00:00:01
we're here today with Wharton and ped
00:00:02
med professor david ash to talk about
00:00:04
his research on behavioral economics and
00:00:06
how they can help us develop better
00:00:08
health incentive programs for employees
00:00:10
David thanks for being here today well
00:00:11
thanks for inviting me so can you give
00:00:13
us a brief overview of your research
00:00:15
sure well i mean i think the background
00:00:17
is this that for so many american
00:00:20
companies and American employers even if
00:00:22
they're not in the healthcare business
00:00:24
health is central to what they do
00:00:26
because it keeps their workforce healthy
00:00:28
and because frankly health care costs
00:00:30
have an incredible impact on their
00:00:32
bottom line everybody knows that health
00:00:35
is determined so much by individual
00:00:38
human behavior and we now have much
00:00:41
better ways of motivating that behavior
00:00:43
from principles of behavioral economics
00:00:45
principles that frankly employers I
00:00:47
think have been a little slow to adopt
00:00:48
and so what are some of the key
00:00:50
takeaways from a research in terms of
00:00:51
how employers could use these principles
00:00:54
to design better health and center for
00:00:56
sure well I mean I think 11 comment
00:00:59
would be to back up a little bit and say
00:01:00
well what are they doing and I think so
00:01:02
much of our approach to health actually
00:01:04
whether it's from a doctor approaching
00:01:06
health care or an employer an HR group
00:01:09
thinking about wellness for their
00:01:10
employees is based on the principle of
00:01:13
education we should just educate
00:01:15
employees that smoking is bad for them
00:01:17
and that fitness is good for them and
00:01:19
they should be adherent to their
00:01:20
medications it's a very rational model
00:01:22
that assumes that if our employees only
00:01:24
knew what to do they would do that and
00:01:26
and as soon as you say those words you
00:01:27
realize how limited that approach really
00:01:30
is because of course people do know that
00:01:32
smoking is bad for them but it's very
00:01:33
hard to quit or they they start up
00:01:35
smoking in the first place they know
00:01:36
they should take their medications so
00:01:39
giving them education rarely is going to
00:01:41
be enough to move the needle and yet so
00:01:44
many programs that employers use are
00:01:46
fundamentally based on that the next
00:01:49
level that employers use actually is
00:01:50
thinking about economic incentives why
00:01:52
don't we give a reduction in a health
00:01:54
insurance premium or some kind of pay
00:01:56
for performance approach to get
00:01:59
employees to move forward and there is
00:02:01
some evidence that that works but the
00:02:02
work that we do has been largely focused
00:02:05
on different kinds of arrangements to
00:02:08
help motivate either employees on the
00:02:10
employer side or patients in the
00:02:12
clinical setting because I'm a physician
00:02:14
to try to get people to take better care
00:02:16
of their health recognizing the
00:02:18
psychological foibles and pitfalls that
00:02:21
we all fall into so if I'm an employer
00:02:23
I'm trying to design an effective health
00:02:26
incentive program what are some
00:02:27
conclusions from your research that
00:02:30
might be of a surprise to me as I'm
00:02:31
sitting down trying to design this plan
00:02:33
yeah so a lot of the research results
00:02:36
are surprising and they're surprising
00:02:37
because they're sort of irrational in
00:02:39
fact they take advantage of the fact
00:02:40
that many of us don't think in rational
00:02:43
ways so here's an example of a recent
00:02:45
study that we did was led by a colleague
00:02:46
mitesh patel also a colleague kevin volp
00:02:49
and i we did this with a group of
00:02:51
employees to try to get them to walk
00:02:53
more and we we're aunt with a randomized
00:02:56
control trial real science thinking how
00:02:59
can we get employees to walk at least
00:03:00
seven thousand steps a day one group got
00:03:03
was a control and group they just were
00:03:06
given feedback about how far they walked
00:03:08
some groups were paid in this case we
00:03:10
paid them up to a dollar we paid them a
00:03:12
dollar forty a day for every day in
00:03:15
which they walk seven thousand steps a
00:03:17
third group got actually what we would
00:03:21
call a loss framed incentive exactly the
00:03:23
same amount of money a dollar forty a
00:03:24
day is 42 dollars a month what we did is
00:03:28
we gave them 42 dollars a month in a
00:03:30
virtual account before it started and we
00:03:32
took a dollar forty back for every day
00:03:35
they didn't walk seven thousand steps
00:03:37
now mathematically those two incentive
00:03:39
arms are the same right a dollar forty
00:03:41
gain if you walk or the failure to lose
00:03:43
a dollar forty if you walk the same
00:03:47
thing it turns out those who got the
00:03:49
gain incentive that were promised a
00:03:50
dollar forty four walking didn't walk
00:03:52
anymore than the control group the group
00:03:55
that got the same dollar forty framed as
00:03:57
a loss incentive walked fifty percent
00:04:00
more than the control group it's
00:04:02
irrational it doesn't make any sense
00:04:04
it's the truth real data that shows that
00:04:08
and it's consistent with theories of
00:04:10
behavioral economics but inconsistent
00:04:11
with how a lot of employers or doctors
00:04:14
think about the best way to take care of
00:04:15
their patients so what are some of the
00:04:17
underlying forces here in terms of like
00:04:19
that's work here like if we're thinking
00:04:21
about this like if I'm not necessarily
00:04:22
doing walking but I'm doing any sort of
00:04:24
incentive program or any sort of
00:04:26
employee health program
00:04:28
and I think about this using behavioral
00:04:29
economics to kind of design something
00:04:31
that will make people that will actually
00:04:34
get the result that I want which is
00:04:35
healthier employees like what are some
00:04:37
of these irrational ways of thinking
00:04:39
maybe that would help me out like how
00:04:41
would I do the back what is the best way
00:04:43
that you find to do incentives what is
00:04:44
the best way to sort of make people do
00:04:46
the things that we all know we should do
00:04:48
but that doesn't mean we're going to do
00:04:49
them yeah I mean I think it just phrased
00:04:51
it perfectly how do we do the things we
00:04:52
all know we should do but don't do it
00:04:53
anyway and as soon as you say that you
00:04:56
realize what strategies you have to use
00:04:58
because it's not going to be the
00:04:59
rational one because you just gave a
00:05:00
very rational argument of course I
00:05:02
should do this and the rational view is
00:05:04
well then I do it but of course that's
00:05:06
not what really happened so you have to
00:05:07
move beyond the typical education i i'm
00:05:10
not against education i'm a professor
00:05:12
write this I'm in the business of
00:05:13
education but it can be a bit of a decoy
00:05:15
when when something more effective might
00:05:17
be in reach and so I think the answer to
00:05:20
your question the more direct answer to
00:05:21
your question is to think well what are
00:05:23
the pitfalls we normally fall into so
00:05:25
one of them is that I just mentioned we
00:05:28
see losses as more potent than gains so
00:05:31
let's see how we can frame things as
00:05:32
losses another pitfall is that objects
00:05:36
right in front of us or goals right in
00:05:38
front of us are so much more salient so
00:05:40
much more meaningful than things far in
00:05:42
the distance right this is what we don't
00:05:43
save for retirement enough it's why we
00:05:45
eat that piece of chocolate cake right
00:05:47
in front of us even though we're on a
00:05:48
diet it's why we don't take our blood
00:05:50
pressure medicine even though we
00:05:52
desperately want to avoid a stroke that
00:05:53
might happen years later all of these
00:05:55
things reflect errors in which we focus
00:05:58
too much on the present and not enough
00:06:00
on a future how do we fix that book can
00:06:02
we accelerate the benefits can we create
00:06:05
incentives today for taking my
00:06:08
antihypertensive incentives today for
00:06:10
not eating the chocolate cake incentives
00:06:12
today for going to the gym not the
00:06:14
incentive we would normally get which is
00:06:15
a you know thinner healthier person 10
00:06:19
years from now or several months from
00:06:21
now and financial incentives our way to
00:06:23
do that but there may be other
00:06:24
incentives to social incentives right if
00:06:26
I if I set up a system to go to the gym
00:06:29
with a buddy right you don't we don't
00:06:30
need research to tell us this people are
00:06:32
more likely to go to the gym if they've
00:06:33
made a little pact with a buddy to show
00:06:35
up because you feel guilty if you don't
00:06:37
show up those are sort of social
00:06:39
incentives those are very powerful also
00:06:41
at in healthcare we often think of
00:06:43
healthcare so private that we don't want
00:06:44
to engage two people together with it
00:06:47
but we do that with so many other
00:06:49
aspects of her life and going to the gym
00:06:51
or taking your medicines it's really not
00:06:52
that private so what a misperception I
00:06:54
thought that with the research address
00:06:56
that I thought was very interesting is
00:06:57
this idea of you know like you said
00:06:58
education is obviously important and
00:07:00
it's obvious important to be transparent
00:07:02
about what you're offering someone what
00:07:04
you're doing but at the same time one of
00:07:05
the things i think the research gets
00:07:07
that that's very interesting a
00:07:08
misperception it kind of corrects is
00:07:09
this idea of sometimes it is possible to
00:07:11
give people too much information if you
00:07:13
maybe throw too much at them that they
00:07:14
actually miss what you're actually
00:07:16
trying to tell them or get them to do
00:07:18
absolutely and and so there is like a
00:07:21
fine line between enough information and
00:07:24
too much information I think the best
00:07:25
example for that are the very health
00:07:27
insurance plans that employers you know
00:07:30
give to their employees so it's April
00:07:33
right now the University of Pennsylvania
00:07:35
has open season few no benefits open
00:07:38
season I've got like I've got to do that
00:07:41
I think it's actually due tomorrow or
00:07:42
something and I have to select a bunch
00:07:45
of hell I have to select the best health
00:07:46
care plan for me and my family well at
00:07:48
Penn I spent 15 years as the director of
00:07:51
the Leonard Davis institute of health
00:07:52
economics I have to think of myself as
00:07:55
someone pretty knowledgeable about
00:07:56
health insurance and health benefits and
00:07:58
I can't figure out what the best health
00:07:59
insurance is for me and my family and
00:08:01
the reason is is that there are these
00:08:03
complicated tables of deductibles and
00:08:06
co-payments and out-of-pocket maximum
00:08:08
HSAs and all these other kinds of things
00:08:10
each one of those is a financial
00:08:13
incentive designed to try to help me
00:08:16
move my health care in a particular
00:08:18
direction but I would submit that a set
00:08:21
of financial incentives you don't
00:08:22
understand is a set of financial
00:08:24
incentives that can't be effective it's
00:08:26
just too complicated I think we need to
00:08:29
greatly simplify our almost our approach
00:08:32
design largely by economists and
00:08:34
actuaries and turn it into an approach
00:08:36
that's understandable by humans now I
00:08:39
feel like sometimes maybe they companies
00:08:41
don't do this because they feel like
00:08:42
they're gonna insult people's
00:08:43
intelligence like of course you can
00:08:44
figure this out of course you want all
00:08:45
this information but maybe really people
00:08:48
don't they just want to know what's best
00:08:50
for them really in the end is what they
00:08:52
want to know I don't think it should be
00:08:54
insulting I mean
00:08:55
you know I think I know a lot about
00:08:57
health insurance and yet it's just
00:08:59
complicated and I'm busy and it's not
00:09:01
about cognitive demand or you know
00:09:05
cognitive ability it's just more about
00:09:07
like we should be making it easy for
00:09:09
people and not rely on their ability to
00:09:11
handle hard things lots of our employees
00:09:13
are smart and they can handle hard
00:09:15
things but why should we rely on that we
00:09:17
should be making it easy and I think so
00:09:19
much of behavioral economics is about
00:09:20
that also how do we make it so that it's
00:09:23
easier to walk how do we make it so that
00:09:25
it's easier to take your
00:09:27
antihypertensive medications to lower
00:09:28
your blood pressure why we have
00:09:30
co-payments for antihypertensives
00:09:34
they're not drugs of abuse no one's
00:09:35
going to go and take lots of blood
00:09:37
pressure medicines for the fun of it we
00:09:39
want people to take their blood pressure
00:09:40
medicines why do we make it so hard for
00:09:43
them to get that medication and I think
00:09:44
that there once we take a step back we
00:09:47
can find lots of things that we're doing
00:09:49
that don't have a good reason
00:09:51
psychologically right we could we should
00:09:54
be defaulting patients into automatic
00:09:56
refill plans so they don't run out of
00:09:57
their medications and fall off the wagon
00:09:59
we should be you know lots of approaches
00:10:02
once we recognize what are the steps
00:10:03
required for certain goals and what are
00:10:05
the psychological pitfalls that we all
00:10:07
have as we approach those goals now I
00:10:10
mean health incentive programs i mean
00:10:11
have really been in I mean if not the
00:10:13
news there's sort of in our cultural
00:10:14
lexicon now it seems like every company
00:10:16
has them and if they're not they're
00:10:18
probably gonna have one next year or the
00:10:20
year after that so as companies are
00:10:22
adapting these things I mean do you
00:10:23
think people our company's being
00:10:26
thoughtful enough when they're adopting
00:10:27
these things are they kind of being
00:10:28
adopted in a vacuum without thinking
00:10:30
about the particular particulars of that
00:10:32
company of that company's employee
00:10:34
population or about some of the things
00:10:36
you're talking about about how do we
00:10:37
design a program that doesn't just work
00:10:39
for the short-term to get us what we
00:10:40
want but actually makes for people who
00:10:42
are healthier in the long term yeah so
00:10:43
it is about more than eighty percent of
00:10:46
us companies are using some form of
00:10:48
financial incentive to promote the
00:10:50
health of their employees and or their
00:10:52
spouses and dependents and the leg so it
00:10:54
is widespread is probably going to grow
00:10:56
more than that now what I think is
00:10:58
interesting and I hear it I don't want
00:10:59
to be too insulting but I think that a
00:11:01
lot of employers are not using these
00:11:03
techniques as efficiently as they could
00:11:05
and one of the reasons is you know as
00:11:07
I've alluded to is they've got almost to
00:11:08
ration
00:11:09
own perspective they the assumption is
00:11:12
well I'm using economics to improve
00:11:14
behavior therefore i'm doing behavioral
00:11:16
economics that's not behavioral
00:11:19
economics that's economics right paying
00:11:21
someone to do something is a highly
00:11:22
transactional approach it becomes
00:11:24
behavioral economics and much more
00:11:26
potent when you use psychological tricks
00:11:29
that that require these little clever
00:11:30
design elements and actually that's not
00:11:32
widely understood those those little
00:11:35
tricks you know take take some
00:11:37
experience and it turns out that subtle
00:11:39
difference is as I mentioned the framing
00:11:41
of a reward as a gain or a loss a very
00:11:44
subtle difference can can have a huge
00:11:46
impact so unless you know what you're
00:11:48
doing it's actually hard to make use of
00:11:50
behavioral economics in the workplace so
00:11:53
what would you say sets this research
00:11:54
apart from other research can bid that's
00:11:56
being conducted on this topic so um you
00:11:59
know there isn't a lot of work that is
00:12:00
currently being done and behavioral
00:12:02
economics and health that combines work
00:12:05
in the Wellness space how do I get
00:12:07
people to walk more or to lose weight
00:12:09
with our insights from really clinical
00:12:12
medicine how do I get someone who
00:12:14
actually already has an illness let's
00:12:15
say congestive heart failure and gets
00:12:18
them to stay out of the hospital and so
00:12:20
the that the acknowledgement that we can
00:12:23
learn something from both of those
00:12:25
settings and combine them now if you're
00:12:27
an employer for example it's easy and
00:12:30
comfortable to think about wellness but
00:12:33
actually well this is not going to save
00:12:34
you a lot of money in your employee roop
00:12:36
it's not where the money is the money is
00:12:37
in the very few employees that you have
00:12:40
who have some serious illnesses and so
00:12:42
the ability to improve the health and
00:12:45
health care of those employees is going
00:12:47
to have a much larger effect on your
00:12:48
bottom line then the important but i
00:12:51
would say highly diluted effects of the
00:12:54
wellness activities for your employees
00:12:56
and so what's next for your research
00:12:58
what do you were you going to go from
00:12:59
here so we've done a lot of work on the
00:13:03
use of financial incentives to motivate
00:13:05
people and again financial incentives
00:13:06
that are designed with psychology and
00:13:09
mind not just transactionally and
00:13:11
there's been tremendous success in
00:13:13
identifying ways to manipulate financial
00:13:15
incentives to motivate employees I think
00:13:18
we're increasingly interested in using
00:13:20
social incentives people sometimes
00:13:21
freaked out about it financial
00:13:23
are you paying people to lose weight or
00:13:24
to stop smoking can seem a little creepy
00:13:27
or like a little bit of a nanny state
00:13:28
but actually I think people are much
00:13:30
more motivated by the social
00:13:32
interactions that they have they want to
00:13:34
feel good among their colleagues they
00:13:36
don't want to be a dead weight they care
00:13:38
what other people people think about
00:13:40
them and to the extent that we can
00:13:41
harness those very legitimate and very
00:13:44
pro-social emotions and apply them to
00:13:46
health settings I think we can advance
00:13:48
individual and population health forward
00:13:51
a lot one of the barriers to that has
00:13:54
been as i mentioned that people have
00:13:56
often thought of health and healthcare
00:13:58
is very very private but if you were to
00:14:00
watch what people say about themselves
00:14:02
on twitter or on facebook you would
00:14:04
realize in an instant that actually
00:14:06
people are quite open about their health
00:14:08
and I think that privacy is important
00:14:09
but honestly it's probably not as
00:14:11
important as many people think it is
00:14:13
David thank you so much for being here
00:14:15
today thank you
00:14:31
you
