2017 Public Health Grand Rounds 04/21

PUBLIC HEALTH GRAND ROUNDS Linking Research to Community Health Improvement Jointly sponsored by the Department of Public Health Sciences & URMC Center for Community Health

- Welcome, everyone, to the grand rounds.
So, today we are very pleased to have Dr. Kathi Heffner
to present one of her studies,
aging-related chronic pain and poor sleep.
So, Dr. Kathi Heffner is an Associate Professor
in the School of Medicine and Department of Psychiatry
in the URMC here.
In 2008 she joined the Rochester Center
for Mind-Body Research, and is now leading the development
of its evolution into what is burgeoning
as the Healthy Aging and Research Program.
Dr. Heffner's healthy aging research is aimed
at understanding the role of social, cognitive,
and behavioral factors, particularly sleep
in older adults, physiological stress, regulation,
and immune function.
She applies these pathways to understanding complex
clinical conditions, including chronic pain,
as well as depressive and anxiety symptom comorbidity.
Dr. Heffner received her PHD in 2001
from the Interdisciplinary Social Psychology Program
at the University of Nevada, Reno,
where she specialized in cardiovascular psychophysiology
and health.
She then completed a postdoc through a fellowship
in psychoneuroimmunology at Ohio State University.
So, welcome, Casey.
- Thank you.
(audience applauding)
Hopefully you can hear me.
Does that sound good?
Thank you.
Let me get my slides up here.
So, thank you very much.
It's a pleasure to be here today to talk with you
about some work that I and my collaborators
have been doing in the area of chronic pain and poor sleep,
and in particular we've been trying to understand
how we can use sleep interventions to improve not only
the insomnia that tends to rampant
in chronic pain populations, but also to improve pain.
So, what I'd like to do today is just to describe
what we do now about these relationships
between chronic pain and poor sleep,
and I'll be also talking about a promising
behavioral intervention that's been shown
to be highly efficacious for insomnia treatment,
but that also shows promise for improving pain,
particularly, and as what I'm interested in, older adults
with chronic pain.
And, so, I'll also be describing our work aimed
at understanding inflammation as a mechanism
linking sleep and pain improvement,
and I'll describe a pilot study
that we just recently completed that targets older adults
with osteoarthritis pain.
And then, finally, what I hope to do is to have enough time
that we can discuss a bit more about the strategies
that I think are needed in order to really advance
the treatment of comorbid pain and insomnia,
and this is, of course, an audience that I think can help
us think about these bigger-picture questions
about how population health and public health
can both contribute to maximizing reach
of these interventions, and also defining the populations
who may benefit most from them.
So, I'll start by underscoring the public health challenge
that is chronic pain, and many of you are familiar
with these data.
Pain affects around 100 million adults in the US.
It's very costly to treat and to manage.
Estimates are around $600 billion is spent annually
dealing with chronic pain.
And it affects specific subpopulations disparately,
and I'm very interested in older adult and the elderly,
for whom chronic pain is often undertreated;
however, even with this undertreatment,
medicare programs tend to bear a lot of the cost
of medical expenditures for pain.
And estimates are that almost 25% of those costs
are born by medicare programs.
So, when we're trying to think about interventions for pain,
a number of challenges that are presented
because chronic pain is very unique as far as a condition.
It's very complex.
As the International Association for the Study of Pain
defines it, it's an unpleasant sensory
and emotional experience associated with actual
or potential tissue damage.
So, we have this very complex phenomenon
that's not just simply a physical sensation,
but it's also an emotional experience;
and our experiences can affect those pain sensations.
So, we have to approach pain treatment,
really targeting these multiple facets of pain.
And then chronic pain can present a challenge
because this pain does persist,
even once the healing has happened,
and there's no more evidence of tissue damage.
And, so, you have to work with these situations with people
to try to work them through the pain experience.
And then chronic pain itself has a host of comorbidities
that add to this complexity,
and some of the most common are depression and anxiety.
And, in fact, 30% of chronic pain patients
will show evidence of elevated anxiety symptoms,
depression occurs.
I'll show you some data in a bit,
but depression and anxiety are highly comorbid.
But the comorbidity that I'm very interested in
with regard to chronic pain is insomnia.
Or clinical sleep disturbance, as we also refer to it.
And sleep problems are a significant complaint
among chronic pain patients.
It's a big part of their experience,
and up to 90% will have some sort of sleep disturbance.
Clinical levels of insomnia are highly prevalent
among chronic pain patients.
This is just one example from a study
of chronic back pain patients,
where more than half report at least moderate
clinical levels of insomnia that is meeting
diagnostic criteria.
But what's important to recognize with insomnia
in the context of chronic pain is that it's not solely
a secondary symptom of the chronic pain;
and really the field has moved towards recognizing
that insomnia is its own entity,
even in the context of chronic pain,
and that we can't see it solely as a symptom of depression
or of pain.
And, so, we need to target it specifically
in order to improve quality of life in these patients.
This is because sleep disturbance itself
is its own public health issue.
And in 2006 the Institute of Medicine
identified sleep disorders and sleep deprivation
as unmet public health problems.
Today the Center for Disease Control highlights
insufficient sleep as a public health problem.
And we know now that there are multiple determinates
of poor sleep.
When we think about these factors at the societal level,
we can think about how we define our work weeks
societally and culturally, and how professions
are structured, and what professions we think require
more hours of work and different times of day.
At the social level, you can think about social development
and how our sleep changes with that.
I think a great example of this is college students.
College students have the whackiest sleep schedules
on the planet, and we actually exclude them
from sleep studies because they're so chaotic.
So, this is the time in life when sleep is very different.
At the individual level,
there are personality characteristics like neuroticism,
and other associated individual traits
like rumination and worry that can increase susceptibility
and risk for insomnia.
And what's important here is that what the literature shows
is that this disturbed sleep, regardless of the source,
increases risk for a host of adverse health outcomes,
including increasing risk for cardiovascular disease,
metabolic risk, poor mental health,
it affects the immune system;
and, of course, all of these adverse health outcomes
increase mortality risk.
And when it comes to older adults,
sleep problems are typically undertreated.
They're not diagnosed, and they don't tend to be,
even when they're brought up in the clinical context,
they're not addressed because there is
an aging-related stereotype that we perceive our sleep
gets bad as we age; and that is not the case.
Sleep does change over the course of our lifetime,
but sleep problems are not an inherent part
of the aging process.
And, so, this is an educational piece
that I think is very important as far as raising awareness
of the role of sleep disturbance and aging.
So, with this attention to sleep
as its own public health issue,
there's really been a drive to develop
behavioral sleep interventions that can improve sleep.
And this has come from really the patient population,
in particular older adults who are often prescribed
sleep medications.
But many individuals, young and old, do not like
pharmaceuticals for their sleep improvement.
There are a lot of side effects.
Often people will report that they do sleep,
but then they feel groggy the next day.
You don't have a refreshing sleep,
and, so, there is a demand
for behavioral sleep intervention.
The sleep intervention I'm gonna highlight today
has been developed over time and has demonstrated efficacy,
very strong efficacy for improving sleep,
and that's cognitive behavioral therapy for insomnia.
Or CBTI, I'm gonna refer to it as for the rest of the talk.
So, CBTI is based on cognitive behavioral therapy principles
so it has multiple components that cover
from the cognitive and behavioral domains,
but they all target sleep specifically.
So, cognitive component focuses on the cognitive distortions
about sleep that can increase anxiety.
I know we see a lot of anxiety about sleep
in insomnia patients.
There's relaxation training in order to reduce
that cognitive arousal and the physiological arousal
that's associated with sleep.
Often this begins a few hours before bedtime
and then a patient is in bed.
So, they're provided with relaxation training
to overcome that physiological arousal.
And then the strong behavioral component of CBTI
is what we call sleep restriction,
and this is the idea that we actually in the early stages
of the intervention, we wanna restrict sleep.
So, essentially how that's operationalized
is that patients are asked to go to bed much later
than they normally do, but get up at the same time.
So, now you've created a shorter duration of time in bed.
Because the patient's going to bed later,
they're going to be more tired,
and they're more likely to fall asleep relatively easily,
and sleep through the night.
So, it's what we call sleep continuity.
Then you ask them to do that for multiple stays;
you've now given them sleep debt,
and there will be a physiological need for sleep.
The whole idea is just to increase the number of hours
that they're sleeping while in bed.
And, so, gradually you can add more hours
to that bedtime duration.
And that is actually a very powerful component
of the sleep intervention.
There have been studies looking solely at sleep restriction,
and it's highly effective at improving sleep
without any of the other components of CBTI.
And then, finally, there's stimulus control.
So, this is basically making your bedroom a cue for sleep.
So, a patient is told you can only use the bedroom
and only use your bed for sleep and sexual activity.
That means no TV in the bedroom,
and you can't bring your devices in the bedroom.
No iPhones, no tablets.
I see a few people smiling 'cause I know you do this.
(laughs) (audience laughing)
And this is step number one in therapy.
So, get rid of the devices.
So, because this particular insomnia treatment
has been shown to be highly effective,
chronic pain clinicians and researchers have turned to it
to see whether it can be effective in chronic pain patients
to improve sleep.
And this is for two reasons.
I've told you that insomnia carries its own risk.
So, then the idea is that sleep improvement should
in and of itself improve well-being
in chronic pain patients.
So, we can target the sleep, improve the sleep,
and despite the pain, perhaps there'll be a better
quality of life.
However, what we know from mechanistic studies now
is that sleep disturbance can actually increase
our pain sensitivity.
And this is from experimental studies,
where people are brought into a sleep laboratory
and they're told they're allowed to sleep
through the night as they normally would.
Other people are sleep deprived,
and they're kept up all night through horrible means
like noises and lots of ways of keeping the lights on,
and they have to stay up all night.
So, this is sleep deprivation.
Other participants are told they can sleep
part of the night.
So, they're allowed to sleep three or four hours
in the evening.
If you then test the participant the next morning
by painfully stimulating them,
so engaging them in laboratory pain testing.
This would be things like putting their hand
on a hot plate, putting their hand in ice cold water,
and asking them to report their pain or tolerate the pain.
What we find is that the participants
who were sleep deprived, or even partially sleep deprived
have greater pain sensitivity after that sleep deprivation.
So, sleep itself, we know when we're feeling pain
that's gonna disrupt our sleep,
and that's why pain and insomnia are connected;
but, in fact, it goes the other way.
When we have poor sleep, it actually increases
our pain sensations and reduces our pain tolerance.
So, the idea then is that sleep improvement
should actually improve pain.
So, what do we know about this at this point?
In 2014 Michael Smith and his colleagues published a review
of CBTI for comorbid insomnia and chronic pain.
And what they concluded after I reviewed the literature,
is that CBTI is highly effective at improving sleep
in chronic pain populations.
However, it remains unclear, although there is promise,
as to whether CBTI can actually improve pain
and reduce pain symptoms.
And I'll give you a couple of examples
of more recent research from this line of thinking.
So, this is from a clinical trial that was conducted
by Michael Vitiello from the University of Seattle.
And what they did is they randomly assigned older adults
with osteoarthritis pain to either
an education control group or a cognitive behavioral therapy
for pain group.
So, this was a CBT intervention targeting pain symptoms.
Or they randomly assigned the participants to a CBT
for pain and insomnia.
So, this CBT was targeting both pain,
as well as targeting the insomnia.
What they found,
and this is represented in the top two graphs,
is that the CBTI was highly effective at improving sleep.
However, despite that one bar there for CBT
for pain and insomnia, the sleep treatment,
and actually none of the groups differed with regard
to the effect of the intervention on pain.
So, although the CBT for pain and insomnia
was highly effective and more effective at improving sleep,
it did not show strong effects on pain.
And then in another trial, similar findings emerged.
This was by Michael Smith and his colleagues
at Johns Hopkins, where they randomly assigned,
again, older adults with osteoarthritis,
specifically here of the knee,
to either CBTI or to a control group
with a placebo sham intervention.
And here what they're looking at is what we call WASO,
or wake after sleep onset.
So, this is the amount of time you spend awake
after you've fallen asleep,
and this is a specific type of insomnia
called sleep maintenance insomnia.
And, so, what they found is that CBTI was highly effective
at improving sleep maintenance insomnia.
These participants slept longer after they fell asleep.
However, if you look at that last chart labeled D,
you can see that the clinical pain findings
were not as robust.
What they found is that there was a decline over time
in pain for both groups.
There was some suggestion at three months that maybe
the CBTI group was fairing better;
but there still wasn't a strong,
there wasn't strong evidence
that this particular intervention affecting patients' pain.
These two studies, in the context of other trials,
and pilot work that suggests there might be some benefit
for patients if we can figure out who to target
and how to target them for using CBTI to improve pain.
But what we think is that we need to spend more time
looking at the how.
How is that improving sleep will improve pain,
and flipping that.
How is that poor sleep actually up-regulates
our pain sensitivity?
Because I would suggest that understanding
how sleep disturbance affects pain and the converse
may help identify how best to intervene and with whom.
So, we really need to disentangle this mechanism.
So, now I'd like to turn to some work that really addresses
this mechanism.
And, as I mentioned, I'm interested in inflammation
as the pathway through which poor sleep might increase pain.
When I'm talking about inflammation,
I wanna define it a little bit and give you a context
for how we're thinking of this.
In general, inflammation is an early response
by the immune system to infection.
When we get a bacteria or a virus,
we have an inflammatory response.
We also have an inflammatory response
when we have injured tissue.
And this inflammatory response is mediated
by inflammatory cytokines, cytokines like interleukin 6,
or IL6, and tumor necrosis factor alpha, or TNF alpha.
These cytokines are released by immune cells,
and they coordinate immune responses
to get rid of the infection, the bacteria or virus,
or to heal the wound.
Interestingly, we've learned a lot about how these cytokines
relate to pain.
And in a large number of animal studies,
what is found is that these inflammatory cytokines
actually amplify and contribute to persistent pain,
and they amplify pain sensitivity;
and they do this by interacting
with the central nervous system.
And, so, if you think about it,
I don't know if you're like me,
but if I get a fever because I have a virus,
I experience a lot of bodily pain,
just this general defused muscle aches.
Those are your cytokines working on your brain (laughs)
to up-regulate pain.
In addition to having an inflammatory response
to infection or injury, we also have what we call
neurogenic inflammation;
and this is an inflammatory response that's stimulated
by nervous system activation.
So, if you're familiar with the sympathetic nervous system
that supports our fight or flight response,
when we have one of those responses,
we also have an increase in release
of these inflammatory cytokines.
What the literature shows is that stressors like pain
or like psychological distress can lead to an increase
in these markers of inflammation.
And what we're finding in a number of studies
is that sleep disturbance acts like a stressor
and can up-regulate these inflammatory processes, as well,
and I'll give you some examples of that.
So, in general, about seven years ago
there was a nice review article that reviewed
the literature on sleep disturbance,
both in terms of poor sleep quality,
as well as patients with insomnia.
And, in this review, they underscored the idea
that poor sleepers are indeed characterized
by physiological stress dysregulation,
and that part of this dysregulation manifests
in higher circulating of inflammatory markers.
So, we looked at this association
in the context of chronic pain.
And, so, what we did is we recruited individual,
these were adults.
They were age and gender matched,
who had chronic low back pain
and those with no chronic pain.
And what we found is that
among the chronic low back pain patients,
when they reported poorer sleep,
they showed higher levels of circulating IL6.
We did not find this same relationship
among the adults without chronic pain,
and nor did we find that there was much difference
in their reported sleep.
So, what we suggest is that,
these findings suggest that chronic pain may increase
vulnerability to sleep-related inflammation.
We've also taken a look at the way people respond
to stressors and how chronic sleep disturbance,
or poor sleep, might affect that response,
as another way to look at how the inflammatory system
might be sensitized.
And, so, what we and others have found is that
in response to stressors, poor sleepers tend to show
larger increases in IL6 response.
And, so, this is an example from a study of older adults,
and these are very healthy older adults.
What we found is that those who reported poor sleep,
on the Pittsburgh Sleep Quality Index,
had significantly larger IL6 responses
to cognitive stressors.
We had them engage in a series of challenging
cognitive tasks, and this was relative to those
who reported good sleep.
So, given this accumulating evidence
that sleep disturbance may, in fact, sensitize
these inflammatory pathways,
we're working with the idea that sleep disturbance
may sensitize the pathways in ways in which
that may result in exacerbation of the pain experience.
So, this is a picture of what I just said,
and you can see here, so, we're working with this idea
that chronic sleep disturbance can alter
inflammatory cytokine regulation,
which in turn may increase pain sensitivity
and clinical pain, but I do wanna point out
and acknowledge here that we suggest these
are bidirectional pathways; this moves both ways.
So, if you have increased pain sensitivity,
more clinical pain, this is gonna feed back
into worse sleep.
So, this is a picture of a very vicious cycle
that we need to throw a wrench in.
And, so, if it is the case that these pathways
relate to one another, then we would suggest
that improving sleep should alter this inflammatory process
in ways that reduces pain sensitivity
and improves clinical pain.
So, now I'm going to talk a little bit about a pilot study
that was our first go at testing proof of concept
for this model.
So, this was a small pilot study funded
by the National Institute on Aging.
And my co-conspirator in all of this work, Will Pigeon,
was a multi-PI with me on this.
And in it we hypothesized three pathways.
First we hypothesized that improving
chronic sleep disturbance would improve clinical pain.
Second it would reduce acute pain sensitivity,
and, third, it would alter inflammatory response to pain.
And, so, the idea was if we could support these hypotheses,
we're providing proof of concept for this model.
This was a study of older adults, 55 to 75 years of age,
who had knee osteoarthritis that was confirmed by x-ray
or by their physician.
And I should highlight, I know I've highlighted a number
of studies now on osteoarthritis with older adults,
and there are a couple of reasons for that.
In chronic pain work and aging, unfortunately,
osteoarthritis is a normative aging process.
It happens to a lot of us.
And what's interesting about it, though,
is it manifests very different.
There's a lot of variability in the pain
that people experience as a result
of osteoarthritis development.
And, importantly, almost 80% of older adults
with osteoarthritis report some sleep disturbance, as well.
So, it's a very useful model to try to understand
chronic pain and aging.
So, people had to report knee pain,
they had to meet diagnostic criteria for insomnia,
they could have no other sleep disorders.
And, in this study, we excluded if they had elevated levels
of depression or if they had a diagnosis of depression,
or I should say that we included them,
though if they had a diagnosis of depression
but it was treated.
So, I wanna walk you through the procedures for this study.
So, we enrolled 48 older adults into this study,
and we had them come in for an overnight screening night
in the sleep lab to rule out any sleep disorders
other than insomnia.
And then once we did that, and we actually had a number
of people who had undiagnosed sleep apnea,
periodic leg movement disorder.
Those were the two most common,
and, so, they were referred to a sleep medicine clinic.
And, so, in the end we had 33 participants
who were eligible to come in for our baseline assessments.
And this comprised an overnight sleep study,
again, in a sleep lab, where we assessed polysomnography.
And I'm not gonna be talking about those data today.
We're still in the process of examining those outcomes,
although I can say disappointingly we're not finding
as robust (laughs) findings as we had hoped
with the polysomnography data,
which actually makes much of the other data
very interesting.
So, after the overnight in the sleep lab,
we escorted people to the clinical research center,
which is the research unit the hospital,
and we had them report
on their clinical osteoarthritis pain
on the Western Ontario and McMaster Universities
Osteoarthritis Scale; this is the gold standard
clinical osteoarthritis scale.
And we also had them engage in a series of laboratory
pain testing.
We used two types of pain testing.
The first was ankle shock, and we did not refer to it
as that with our participants.
(laughs) It was electrical stimulation of the ankle.
We had multiple trials of this,
and this was in the context of another
pain assessment procedure.
I'm not gonna go into detail about today,
but it provided us with a means for pain stimulation
and to assess subjective acute pain responses.
Following the ankle shock, we did a cold pressor task,
and we asked the participants to put either their hand
or their foot, it was their choice, into ice water.
I actually think that is worse than the ankle shock myself.
So, through the ankle shock,
and throughout the cold pressor task,
we had multiple trials.
And, so, we asked people about their subjective pain
on a scale from zero to 100
during those pain testing procedures.
And then throughout the study,
so before, and in the middle, and following
during our recovery period,
we did serial blood draws through an indwelling catheter
to look at, in particular here, IL6 and TNF alpha.
So, following that baseline assessment,
participants were then randomly assigned
to either CBTI or to a no-CBTI control group.
So, our CBTI-assigned participants engaged
in a six-week CBTI program,
where they met once a week with a sleep therapist.
Most of the sessions are 60 minutes;
the first one's 90 minutes.
And then the no-CBTI group just received weekly calls
from the study staff just to check in
and see how they were doing.
And after the intervention period,
we brought everybody back,
and then we did all of those assessments all over again.
So, I'm just gonna jump into what we found thus far.
So, first, we wanted to ask our main question.
Does CBTI improve insomnia?
And, for us, because we're interested in sleep improvement,
this was a manipulation check for us.
So, what we found is that, yes, CBTI was highly effective
at improving sleep in our older adults
with knee osteoarthritis.
And this was on average quite a large effect,
and on average they were well below the insomnia
diagnostic criteria at that point.
And we actually set our bar a little high,
or I should say a little lower in order to determine
whether or not they met insomnia criteria.
So, we used 10 as an enrollment number,
but we used eight as our criterion for no longer
showing evidence of insomnia.
However, when we looked closely at the individuals
who were assigned to each group,
what we found is that, yes, CBTI on average
was highly effective, but we had some people
who didn't respond to the treatment.
So, we had six people assigned to CBTI
who still showed evidence of insomnia at followup,
and that was almost 40% of our small sample.
We also had two controls whose insomnia had improved,
and they no longer met diagnostic criteria at the followup.
So, what we did is because our hypotheses
were about sleep improvement affecting these outcomes,
we have done analyses based on whether or not people
do have insomnia or don't have insomnia at followup
'cause, remember, everybody had insomnia at the beginning.
So, what you'll see are groups designated
by their insomnia status at followup.
So, was insomnia improvement related
to improvement in OA pain?
This was our primary outcome.
Could we see effects of insomnia improvement
on that WOMAC scale?
And, in fact, we did find that.
So, we actually found significant and clinically meaningful
differences in the pain ratings based on the WOMAC scale
between our participants who did and did not show
insomnia at followup.
And our group who had insomnia improvement
also showed a minimal clinically important improvement
in their insomnia symptoms, that is they had a score change
that was greater than eight.
Eight points on that, which is the criterion
to determine clinically meaningful change.
So, that was very exciting.
We wanted to look at the acute subjective responses
to the pain testing.
Did sleep improvement have an effect on those?
So, I will tell you, I don't have figures to show it,
mostly 'cause it's disappointing.
When we asked them about their subjective pain
while we were shocking them and while they had their hand
in the ice cold water, we did not see effects
of sleep improvement on those ratings.
They were very similar.
However, we asked people throughout the study
to mark along a line their current pain,
the point on the line that represented their current pain.
We did this prior to the pain testing.
We did this after the ankle shock
but before the cold pressor task,
and then we did it after the cold pressor task.
And what we found, I'm gonna show you the baseline data.
So, remember, everybody had insomnia at baseline,
but we wanted to make sure that our group
that no longer had insomnia didn't look different
from our post-insomnia group when they all came in
with insomnia.
And we found, these are the pain ratings
based on this visual analog scale.
So, we didn't see baseline differences
between these groups then.
However, at the followup we found that those
who no longer showed evidence of insomnia,
so whose insomnia had improved,
they had significantly lower ratings of current pain
relative to the participants whose insomnia
did not improve by followup.
And then I was very excited to find out what happened
with the inflammatory cytokines.
So, for the first day I'm gonna show you our IL6,
and I'm gonna show you the baseline data first.
So, this is IL6 levels across the pain-testing session
from baseline to 90 minutes post testing.
And we looked that far out because when we use
any kind of stressor or a pain stimulus,
there's some delay in seeing an increase
in inflammatory cytokines in circulation.
So, we tend to make people sit around
for an hour and a half, and then we continue
to sample their blood.
So, what you see here is that we don't see
a statistically significant difference in our groups.
So, again, these lines represent the groups
that are defined by their insomnia status at followup,
but this is when they came in when they all had insomnia.
And, so, we don't see a statistically significant difference
but we do see that those who
no longer have insomnia at followup,
at their baseline they were slightly lower
in terms of their IL6 levels.
So, what happened after the insomnia improvement?
So, we do see now a statistically significant difference
between the groups.
So, what we essentially see is that the group
who no longer had insomnia at followup
had significantly lower levels of IL6 at multiple points
throughout the pain-testing session.
I'm gonna show you TNF alpha.
So, this is somewhat promising,
and I had a lot of thoughts about the pattern that,
if anybody's interested, we can talk about offline.
But let me show you here TNF alpha.
So, these are the TNF alpha baseline data.
And TNF alpha is different than IL6.
So, with a robust sympathetic nervous system response,
you see declines in TNF alpha.
So, that's why the trend looks the opposite to IL6.
So, here at baseline, our groups at followup who differed
by insomnia status didn't show any difference
in their baseline TNF alpha response.
What happened post?
So, this is at followup.
So, now what we see is that the participants
whose insomnia improved actually had very stable TNF alpha
levels throughout the pain-testing session;
but those whose insomnia did not improve,
they still met diagnostic criteria,
showed the same decline in response to the pain testing
that we found at baseline.
So, this was, and is, we're working on the paper now,
very exciting 'cause we think that we are gathering
some evidence for proof of concept of our model.
So, what we're finding is that here
improved sleep disturbance did, in fact,
improve clinical pain in this relatively small sample.
And, I should say, this same of older adults,
aside from having osteoarthritis, knee pain,
and insomnia was otherwise very healthy.
'Cause when we're looking at these inflammatory markers,
we have to exclude for a host of conditions,
especially inflammatory-related conditions.
So, in general, we have very healthy folks in the study,
which may have, it provides a nice model,
but it may also be why we found such robust findings
in terms of the clinical pain.
And we found some evidence that there was reduced
current subjective pain with sleep improvement,
as well as an altered inflammatory response to pain.
So, based on our preliminary findings,
we're interested in developing this further
and conducting a larger trial,
and we think these kinds of larger trials are warranted,
and should test these integrative models of sleep and pain.
But what we emphasize is that we have to continue
to understand these mechanisms that link sleep to pain
in order to understand fully how best to intervene.
In terms of our program of work,
we're interested in expanding our mediational model
to include other physiological stress systems
because inflammation doesn't operate in isolation,
and it's like, as I mentioned,
or I mentioned the sympathetic system,
that feeds into the inflammatory response.
And there are certainly other stress hormones
that help regulate the immune response
that leads to inflammation.
So, these all need to be considered
if we really wanna have a comprehensive model.
I would also suggest that, as I mentioned,
we had very healthy people.
These were non-depressed chronic pain patients.
We can't ignore depression and anxiety in these models
because it's so highly prevalent
and it's a part of the chronic pain experience.
So, that needs to be worked into this, as well.
I think it's really important that we focus
on these mechanisms, and we just don't keep trying
to how can we use CBTI to improve pain,
and how could we change CBTI?
We have to understand how this works
because it's likely that those who show,
and this was a hypothesis that we've developed,
is that those who show evidence of sleep-related
physiological dysregulation, maybe the patients whose pain
will benefit most from sleep intervention.
So, we need to begin to understand the mechanisms
so we can identify who's likely to respond
to sleep intervention in terms of improvements in pain.
This is where I think we need to look at the bigger picture
because I am very interested in these mechanisms,
but I am interested in using them to put them
into the bigger picture of clinical care.
While we're taking this in a direction
of trying to disentangle these physiological and systems
that are part of this picture,
I think we also have to consider how we can translate
what we're finding here into the clinical context.
And that's really where I think the population
health level strategies and public health strategies
are gonna be important.
I think that we have to think about these things
in parallel.
Just last year the Interagency Pain Research
Coordinating Committee, which is an NIH committee,
was commissioned by the Department of Health
and Human Services to come up with a national pain strategy.
So, recognizing the public health challenge
that is chronic pain.
Trying to determine what direction do we need to take
to really reduce this public health burden?
Two of the strategies that they target in the report,
and two goals, are to define and evaluate integrated,
multimodal, and interdisciplinary care for chronic pain,
as well as enhance the evidence base for pain care.
That's great for us and our interests in understanding
how this intervention works and how we can improve pain
by improving sleep, but there are some other
bigger-picture questions that we need to answer
so that we can do this most effectively,
and really translate this work into clinical practice.
So, I'm not gonna say too much about this.
We've been thinking along the lines of how to identify
those who are most at risk for these kinds
of physiological disturbances that link their sleep
to their pain outcomes.
And, so, we had to figure out how would we even identify
those people?
If we can say, yes, there are certain people
who are more at risk for sleep-disturbed
physiological dysregulation and pain, that's great,
but we need to be able to identify these folks
in clinical care.
So, that's gonna require surveillance,
and it's going to require leveraging the medical record
to be able to identify those patients.
Currently we don't have a good way to identify patients
who have insomnia.
It's so underdiagnosed and undertreated in primary care,
and often people are misdiagnosed.
Especially older adults,
when people start talking about sleep problems,
then they're prescribed an antidepressant.
And antidepressants do help with sleep,
and that's why it's common for them to be prescribed,
but it's not necessarily targeting what should be
being targeted, and it could be targeted perhaps
through behavioral intervention.
So, we think we need to take a closer look
at how we're monitoring the sleep problems
that are in the patient population.
And we also don't do a great job of monitoring
and recording chronic pain systematically,
and that's gonna be more challenging.
You can certainly identify conditions
that are associated with pain,
but, as I mentioned, osteoarthritis
doesn't necessarily manifest
in excruciating pain experiences.
Some people manage very well with osteoarthritis.
If you do imaging, you can see a lot of osteoarthritis
disease activity, but the person isn't experiencing pain,
and often this is due to physical activity
or other health behaviors that can mitigate
the disease effects on pain.
So, it's not necessarily the case that just the disease
is going to identify who's experiencing chronic pain.
So, we need to get better at the surveillance.
And then, two, I'm always thinking
because my interest psychoneuroimmunology,
and these different inflammatory factors,
and other markers of physiological dysregulation,
that's all well and good for trying
to disentangle mechanisms, but how can we use
what we're learning from that,
and, again, leverage what we have
in the clinical medical record to identify risk?
We have a lot of clinical markers in the medical record
that are routine clinical measures.
So, when we think about markers of metabolic risk,
even things like BMI and cholesterol level,
we have measures of inflammation in the medical records.
C-reactive protein is a marker of inflammation.
We typically only assess that if somebody shows other
evidence of cardiovascular disease risk
'cause C-reactive protein is a risk factor,
or a risk marker.
However, that's something that can easily be done.
We've been thinking about how we can leverage
some of what's in the medical record,
what's done routinely, to be able to bring in
the same kind of markers that we're using
to understand mechanism to then identify those patients
who may be most at risk.
The other useful information we can get
from those clinical markers relates to outcomes.
So, if we can start identifying who's most at risk,
and do those sleep interventions with them,
we can use those clinical markers as outcomes
to evaluate over time the effectiveness
of those interventions at the population level.
So, I think these are all things that I've been
thinking about and I'm hoping that folks here
may have some ideas about that,
and help us think through some of how we can translate
this work into clinical practice,
as well as monitoring population health.
And I just wanna add one more critical point
on the other side.
So, disseminating these kind of interventions.
So, another part of the national pain strategy
has been to increase awareness of sleep
as a public health problem, as well as increased reach
of interventions to the public.
And I wanna give you an example of how this is being done
in the sleep world.
The sleep intervention world and the VA has developed
what's called the CBTI Coach.
So, this is an app that can be downloaded on anybody's phone
and the intention, and it's advertised as being
a useful tool if you're in CBTI treatment,
or if you've recently completed CBTI treatment,
but it's also being targeted to veterans
who just wanna improve their own sleep.
Because this tool doesn't require that you are
in cognitive behavioral therapy for insomnia.
There's a lot of education that's a part of this tool.
So, a lot of it is raising awareness
and a lot of self-monitoring about sleep habits.
So, it takes the principles of CBTI
and puts it in a package that can reach a lot of people.
And I think that he VA in general is leading
a lot of this effort into trying to reach people
and disseminate what we know about interventions
to those who may not be formally enrolled
in these types of therapies.
And then, finally, I think that in general
the national pain strategy, as I mentioned,
they really wanna push raising public awareness about pain.
A lot of the emphasis has been on pain and pain management.
As many of you know, and you've all heard
of the opioid crisis and a need to raise awareness
about other ways of managing pain.
There's also emphasis in the pain report about depression,
and how that's often undertreated
in the chronic pain population.
One thing that was absent from that report was insomnia.
And, so, that I think that we really need to raise
the awareness of insomnia as a critical problem
in the chronic pain picture.
And, so, with that I wanted to make sure I left time
for discussion.
I wanna thank my collaborators and staff.
As I mentioned, Will Pigeon has been
my closest collaborator in this work.
Christopher France is at Ohio University.
He's the one who taught us how to shock people
in the ankle. (laughs)
And then we have Mike Maloney and Brian Giordano
from orthopedics who have been very helpful
in identifying the study population for us.
And then all of our wonderful staff,
and the clinical research center nursing staff
who helped us with this,
and the National Institute on Aging
who supported much of this work.
So, thank you, and I will be happy to take questions.
(audience applauding) (laughs)
- [Woman] There are two comments I'd like to make.
First, I'm very impressed that all those people
who were shocked in the ankle stuck with it and came back.
- I know, it was amazing.
- [Woman] The second followup, that was very persistent.
My second question, though, really leads
to a very practical sense.
Were the people who participated in this study
that showed improvement, were they able to talk
also about how they functioned in general?
I don't know if you went into that at all.
- We haven't looked at the data.
We have quality-of-life measures.
We did find that, when we started looking at,
on the osteoarthritis scale, you can disentangle
the symptoms, pain symptoms.
So, the experience of pain sensitivity
versus pain disability, and how much interference there is
by pain in their everyday activities.
And we are finding that in addition
to them reporting reduced pain,
many report that they're just able to not let the pain
interfere with their daily function.
And, in fact, in most studies of pain interventions,
it's very difficult to move pain.
Even in cognitive behavioral therapy for pain,
which has been developed to target pain symptoms,
you find most of the time that patients
don't report less pain, but they report less interference
by their pain.
So, they can do more daily activities and manage the pain
a bit better.
So, that's always a challenge in this research.
I think it just speaks to we're not gonna necessarily cure
the pain, but it's about management,
and managing that symptom.
Did that answer your question?
- [Woman] Yes.
- I'm sorry, what was that?
- [Man] You mentioned using polysomnography in your study.
What's that dealing with?
Did you refer to that or is screening your format?
- Yeah, so, we do that on a screening night;
we also call that night the adaptation night.
So, often people come into the lab,
and the first night they sleep there,
they don't sleep like they do at home,
and it can go either way.
Either people will sleep horribly
'cause they're in the basement of the medical center,
(laughs) and it's just this weird environment,
and you have all these electrodes on your head,
or they report they had the best sleep of their life
(laughs) in there because there is,
if you go into the room, and Will Pigeon's the director
of the sleep lab, and the rooms are set up
so when you go in and you turn off the lights,
it's like a sensory-deprivation chamber.
There's no light coming in, you hear no noise.
So, people often sleep well.
So, the first night we have them come in to adapt
to that environment because what we typically see then
the second night, which is their study night
when we actually record for the purposes
of understanding sleep stages,
then it reflects more of their normal sleep habit.
The first night is really to rule out
any other kind of sleep disorder.
'Cause the first night, too, we're looking at their apnea.
So, they have to have the mask on,
and that also interferes.
So, if they don't have apnea and they come back
for the study, they only have the EEG electrodes.
- [Woman] I'm just curious about the sample
of any experiment like this where you're looking
at it through impact on the changes?
But the proof that's coming in is so selective in a way.
They want to get their medicine,
but their mental state is very different from others.
So, how do you take that into consideration?
- We don't in this work.
It's 'cause we are interested in people who are likely to,
whose sleep is likely to improve if we give them
sleep intervention.
But it is a really important part I think
of the bigger picture.
One of the things that I think is so great about CBTI
is that once you can educate people
and make them aware of what it is and what's it about,
it tends to be well received,
and people are very interested in it.
So, even people who would never go to therapy
for depression, for instance, will be interested in CBTI
because it's about sleep.
We all have sleep problems, and we can talk about those,
and there's no stigma around that.
And then the therapy is targeting sleep.
It is solely about how you think about sleep,
what behaviors you're doing that might sabotage your sleep.
And, so, we're actually using it in another study
for women with an interpersonal violence,
who have depression and PTSD who don't go to therapy.
And we're doing it first to see if improving their sleep
might improve their intervention outcomes,
but we did that because we saw it as maybe
a stepping stone therapy.
If we could get people in, it's very palatable.
People say, "Oh, yeah, I wanna fix my sleep.
"I'll definitely come to that."
Then once they see, "Oh, yeah, I can talk to somebody
"about what I'm thinking and how I'm feeling,
"so now I can do the more difficult work
"of talking about depression, and trauma,
"and those sorts of things."
But we do find that if you have an opportunity
to educate people who are resistance to psychotherapy,
that they're really willing to do this intervention.
And, so, that's why I latched on to it,
not only 'cause it's so highly effective
so we can test our models,
but it's something that can actually be disseminated
fairly easily.
- [Man] Do you study, or are there studies
of the use of NSAIDs or opioids on the cytokines,
again, related to sleep disturbance?
- We haven't done that and, in fact,
that's part of our inclusion-exclusion criteria.
If people were taking opioid medications
or sedative medications for sleep,
if they wanted to be in the study,
they had to agree to a two-week washout period
under the guidance of their doctor's observation.
I don't think there are good studies showing
what the opioids do to inflammatory outcomes,
but what's interesting about it,
and there's some animal work to suggest that,
obviously with opioid use, there's an increase in tolerance.
And the opioid system regulates stress responses.
So, our natural, endogenous opioids,
when they're released in the brain,
that dampens our stress response,
which also dampens inflammatory response.
But over time with chronic use of opioids,
you see that the endogenous opioid system
starts making more opioids.
They're resistant to the opioid,
or they respond to the opioid medication
by basically doing too much.
And then what happens is you have lots of opioids,
and then the tissues that respond to opioids,
now the receptors that the opioids talk to those tissues
down-regulate; so not the opioids are less effective.
And, so, there is this cycle,
and, so, you'd expect that that's gonna have a bearing
on those inflammatory processes.
We don't have clinically good data to know that,
but conceptually and theoretically that should be the case.
So, it would be interesting to look at that.
It gets messy with medication use.
- [Woman] Speaking of animal work.
But with animals, the rules on regulating things
are a little unclear.
The university where it schedules themselves.
Have you looked at any of the research,
or are you interested in also collaborating
with them on animal work to bridge the gap
between what you could do to a human to the animals?
- Yes and, in fact, most of the work that led to this work
in the clinical area that I've been doing
was informed by animal research
'cause there's more, even in the sleep area,
there's more research in animals,
showing that if you disturb sleep for a long period of time
in animals, typically rats or mice,
that this alters their physiological stress system.
This is like a little history of this work.
My work prior to coming to the medical center
was really focused on bringing people into a laboratory,
stressing them out, seeing what their body did,
and I'm not sadistic. (laughs)
That's just how I tested these models.
And since coming to the medical center
and then pairing up with clinical interventionists,
I've been using the interventions to try to test
the same pathways.
But instead of making people, make them better,
and seeing what changes.
But there are some things that we can't do with people that,
I think, yes, I would love to pair up
with animal researchers because I think there are ways
to do the work that compliments,
you can take what's happening,
and manipulate things in animal research,
and then figure out how to apply that to clinical research
in ways that obviously don't do that kind of harm.
- [Host] Any other questions?
Thank you, Casey. - Thank you very much.
- [Host] You're welcome.
(audience applauding)