on the rates of enrollment in
high-deductible health plans as a percentage of people
with private health insurance. And going back to 2010, so
again, this is seven years after this federal
legislation created HSAs, you can see that at that point in time 25% of people, these
are non-elderly people with private health
insurance is the denominator. 25% of those people at that point in time were enrolled in a high-deductible plan. And you can see here, if you look at the upper
part in the legend there, most people at that time and
this is still true today, who have a high-deductible health plan do not have a health savings account even though that kind of plan is eligible to be linked to an HSA. Fast-forward to last year,
that rate has now grown to 43%. Okay, 44% actually. So out of all non-elderly people with private health insurance now 44% of them are enrolled in a
high-deductible health plan. So you’ve seen a two to three percentage absolute increase in rates of enrollment of high-deductible plans as a proportion of private health
insurance year over year. If you project that out into the future by 2020 or maybe a little bit thereafter most people who have
private health insurance will be enrolled in a
high-deductible plan. Okay, fast forward even
further beyond that this will far and away
be the dominant kind of private health insurance
going far into the future. As I’ll illustrate in a minute there’s not a lot on the landscape that would suggest that’s
going to change anytime soon. So as I talk about all
of this going forward I want you to think about this as something that will be relevant not only now but far into the future. Just to drill down on
that a little bit more to help you understand
some of the dynamics that led to where we’re
at in rates of enrollment in high-deductible plans, these are data from the
Kaiser Family Foundation and the Health Research
and Educational Trust. These are data on
employer-sponsored plans. So the private health insurance market is comprised of employer-sponsored plans, exchange plans and then private plans purchased off the exchanges essentially. So these are looking just
at employer-sponsored plans. And what you see here is the proportion of employer-sponsored plans that have any deductible at all. So there was a point in time when most private health insurance plans did not have deductibles; There
was first dollar coverage. You didn’t have anything
you had to pay out of pocket for your healthcare before
the coverage kicked in. If you go back to 2006 so three years after HSAs were created, just over half of those
plans had a deductible. You fast-forward to 2016, this is now 83% of private
health insurance plans have any deductible at all. If we look then at those
deductible amounts, so this is only looking at those plans that have a deductible and
again, that’s 83% of plans now you can see back in 2006 the average deductible amount was $584, out of all those plans with a deductible. In 2016, that had grown to almost $1,500; This is the average deductible amount among the vast majority of
private health insurance plans that have a deductible. So we’ve seen dramatic growth in just having any deductible
with your private plan. When you do have a deductible we’ve seen dramatic growth of
the size of that deductible. Those are employer-sponsored plans. Of course, for private insurance plans that are outside of the
employer-sponsored market there’s large deductibles as well. And of course, there’s
been a lot that’s been written about this in the
media especially when the, after the exchanges had been created as part of the Affordable Care Act. What you see here are
the different metal tiers and the average deductible amounts of private insurance plans in each of those different metal tiers. And of course, plans purchased
through the exchange, the vast majority of them are either bronze or silver plans. These are the average
deductibles in those plans. If you look at the gold or platinum plans which are enrolled in at much lower rates than the bronze and silver plans the deductible levels are much lower. Okay, so the vast majority of plans that people are acquiring through both federal and state exchanges are indeed high-deductible plans. Of course, there are some people for whom there are cost-sharing subsidies but that’s only a very narrow
segment of the population that’s eligible for those subsidies. So people that are purchasing
plans through the exchanges the vast majority of them are
subject to high deductibles. Okay, so hopefully I’ve convinced you that this is a really common issue and that there are many people that are facing high
levels of cost sharing in their private health insurance plans. What’s behind this? Well, I think there’s two main drivers of the growth of enrollment
in high-deductible plans. The first you can see here
illustrated on this slide. So again, this is data from Kaiser and from Health Research
and Educational Trust. This is looking at
employer-sponsored plans. And one strategy that
employers have been using to deal with the growth of high
healthcare costs, of course, is that often they will pass that along in the form of premiums. Increasingly, what they’ve been
doing to constrain premiums is: Instead, push that
over as deductibles. Okay, so instead of
increasing the premiums, have been increasing the
deductibles in those plans. So, what you can see here on the lines at the
bottom part of the slide, this is from 2011 to 2016, there’s been a 19% increase
in the cost of premiums over that six year time period. You can see that as higher than the rate of increase in workers’ earnings, higher than the rate of inflation but much lower than the
growth of deductibles. So the growth of deductibles over that same time period has been 63% whereas the growth in
premiums has only been 19%. Again, both of those much higher than the growth in workers’
earnings and inflation. But far and away deductibles are growing at a much faster rate than our premiums. And I think one of the other main driving factors
between the growth of, behind the growth of enrollment
in high-deductible plans has been what has been termed: The skin in the game argument. So this is just the belief or the way of looking at the world that when people are faced with
high levels of cost sharing they have more skin in the game and they will be more
judicious users of healthcare, they will be more cost
conscious healthcare consumers, they will use healthcare more
wisely and more carefully and that they might also even
take better care of themselves because they don’t want to, in the future, accrue
costly health conditions. That is how that line of argument goes. Whether you actually believe that or not is a little bit of a, I’ll be honest, a political Rorschach test. But be that as it may,
that is one of the drivers. And certainly if you look
at a more practical level behind consultants that
have been going to employers and have been suggesting that employers instead of
offering their more generous plans move over to only in some cases offering a high-deductible plan. General Motors, for example,
did this with their workers and their salaried
workers and their retirees many years ago now, actually where they moved all of them into a high-deductible health plan only. That’s called a total
replacement approach. You’ve seen a lot of
employers do that now. Some of it, again, is a way to deal with the growth of healthcare costs and not having to increase
premiums too, too much. But also, again, it’s just belief, that people will be better
healthcare consumers when they’re exposed to those
high levels of cost sharing. So I think those are the two main factors that have driven the growth
of enrollment in these plans. So how then do these
plans affect patients? And that’s really the
critical issue I think, and what I’ve been exploring in my work. This is an area where
what’s old is new again. So if we go back to the RAND
Health Insurance Experiment, it was conducted in the ’70s and ’80s, remains, of course, one of
the most important studies and probably the most important
at least randomized trial in the history of health policy research. It has some really relevant insights in terms of thinking about the effects of high cost sharing on patients’ access to care
and on their decision making. So just briefly, many of you have heard of the RAND Health Insurance Experiment. Essentially randomized people to different kinds of
health insurance plans and then followed them
for a number of years into the future. There were several different
levels of cost sharing and then there was a free care plan. One of the policy
questions at that time was: Well, what happens when we
give people free healthcare? You don’t see that as a policy relevant
question as much these days but that was something
that was on the table, of course back then. If we flip that around now and we say: Well, what about the people with high levels of cost sharing, how does that compare to people who didn’t have cost sharing or had lower levels of of cost sharing? And I think the results from the RAND Health Insurance Experiment remain very, very relevant to this day. Now, what you see here in this figure are different kinds of healthcare, healthcare that is very, very effective and healthcare that is
much less effective. I think in 2018 we would call
that high-value healthcare. Or we would call that
low-value healthcare. And one of the main findings from the RAND Health Insurance Experiment was that when people faced cost sharing, so if you lumped together all the plans where people had cost sharing, compare that to the
people who had free care, people who had cost sharing reduced their use of less effective care. That’s a good thing. People avoided getting care they probably shouldn’t have been
getting in the first place. Okay, the problem, of
course, is at the same time people also went without getting care that was highly effective,
care for acute conditions, care for chronic conditions,
care that was gonna help people that they probably should be getting. They got less of that as well. And in fact, they got
less less effective care and less highly effective care
at relatively equal rates. So in other words, people
decreased their use of everything across the board whether or not that care was less effective or highly effective. Fast forward now to the 20 to 30 years since the RAND Health Insurance Experiment and looking at more contemporary plans we see a lot of those findings
echoed in study after study. First of all, we’ve learned
from the research in this area that it is, of course,
difficult for patients to differentiate high-value services from low-value services. One challenge in the whole consumer directed healthcare movement and when people are talking about consumer-directed healthcare they’re talking about
high cost-sharing care with a health savings account typically. We know that we as providers
and I’m an internist and us as providers, part of our job is to be an agent for our patients and help them deal with
asymmetric information. If you’re putting more
of that responsibility on the side of the patient, of course, it’s not surprising that
they’re gonna have difficulty sorting out what kind of healthcare they should and should not get. But we have a number of studies now saying in these current plans
that’s a major issue. We know that people reduce
needed healthcare due to cost, again, hearkening back to the findings of the RAND Health Insurance Experiment. We know that people in
high-deductible plans decrease use of
high-priority office visits, long-term medications
for chronic conditions, and even use of clinical
preventive services that don’t have cost sharing. It is complex for people
to use these plans and understand: What is my deductible? And: Am I at or below it? Some of these plans exempt
some services from deductibles. In many cases, people go without clinical preventive services even when in this day and age evidence based clinical preventive services
are exempt from deductibles people will often still
go without those services because of the perceived cost of them. Of course there are greater
adverse financial outcomes that people face in these plans in terms of healthcare debt, ability affording healthcare
bills and the like. We’ll talk more about that a little later. These plans do, that
being said, lead to lower at least in the short
term, healthcare spending. People decrease their use of healthcare, it turns out that they’re less costly. Again, this is what many employers are attracted to in terms
of offering these plans. And in fact, when we
look cross-sectionally, people in high-deductible
plans have healthier behaviors. But it looks like that is
almost entirely due to selection or people who are otherwise healthy and engaged in healthier
behaviors choosing those plans as opposed to the plans
encouraging them to be healthy. We also know at the same time that there are particularly
vulnerable populations that are increasingly
enrolling in these plans. One population that we
looked at in a study we did several years ago, back actually when I
was a resident in Boston and this was in partnership with Harvard Pilgrim Health Care, we looked at the experience of lower income families
in high-deductible plans. And we defined lower income as being less than 300% of the
federal poverty level. This was at a time and in a place where there was state
based universal coverage that was trying to be achieved through the Connector in Massachusetts. And they drew the line for lower income as being at 300% of the
federal poverty level, in terms of that’s when
people were eligible for subsidies to help them
defray high healthcare costs. And what we found is that
families with lower incomes delayed or went without needed
care at much higher rates than did families with
higher income levels. And this was one of the first studies to look at the experience of lower income families in these plans and how that relates to them
going without needed care. And we looked both at care that
was delayed or went without for both children and for adults and we found that families
were going without both of those at higher rates
when they had lower incomes. We’ve also looked more recently at the experience of people
with chronic conditions. So you could imagine people
that have high healthcare needs, the need for more healthcare, that they would be more vulnerable to the high cost sharing of these plans. It’s important to note and Mark
Fendrick is not here today; He’s probably out working on
this very problem, of course talking to folks at the
Internal Revenue Service that has been really engaged. And some of you have been
involved with this I know as well, engaged with allowing health insurance, plans to still be linked
to a health savings account but exempting evidence-based treatment for chronic conditions from deductibles. So introducing value-based
insurance design into high deductible plans. Currently, it turns out
that your medications for diabetes and
hypertension and all the rest are subject to cost
sharing in these plans, that plans cannot legally exempt those things from deductibles and still be paired with
a health savings account. So what we found then is that when people have high deductibles and they have no chronic conditions, one or more chronic conditions, or two or more chronic conditions, that they face much higher rates of what has been termed
by the Commonwealth Fund, high medical cost burden. So one of the arguments
that these plans are you know, they have high deductibles but they have a lot lower premiums. So for many people it may be a better deal if we look at what people have
to pay for their healthcare both in terms of the
premiums plus the deductibles and add those together, that’s how most people think about their healthcare spending not just the cost sharing
separate from the premiums. If we lump those two together we see that when people
have high deductibles, that when that amount they’re
spending on healthcare exceeds 10% of their health,
their household income, they experience those
high medical cost burdens at much greater rates than
people with lower deductibles or no deductibles at all. And this was work with Joel Segel when he was a doctoral student here and is now on faculty at Penn
State as a health economist. So how could we solve these problems? And that’s what the work
that I’ll be talking about for the rest of the time
has been addressing. If we step back and think about: Okay, how could we deal
with these problems? Well, naturally when
people have to pay for a lot of their healthcare
out of their own pockets, it’d be nice if we could just
make healthcare cheaper right? I think we’re almost there; We’re so close to just making
healthcare cheaper for people. But, you know, theoretically,
that could be a solution. That is a longer term enterprise
and a lot of people in IHPI of course are looking
at this very problem. But we’re still far from being
able to realize that promise. Of course, we could make
health insurance more generous. At the same time, I’ve showed you trends going in the exact opposite direction. And there’s nothing on
the policy landscape to suggest we’re going
to walk back from that. In fact, if it ever
actually gets implemented the Cadillac tax on employer
sponsored insurance plans will make employer sponsored
plans less generous. It keeps getting pushed
back further and further. And the House I think is
poised maybe to do that again. But at some point in time they probably will have
to actually implement that part of the Affordable Care Act. And so it is unlikely that insurance will get more generous in the near term. Although again, you of course start to see more conversations about
single-payer health insurance. But again I think most of us would agree that we are still very far from that being a reality as well. So cost sharing will continue to grow at least in the short term. We could, of course, make
cost sharing smarter. I mentioned some of Mark’s
work and the work of others trying to implement more
value-based insurance design. Reference based pricing is another tool where health plans will say, okay we’re gonna cover this service but we will only pay so much
for that particular service; Let’s pair that with price
transparency information and then try to help people
better afford needable care and select lower cost providers. There’s been some wins I
think, in the short term in terms of implementing
that kind of approach. But in terms of that being
a panacea to these problems, I don’t think we’re quite there yet. Of course tiered networks as well. Let’s have different tiers
of networks of providers depending on the value,
the care that they provide. You know, I think all of these things another challenge, of course,
getting back to consumers being able to understand their plans makes healthcare and cost
sharing much more complicated. So that’s a constant
push and pull going on. If we try to make cost sharing less blunt it will get more confusing for people. And I’ll illustrate in a little bit health insurance literacy and how that relates to people being able to use these kinds of plans well. That could potentially
be more of a problem if we make cost sharing more complex. So what I’ve been focusing on is helping people better use
high-deductible plans. So being a bit of a pragmatist and thinking about we are where we are. You can argue about whether this is good or bad where we’re at, depending on where you look at these, the lens through which
you look at these issues. Let’s take it where it is. And let’s figure out ways
that we can help people better use high-deductible plans to get the care that they need
at a more affordable price. So what would that mean for people to be able to better use
high-deductible plans? Well, you know, how many of you just, I’m not gonna call on
anybody in particular. How many of you have a
high-deductible plan? One, two, okay. So University of Michigan, so most people are in
Premier Care, of course, which is not a high-deductible plan. Some people may have VA
coverage and the like where you can get high-deductible or consumer-directed health plans. These plans, imagine, if you’re not in one having a three, four, $5,000 deductible. Just imagine that for a moment ’cause many of us are not currently in that place with our health insurance. That would require you potentially to think about your
healthcare differently. Especially if you have
a chronic condition. If you don’t have one,
imagine that you do. Or if you’re a provider,
think about your own patients. Imagine you have diabetes and there’s a certain set of services that you’re gonna need every year no matter whether your
diabetes is well controlled or poorly controlled. Hemoglobin, A1C a couple times
a year, urine microalbumin, gonna need medications,
blood pressure monitoring, an eye examine, a test to look for loss of sensation in your feet and probably some other
services too, a statin. These are all things that
high-deductible plans do not cover underneath a deductible. People face the cost
sharing for those services in high-deductible plans, okay? So again, we as providers, as patients, when you have chronic conditions if providers and patients
could work together to say: Well, I’ve got these chronic conditions, what kind of care am I gonna need for that condition over the next year? That way perhaps, I could
use price information that is becoming increasingly available to estimate the cost of those services and be able to save for that
if I’m financially able to. That would be one skill
that is often encouraged among people with high-deductible plans. Could of course use price information that is becoming increasingly available. Health insurance plan,
there’s not great data on this of course because there’s not, people don’t track well
what private health plans are offering in terms of
the bells and whistles that are attached to those plans. But increasingly, at least anecdotally, more health plans are offering price comparison tools to
members in those plans. There are of course state-based efforts, regional efforts that are trying to, especially through all
payer claims databases, trying to communicate
to patients or consumers the prices that they might
face at different facilities. That is rapidly increasing in terms of the frequency and the spread of that. Patients could potentially
use that information to choose where and when
they will get healthcare. They also could talk to their clinicians about their cost sharing challenges and try to make decisions about what kind of healthcare services, what to get and what services to forgo with consideration of out
a pocket costs in mind. This has been recently
framed in the literature by Peter Ubel and others as the financial toxicity of healthcare. In many cases, if we are
ordering low-value services for patients who are going
to have to pay for them we’re probably doing them a disservice. And it may be the case that
if we have conversations about whether this service
makes sense for you let’s also take into account
how much it may cost. That potentially, if
there were more and more conversations like that happening that could lead to better outcomes not only medically but
financially for patients. And then there’re actually reports in the media, believe it or not, of patients trying to
negotiate prices for services. This actually take advantage, somewhat, of the smoke and mirrors
of healthcare prices. And the fact that often prices
that have been negotiated between healthcare systems and plans are based on historically negotiated rates that are increased over time and then there’s a lot of
cost shifting that goes on especially around bad
debt and things like that. And there are certainly plenty
of media reports out there of people going to healthcare
systems and saying: Gee, I could get this service
somewhere else cheaper. Would you be able to honor this price? And saying: Yeah, sure, that’s fine. Because increasingly, as well, healthcare systems are faced with bad debt from people facing high deductibles and then not paying their bills, right? So if potentially the healthcare system could accept a lower rate but actually get something
instead of nothing you could imagine many
healthcare systems potentially being open to those kinds
of inquiries from patients. So this idea, then, of helping people better use their plans I think raises some really
fundamental policy questions. First of all, how often are
people doing these things? I will say when I started
to talk about this a few years ago, I got more
puzzled looks from audiences about well, negotiating prices, using price information,
that’s kinda weird. I think this has gotten
woven more and more in the fabric of, you know, media coverage about healthcare policy as
well as people’s experience in private health insurance plans. But it still begs the question: How often are people
actually doing these things? Also, what kinds of patients
are doing these things? Is it simply the people that often immediately come to mind those who are more engaged
in their healthcare, maybe higher levels of education, higher socioeconomic status, people who are more willing and able to engage in these kinds of behaviors? Might be the ones who are more willing to engage in those behaviors. We’ll talk abut that in a second. When are people doing it? For what kind of healthcare services? And then also, how
helpful are these things? Okay, so it may be the case that people may be using price information maybe getting helped a lot or maybe not. We actually don’t know
a lot about how helpful patients find these kinds of strategies. And then if we’re thinking
about implementing these kinds of strategies more broadly or better supporting patients
in high-deductible plans it’s important to know
what kinds of things facilitate and impede their
engagement in these strategies. So instead of doing what
health plans typically do now is simple creating
price-transparency tools and pushing those out to plan members and then bemoaning the fact that people do not use those price-comparison tools maybe health plans could think differently about engaging consumers
and better support them to use their plans to get care they need at a price they can afford. So the way that we’ve been
exploring these policy questions most recently then was
through an Internet survey that we conducted two years ago in partnership with GfK
and their knowledge panel. So this was a survey of 1,637 GfK Knowledge Panel participants who’d been in a high-deductible
plan for at least a year. GfK’s Knowledge Panel used to
ne called Knowledge Networks. For those of you that are
more familiar with that term, this is the panel that
is currently used now for the National Poll on Healthy Aging conducted by IHPI in
partnership with AARP. For those of you that are not familiar, a nationally representative
Internet survey panel of 55,000 U.S. adults. They can, I think, pretty reasonably claim national representativeness relative to other Internet survey panels because instead of creating a website and having people opt-in to that they do address-based sampling. So what they do is they
look at people’s addresses, they do probability-based
sampling based on that, send people letters in the mail, say: You’ve been selected
to join this panel. They then reach out to
them over telephone. If people do not have a
computer, they give them one. If they do not have Internet access, they give that to them to. They also then pay
people to do the surveys. Very different kind of strategy than if you create a website
and had people opt-in. We administered the survey
again two years ago. We over-sampled people
with chronic conditions because we were interested
in whether people are engaging in these
behaviors at higher rates when they have chronic conditions. Spoiler alert, they’re not. I don’t even have that on the slide because the rates were
exactly the same based on whether or not people
had a chronic condition. And then the response rate
for the survey was 55%. So in our survey we measured engagement in these behaviors in the last year. When people had engaged
in one of those behaviors we asked them: What they did it for, what kind of healthcare
service it was for, how much they felt it
helped them and in what way? When people hadn’t, we
also asked them about what kinds of things facilitated or supported that engagement, what were things that helped you be able to engage in that strategy? When people had not
engaged in those behaviors we asked them: Why not? ‘Cause again that’s a
really critically important issue or unknown in this particular area. Because a lot of these were new areas being explored in survey
research for the first time we had to do a lot of iterative testing in our survey developments. We did seventeen cognitive interviews with people in high-deductible
plans from all over America. So actually it was, I think
maybe five or six rounds of two to three people every time. We spent a very long time ensuring that we got to a
reasonable level of validity in our survey questions
and kind of inteligibility. And that the responses
adequately reflected people’s experiences in these plans. So hopefully that gives
you some confidence that what we’re measuring here has a reasonable internal validity. So these were the
estimated characteristics of the population that we were serving. You can see here the age breakdown. Again, these are non-elderly Americans in high-deductible plans. So people with private health
insurance, non elderly. You can see the rates
of chronic condition, fair or poor health, the income breakdown. Then most people had
employer sponsor coverage and many fewer had coverage
through a marketplace. When we look at data from the National Health Interview Survey probably the most long-standing data on rates of enrollment in these plans our estimates of the
population characteristics of people in high-deductible plans looks very similar to what NHIS has found. Okay, so how often are
people in these plans actually engaging in these behaviors? Well, we found that the most common thing that people were doing in terms
of these consumer behaviors was saving for healthcare they knew they were
gonna need in the future. Which 40% of Americans in these plans have done in the last year. So this is not saving just generally, this is not saving for
healthcare generally. This is saving for healthcare services you knew you were gonna need in the future is how we framed that question. 1/4 of people had discussed the cost of a service with a provider. 14% had compared prices for a service in their decision making. And 7% had tried to negotiate
a price for a service. We ended up framing that question based on cognitive interviewing, based on, you know,
irrespective of whether that was before you actually got the service or after you got it and got the bill and then negotiated with
the healthcare system. We found there was a lot
of the latter going on in our cognitive interviews. And so we wanted to
ensure that we, you know, set a, cast a pretty wide net for attempts to negotiate with providers. So which patients then
in high-deductible plans are engaging in consumer behaviors? So what I’m gonna show over
the next several slides is going to be different
characteristics of people, okay? And then the adjusted
prevalence of engagement in each of those four behaviors by, separated out by
different characteristics. And then these are marginal estimates from logistic regression models. So this is adjusting for a
whole host of other factors that could confound those relationships. So what we’re looking at here first then is that people with
higher levels of education were more likely to have
saved for healthcare services they knew they were
gonna need in the future relative to people with
lower levels of education. We also found that people with the highest levels of education,
with a graduate degree were more likely to have discussed healthcare costs with their providers. But it’s worth noting, you’ll see this again and again that even for people with the
highest levels of education, most people in these plans are not engaging in these behaviors. It’s a very, very important point. We also in our survey, we used a scale from the marketing literature. We were very interested
in looking at whether people who are more likely
to be bargain shoppers are more likely to engage in these consumer healthcare behaviors. So this is, again, a scale
from the marketing literature and then we separated people
into different tertiles by virtue of where they
stood on that scale. And what we found is that, indeed, people who are more likely
to be bargain shoppers are more likely to also
compare healthcare prices and talk to their doctors about cost. But again, even the people who are, you know, the coupon clippers and who are always looking for deals do not engage in these consumer behaviors at very high rates in these plans. Another really important
point and why I earlier really hit on the issue of most people in high-deductible plans do not have a health savings account. We did find that having
a health savings account was associated with a much higher adjusted prevalence of having saved
for healthcare services people knew they were
gonna need in the future. It’s a really important point
especially as it relates to potential policy interventions
around these issues. We also wanted to understand other factors that could potentially mediate engagement in these behaviors. People being confident in their ability, take into account all that would you know, come along with being confident, having the resources, the tools, the knowledge, so on and so forth, the timing, being able to do it. We looked at people’s level of confidence in being able to engage in those behaviors and how that relates to
their actual engagement. And we found that people who had high levels of confidence in
being able to save for care or discuss costs with a
provider and negotiate prices were, indeed, more likely to do so. But again, even people with a very high, highest levels of confidence most of them did not engage
in one of these behaviors. And then the final thing I’ll show you relates to health insurance
literacy and financial literacy. And we found that people with higher levels of health insurance literacy were more likely to have
saved for healthcare services they knew they would need in the future. But they were not more likely to engage in any of the other consumer
behaviors that we looked at. Another concept or construct
that employers especially have been increasingly interested
in is financial literacy. Especially as it relates to
things like retirement saving. We thought a lot about that as a construct and how that’s distinctly different from things like health
insurance literacy, numeracy, people being able to
understand and use well money is a really important construct or skill in terms of being able to use
these kinds of plans well. And, indeed, we found
people with higher levels of financial literacy were more likely to have compared prices for a service in their decision making. But again, even people
with the highest level of financial literacy, most of them did not engage
in one of these behaviors. Okay, so that looks at, we looked at the rates of
engagement in these behaviors, we looked at the factors
associated with that engagement. Now we’ll look at what kinds of services people are engaging in
these behaviors for, how much they felt like it helped them. What kinds of things supported
them in being able to do so and what sort of things
impeded or prevented people from engaging in those
behaviors when they did not which, of course, is a
much more common scenario. And we’ll go through these one by one. So first looking at saving for healthcare. So we actually did not ask in our survey about the kinds of services for which people saved for healthcare. We wanted to do so and we did that in cognitive interviewing. It completely fell flat. People did not understand it and they did not think about
healthcare saving in that way. They thought about generally
saving for healthcare I knew I would need in the future and a general bucket of services. Very hard to pin them down in a survey in terms of the individual
services they saved for. So these were the perceived
effects of saving. So just over half of people
felt like it helped them get needed care and have
less debt from healthcare. And a much smaller proportion of people said it helped them free
up money for other things. So the fact they’d set
aside money for healthcare then they ended up needing
healthcare, they had that on hand that then freed up their ability to use money for other things. What I’m showing you on these slides are just the three most common things that people told us in
each of these areas. In terms of the things
that facilitated saving, employer contributions
to healthcare savings was the most common thing that helped people be able to save. People’s health plans,
presumably communications from their health plans and the like also was a common facilitator. So, what stood in the way of people being able
to save for healthcare? When they did not save
for healthcare services we asked them: Why not? And the most common thing that we found why people didn’t save is they said: Well, I didn’t consider doing so. Another spoiler alert, this
is the most common reason from people not engaging
in these behaviors across the board for each one of them. People not thinking of doing it in planning ahead for their healthcare. Of course, the one response
that immediately comes to mind especially for people with
lower socioeconomic status, not being able to afford to. That was a common reason but
it was not the most common one. And then, of course, people saying: Well, it wouldn’t’ve changed my decision. I think this, again, speaks
a little bit to the fact in which people are using these plans. Even if I had set aside money for my care I would’ve gotten what
I was gonna get anyways. So again, maybe planning
a little bit of a seed in terms of thinking about helping people use these plans differently when their cost sharing requires them to. Alright, so that was saving for care. Discussing cost with a provider
is what we’ll turn to next. So what are people talking
with their doctors about when they’re talking about cost? The most common thing is a
cost that is on the mind, of course, of a lot of Americans especially those with chronic conditions and low incomes and that’s
prescription medications and the cost of those. The next most common thing
was outpatient visits followed by procedures. So, those were the top three. In terms of the perceived effects of having talked with
one’s doctor about costs, just under half of people said it helped them pay less for a service. You can, you know, debate
whether that is high or low. I think we want that to be
more helpful for people. But that being said, if you look at it from the other direction there were many people that felt like having talked about
cost with their provider indeed help them be able
to pay less for a service. Some people decided that and made a plan with their provider that they were gonna get the service but get it later at a time
they could better afford that. You see a lot of reports
about this in the media especially and when you talk to people who are proceduralists,
especially who deliver more elective or in some
cases discretionary services. They find that they’re
volume often increases at the end of the year, right? When people have spent
through their deductibles and then they don’t face the cost sharing. So deciding to put off a service
until you could afford it would probably fit into that reason. And then deciding a service
wasn’t worth the cost. A big focus of IHPI and
of my work and others is helping people avoid using
low-value healthcare services. So when people talk to the
doctors about the cost of service we don’t know from this particular survey whether or not, whether people should’ve or should not have gotten the service that they talked about. But I think interesting that many people felt like that helped them
and their provider together decide that a service wasn’t
worth the cost to them. Maybe a good thing, maybe a bad thing but a common outcome
of those conversations. Things that facilitated
cost conversations, somebody in the providers office. That makes sense if you’re
having a conversation with somebody in the healthcare system hopefully somebody in
the healthcare system would be helpful in doing so. And health plans as well were
another common facilitator. In terms of things that stood in the way of cost conversations, there was no dominant reason. But the most common one
was people not considering doing so in their healthcare seeking. Another thing we asked
about and this also emerged from the cognitive interviewing, the belief that providers can’t
help with healthcare costs. So we went in each of those questions when people did not
engage in those behaviors we went with some maybe
commonly held beliefs of why people may not
engage in those behaviors. So providers can’t help was one of them. And that was something that just over one in four people endorse as a reason. And then the way that
we asked the question we did not limit people to
just one particular reason, we asked them everything
that was relevant for them. Alright turning next to comparing prices. Again, just like in cost
conversations with providers, the most common things people
were comparing prices for: Prescription drugs, outpatient
visits and procedures. Really, really common kinds of healthcare services for many patients and the things that are
often quite expensive for people in high-deductible plans. The perceived effects, again,
people often felt like it helped them pay less for a service, sometimes felt like that
helped them get needed care. And again, often helped people decide that a service was not
worth the cost to them. In terms of the kinds of
things that facilitated comparing prices: Use of a website, that’s often how this
information is delivered. As well as support from a health plan. Looking at the barriers then, again, the most common thing is people not considering doing so. Another commonly held belief among people who did not compare prices was that prices don’t vary all that much. We of course know that that’s not true. There’s huge variation in almost every kind of healthcare
service in almost every market for a whole host of peculiar reasons. And so when people feeling
like that’s the case and that being a reason for
people not comparing prices I think that’s important. Alright, finally and quickly, negotiating prices for a service. When were the 98 out of the
1,700 people who we surveyed trying to negotiate prices for services? So again, that wasn’t zero people it was a small number of people. But there were some people that, indeed, had attempted to do this. Well it was commonly that
was for outpatient visits. And then followed by
prescription medications and imaging tests. So you did not see that as much in terms of the other behaviors. But when you look at media reports, negotiating what one will pay out a pocket for an imaging test is one of the more common
kinds of healthcare services among people who try to negotiate, around which people may
attempt to get a cheaper price. For the perceived effects. Similar to what we found
for the other behaviors often people feeling like it helped them pay less for a service, have
less debt from healthcare or free up money for other spending. For the things that facilitated that. Very similar to what we’ve
seen for the other behaviors. And then why people, the many more people who did not attempt to negotiate prices why they didn’t do it. Again, the most common thing although not tremendously dominant. The most common thing was people not considering doing so in
their healthcare seeking. As well as closely
followed by people saying, well, prices can’t be negotiated. Alright, so where does this leave us, stepping back from it and where
are the policy implications? And circling back to some
of the policy questions we were trying to address in our research. Well, I think the first thing is a few patients in high-deductible plans are engaging in consumer behaviors that could potentially help them get needed care at an affordable price. We know these are things
that people in these plans are not doing very frequently at least as of 2018 or I should say 2106, more precisely, when we surveyed people. Those who have engaged in these behaviors could be realizing more benefits. Across the board for
each of those behaviors when people engaged in them about half of them,
roughly, felt like it either helped them get needed care
or pay less for healthcare. That also means at the same
time that half of people who had engaged in those behaviors didn’t feel that way about having done so. So people could be potentially
standing to benefit more. We know that not considering
engaging in consumer behaviors and having negative perceptions, the benefits that could come from engaging in those behaviors
were commonly held beliefs and reasons why people are not engaging in these behaviors at higher rates. And I’ve hopefully illustrated for you in terms of the slides I presented earlier that the importance of these issues will only continue to grow as more people with private insurance face the challenges of high deductibles. Probably the most important slide that I will show you is this one. And where I think this
points us for the future in terms of the roles of
potential stakeholders to better support patients
in high-deductible plans. ‘Cause I think that our
research points to some pretty clear directions in this area and some new opportunities. I think employers in
particular could potentially offer health savings accounts alongside high-deductible
health plans at greater rates. Again, most people in
high-deductible plans do not have a health savings
account paired with it. If that could be done more consistently and, indeed, if employers could also be contributing to healthcare
savings at higher rates that could potentially better support more people in these plans. They could also continue
to develop programs to cultivate financial and
health insurance literacy. Again, health insurance literacy doesn’t, seems to be an issue here,
not necessarily the main one. But we know if people could
understand their plans better it’s likely, that at the very least, they’d be able to more
effectively and more consistently save for healthcare services. If people had higher levels
of financial literacy maybe they would use price comparison tools at greater rates. And so that would be the hope there. On the health plan side,
again, what health plans unfortunately often do is that they create price-comparison tools, they push it out to people, they say rates of
engagement in those tools among people with high deductibles being in the realm of five,
10%, if they’re lucky. And then bemoaning the fact
that nobody uses the tool. So it’s that plus your
explanation of your benefits that you get at the beginning of the plan you’re in, that’s about it. I think if health plans could help people to think about consumerism in healthcare as a health behavior that we
could teach people around, that we could cultivate
for people in these plans. So, again, help them
get the care they need at a more affordable price. That could be a new avenue for interventions and programs. Also I think that when health plans do a lot of predictive modeling trying to understand when people
are gonna need healthcare, when they’re likely to utilize it, when we have GIS tools that
identify where people are and maybe try to take all
of those data into account to try to make reasonable predictions about when people are gonna need care they could push information out to people, trying to prompt them to engage in those consumer behaviors at a time we’re doing so. And interacting with that
information maybe more salient. That could potentially
be a new opportunity. So, thinking more about
just in time interventions and how those could be
delivered in a space. And then for healthcare systems, making prices available
at the point of care, I think, is critical. If we don’t have that
information at the point of care and patients and providers
are together making decisions the same time as they’re considering information like: People’s health needs, their health insurance coverage, their adverse reactions to medications, what healthcare they’ve gotten before, things that they’re seeing in EHR, if we don’t have that cost
information alongside that when people are facing the
costs for those services it’s going to be incomplete information and lead to adverse
consequences for people. Direct primary care practices are starting to figure this out and their whole market for
those kinds of providers is the consumer-directed healthcare space. And so they’re often making
very clear and very transparent the price that that individual person is going to face for
that particular service. Not the charge, not the
Medicare negotiated rate, okay, not the average cost of
providing that service, the price that that individual
will face for that service and trying to take that into
account in decision making. To be clear, I’m not suggesting that we, everybody should be going down that route but I’m saying there’s
something to be learned about healthcare systems
that have adapted to that, are trying to meet patients in
these plans where they’re at and help them better
respond to those challenges. And then also potentially health systems could better facilitate conversations about costs of services, help
providers better understand how to talk to patients about these things in the context of 20 minute visits if we’re lucky, in some cases. And then also make pretty
clear for patients: Where can you go to talk
to people about your bills, potentially even negotiate a lower rate for paying for your healthcare
if the choice for you is to not pay that bill at all or maybe be able to pay something for it? So where do I wanna take
this work in the future? I am as, as Dr. Richardson
suggested in the very beginning, I do a lot of behavioral interventions and I am deeply interested in developing and testing strategies to improve outcomes for patients in these plans. Especially developing
partnerships with health systems, health plans, employers, both
on the clinician-facing side, helping us as clinicians better care for patients in these plans, helping patients in those
plans use them better, maybe even doing both together to improve outcomes for these patients, to include reducing their
cost-related access barriers, limit their out of pocket
spending and ultimately optimize the value of what they’re
spending for their healthcare. So, this work has not been just my own, it is a product of a tremendous team that I’ve been really blessed to lead. And that’s included: Angie
Fagerlin who was here up until a couple years ago,
is now at University of Utah who I continue to work with in the space, as well as Mark Fenderick,
Helen Levy at ISR, Brady West, a tremendous
statistician at ISR who’s been really
instrumental in this work, Chris Krenz who’s a research associate at Center for Bioethics and
Social Sciences in Medicine, it’s a mouthful, Betsy Cliff,
a great doctoral student in Health Management and Policy who is a soon to be health economist. And then Simone Singh from
the School of Public Health as well as Tim Petersen. We’ve all been working together
in this line of research and it’s been a really fun ride. I’ve also gotten great
support from colleagues at the Detroit Regional
Chamber who’s really helped us keep our finger on the pulse of questions that are
important to employers and health insurers in this area and ensuring that the
information we’re generating is gonna be relevant to policy
makers and practitioners. Jonathan So is who I’ve
worked with most closely and that work has been really tremendous. And then this work has
been funded and supported by Mcubed, the Robert
Wood Johnson Foundation and then as well as supported
by a career development award from the Department of Veterans Affairs. So I am grateful for all of that. And I am grateful for your attention and I look forward to your questions. (group clapping) – [Caroline] So, the
whole concept of insurance is to pool risk, right? Because there are sort
of unpredictable risks that are relatively rare
across the population. And if it was easy sort of to plan ahead and save for these risks people would be doing it without insurance and you wouldn’t need insurance products. But we know that that’s
not how healthcare works, that healthcare spending is unpredictable and it varies, there’s a lot
of variance from year to year about how much you have to spend. And so when you spend a
lot of time talking about health savings accounts
and expecting people to put some aside money in a
health savings account for it, it sort of seems like an oxymoron to have insurance that
you have to plan ahead for and save for yourself on. It doesn’t seem like it makes sense but I think where we are is
we’ve gotten to this point where chronic disease management is folded into the high-variance
insurance plan coverage and it’s distorting things in a weird way. And I’m just wondering is it really possible
for people to plan ahead for what they’re gonna have to spend in some efficient and effective way? Is that really limited to
chronic disease management which they know they have, they know they’re gonna
have to spend money on and should we maybe be pulling that out of the whole insurance model and putting it into something that isn’t a risk pooling device? And then just a point of clarification. My patients have told me that if they put money into
a health savings account and don’t spend it, it goes away. Is that accurate? – That is not true but
they may be thinking about flexible spending account or an FSA. So one thing about HSA is
the money you put into it it is tax-deductible so
you don’t pay taxes on it, you can actually deduct it
from your taxable income. You can then put that in an account, it rolls over year to year
and you own it, right? So you can take it away with you if you leave that
employer or what have you. Use it for credible medical expenses and you don’t have to pay taxes on that. And then people can withdraw
that on the back end once they hit 65, they have
to pay taxes at that point. If you pull out earlier
there’s a 20% penalty in addition to having to
pay income tax on that. – [Caroline] And so in your numbers where you said people had
health savings accounts, you didn’t include flexible spending? – We did not, no. So we used, yeah, we used the National Health Interview
Survey wording about that item which included things like
health reimbursement accounts but makes a clear distinction
between that and FSA. So we were very confident we
were not capturing FSAs there. – [Caroline] And what percentage of people with these high-deductible accounts have flexible spending accounts? – That I don’t know, I’m not sure, yeah. But so that’s, yeah, that you know and HSAs are kind of tricky. You’ve gotta have a lot
a financial literacy to be able to use that and
high levels of education and numeracy and all of the rest. So it’s clearly not a panacea
to that problem either. I think and you know your
point is an excellent one about chronic disease management. You know, at some point
your absolutely right. I think from a theoretical standpoint that’s a point that’s made, I remember getting my MPH here in Health Management and Policy. We were trying, another goal of insurance is to protect people against catastrophic financial losses, right? Not pay for routine, predictable services. That for better for worse is what health insurance has
become to a lot of people. Now maybe that’s swinging
back in the other direction. That, you know, as you have
more catastrophic plans that’s indeed what they do. They still provide that
but they don’t provide what we are, now used to that we had under first dollar coverage
is paying for everything. So the pendulum swung
back to maybe perhaps a little bit more of the true goal of health insurance coverage. But where that sits then for people with chronic conditions is a major issue. And again, Mark has been
working on that piece for, you know: Could we
allow no cost sharing under these plans for
chronic disease care? That may be an opportunity. That has proven a lot
harder than one might think. But if you think about Washington maybe that’s not surprising, I don’t know. Yes, John. – [John] Jeff, thanks
for your presentation. I have a two part question,
one on the provider side and one on the health plan side. Do providers like University of Michigan know when patients have
high-deductible health plans and if they do could they
flag it for us as providers? – They do not know but they could. And I would love for someone to be willing to test
that as an intervention. I think that simple flag would, may lead to different conversations and may lead providers to be more attuned to some of these issues
especially you know sure providers may not know
the exact dollar amount but I think most of us know what are some of the big ticket items and which are to a lesser extent. Especially too as we think
about medications and things. I think that simple flag and making that clear for providers would lead to different outcomes. And I think we should try to do that. I will say that Michigan Medicine theoretically knows that somewhere, right? They may know the name of
that health insurance product, they may have ways of
being able to estimate from the allowable rates
and what people pay and things like of that nature. They may be able to estimate whether somebody’s in
a high-deductible plan. That’s not easy for them to know. But you could also
imagine a world in which that is pretty clearly communicated at the point of care and
that being beneficial. – [John] Over there like
in patient registration we could ask patients or do it directly from Blue Cross Blue Shield, if they’re, one of our dominant payers ask them to share that information. My second question is
on the health plan side. When a high-deductible
health plan’s offered do they bundle it with other tools that health plans have to keep prices down like in-network versus
out-of-network pricing, reference pricing, narrow networks? Do people get the benefit of
those negotiated discounts if they’re in a
high-deductible health plan before they reach their deductible for services that they use? – So just so I’m understanding
that correctly John so you’re saying our, when people are underneath the deductible are they able to take advantage
of some of those things? – [John] They’re paying out of pocket but the provider knows they’re in United or Blue Cross–
– Yeah, yeah. – [John] High-deductible health plan. Do they get the benefit
of negotiating discounts? – They do indeed, they do indeed. Yeah, so initially thanks. So they do, right. So you then, so for example
if you had Blue Cross coverage you would pay the Blue
Cross negotiated rate even if you’re the one that’s
paying for the whole thing. There are a limited
number of circumstances where sometimes that’s, that is actually more expensive than what, so some imaging facilities, for example have tried to undercut and
are trying to win on volume. And they say, listen this is
our retail price for this, don’t even use your insurance
if you have a big deductible ’cause that may well be cheaper than the negotiated rate with your health plan. Talked to a reporter for
the Philadelphia Enquirer about this a couple months ago. Had encountered this again and
again with different patients who when they went through their insurer they actually paid a
higher rate, surprisingly. But that’s currently how that works is you get the benefit
of what they negotiated. Yeah. And we have just a announcement,
we’re like right at five so I’m happy to stay as
long as you guys want but yeah, sorry, yeah. – So I guess I’m curious about specifically people with chronic disease but how do these behaviors
change for people with chronic disease versus somebody who has a catastrophic
event during the year? And do we have different interventions that we’re gonna be giving people who are experiencing like catastrophic, a fall off a deck, a
motor vehicle accident versus somebody who we know up front is gonna be managing their diabetes? – Yeah, it’s a different question. So maybe I’ll just
re-phrase that a little bit which is that it sounds
like your asking about is: Could there be different interventions for different kinds of health issues for people in these plans? I haven’t really thought of that but yeah I think that’s a great point. You know, maybe again that if
you have a chronic condition ask your doctor what kind of care they think you’re gonna need next year. If you’ve got diabetes, are the things that I’m definitely gonna
need when I’m in this plan. Tell your provider you’re in
a high-deductible plan, right? So there may be certain skills. You have a catastrophic
condition, you know, I think even the most staunch proponents of consumer-directed healthcare when they’re honest about it would say that people should
not be comparing prices when they’re having crushing
substernal chest pain on the way to a hospital, right? That makes no sense. But it may be to the negotiation piece. If your stuck with a big bill
from some catastrophic event you should talk to the healthcare
provider about that bill and whether they’d be
willing to accept less. All they can do is say: No. It turns out sometimes
that actually helps people. You could imagine something like that. Again, that would need
to look very different but that’s a great point and I’ll think more about that, thank you. – Why don’t we adjourn
and move out to the lobby and then we can keep
peppering you with questions. – I would love that, thank you. Thank you everyone. (group clapping)