#173 - AMA #26： Continuous glucose monitors, zone 2 training, and a framework for interventions
Episode Stats
Words per Minute
180.24538
Summary
In this episode, Dr. Bob Kaplan and Dr. Peter Atiyah answer a bunch of follow-up questions to topics covered in the previous AMA. Topics covered in this episode include: - What is the role of drugs in treating diabetes? - What role does exercise play in managing glucose and insulin resistance in diabetics? - Is there a role for exercise in managing insulin resistance? - Should exercise be used in conjunction with diabetes care? - How do we know if exercise interventions are safe and effective? - Are they safe in non-diabetics and non-insulin populations? - Can exercise interventions be helpful in managing diabetes?
Transcript
00:00:00.000
Hey everyone, welcome to a sneak peek, ask me anything or AMA episode of the drive podcast.
00:00:16.500
I'm your host, Peter Atiyah. At the end of this short episode, I'll explain how you can
00:00:20.460
access the AMA episodes in full, along with a ton of other membership benefits we've created,
00:00:25.440
or you can learn more now by going to peteratiyahmd.com forward slash subscribe.
00:00:31.140
So without further delay, here's today's sneak peek of the ask me anything episode.
00:00:39.220
Welcome to ask me anything episode 26. I'm once again joined by Bob Kaplan. In today's episode,
00:00:46.540
we field a bunch of questions that are kind of follow-up questions to topics we've
00:00:51.480
recently done deep dives into. The two main themes are more follow-up on CGM and more follow-up on
00:01:00.440
zone two. Now, when Bob first told me that those were going to be two of the things we're going to
00:01:04.920
talk about, my first thought was, we have nothing else to say on these topics. We have spent so much
00:01:09.380
time on them, but it turned out I was wrong. There were a lot of really good questions here
00:01:13.260
and things I hadn't considered. So if you find yourself wanting on either of those topics,
00:01:18.280
this episode is definitely for you. If you think you know everything on those,
00:01:22.200
it's probably still for you. The time of the recording here was in the middle of the Tour de
00:01:26.340
France. So we had a little digression on performance physiology and cycling. Now, remember if you're a
00:01:31.900
subscriber and you want to watch the full video of the podcast, you can find it on the show notes page.
00:01:37.160
If you're not a subscriber, you can watch a sneak peek of this video on our YouTube page.
00:01:41.440
So without further delay, I hope you'll enjoy AMA 26.
00:02:06.720
Yes. Yeah. To say the least. I don't know if we'll get through everything,
00:02:10.460
but we did get a lot of follow-up questions on CGM based on our previous, actually, I think it was
00:02:18.840
the previous AMA where we talked about glucose. We talked about mean glucose, glucose variability,
00:02:23.660
and glucose spikes. And then we also had that Sunday email on CGM and non-diabetics. It was
00:02:30.800
related to a JAMA perspective talking about it. So there's a bunch of questions on that.
00:02:36.380
We also have a couple of questions on Aura, the sleep wearable, the Aura ring. We've got some
00:02:43.820
exercise-related questions, and we may work in one or two additional questions if we have time.
00:02:52.860
Okay. So Peter, since we have a lot of these CGM-related questions, one of the things that
00:02:59.320
I've heard you talk about is you have a framework for interventions that I think will be really
00:03:03.300
helpful in laying the foundation for how you think about CGM and their use in different
00:03:08.920
populations. So can you start off by telling us a little bit about that framework?
00:03:14.180
Yeah. So again, it doesn't pertain to CGM specifically. It pertains to anything
00:03:19.440
that comes across my plate. And the first time I actually contemplated this was really when I
00:03:25.780
started trying to look critically at the data around meditation. So this was probably about
00:03:31.280
maybe eight to 10 years ago. But anyway, it basically asks a series of questions. So the
00:03:36.320
first question is, what is the risk of harm from doing this thing? That's a direct question. So if you
00:03:42.580
do X, how high is the probability of harm? The second question is obviously the contrapositive
00:03:50.540
of that. If you do X, what is the probability of benefit? And then the third question, and by the
00:03:59.320
way, before I go to the third question, those first two questions are so obvious that they're almost not
00:04:04.060
worth stating. And of course that's mirrored in the way the FDA organizes drug trials, right? So a drug
00:04:10.960
trial is organized first by, well, after you get through the preclinical data, the animal work,
00:04:17.060
you know, after the IND has been filed, your first trial in humans, which is called a phase one trial
00:04:21.980
is looking at harm. I mean, it's typically a small trial with dose escalation that is only trying to
00:04:30.760
understand if as you escalate the dose, do you see an increase in side effects? Very occasionally,
00:04:37.500
you see some benefits in a phase one trial. And if you do, that's interesting, but you generally
00:04:42.100
can't take it to the bank because the study is so small and generally it's quite homogeneous.
00:04:47.600
So that's when you move on to phase two studies, which are geared towards efficacy, i.e. is this thing
00:04:53.360
doing good? And of course, if the phase two trial is positive, you move to a much larger trial called
00:05:00.280
a phase three trial, which really doubles down on efficacy. Of course, both of these trials will
00:05:06.040
continue to be able to pick up any signal of harm. So you're always in the spirit of trying to capture
00:05:11.060
that. But the real point here is you're raising the bar, so to speak, for what you're demanding of
00:05:16.880
this. So again, what's the risk of harm? What's the probability of benefit are two obvious questions.
00:05:22.640
I think the third question then is what's the opportunity cost of this intervention? And I feel
00:05:29.220
like I've talked about this on a previous podcast before, and maybe it was even one of ours, but
00:05:32.660
there was this device, sort of a device that you would listen to and it would supposedly put you
00:05:40.520
in a trance. And the company that was proposing this thing had all sorts of theoretical benefits
00:05:47.160
from using it. If you listen to this device, you were less likely to get breast cancer and all of
00:05:52.780
these other things. So was there any harm in this device? As far as I could tell, no, I really didn't
00:05:59.760
think that listening to this device was harmful in any way. Was there any benefit of this device?
00:06:08.060
Certainly not to the extent that they made claims. That said, I had tried the device because a friend
00:06:14.180
of mine bought it for me. And I have to admit, it was the most relaxing thing I'd ever done. In fact,
00:06:19.360
virtually every time I tried it, I fell asleep. So maybe there was some good in that. Maybe there was
00:06:24.060
some bad in that if it was a daytime nap and you could speak to the disadvantages of daytime napping,
00:06:29.580
but there was an opportunity cost to it. And I don't just mean financial. So the device was
00:06:34.120
pretty expensive. I want to say it was like a thousand bucks if you were going to buy the thing.
00:06:38.580
Well, a thousand bucks is no trivial sum of money for anyone. So that's obviously something that has
00:06:44.360
to be weighed against what else could be done with an opportunity cost. But the other thing you have
00:06:48.620
to keep in mind is their prescription for use was two 20 minute sessions a day, much along the lines
00:06:55.220
of like transcendental meditation, which is similar, but has much better data. And that's where I kind
00:07:01.700
of thought, well, there's a problem because for most people who are super busy, 40 minutes a day
00:07:07.200
for very questionable benefit didn't make a lot of sense if it came at the expense of other things that
00:07:16.680
undoubtedly had benefit such as, could that be 40 minutes a day of actual meditation? Could that be
00:07:23.900
40 additional minutes a day of sleep? Could that be 40 minutes a day of exercise? All things that I
00:07:29.600
would point to as having far greater evidence in favor of. So I think anytime you're thinking about
00:07:36.620
doing something, you want to kind of go through that. And those are especially important questions to
00:07:43.940
be asking when the answer is not readily apparent from RCTs that have generally already answered
00:07:52.400
one and two. Now remember, many RCTs are the easiest RCTs to do are the ones that are based on
00:07:58.980
pharmacology. And they're generally addressing one and two, but they're not really addressing three
00:08:04.680
because there really isn't much of an opportunity cost to taking a pill outside of the economic cost.
00:08:09.360
But the time cost of it is relatively low. Of course, when it comes to RCTs that are more
00:08:15.620
intervention-based, such as exercise, yes, you want to be able to think about this. But as you look to
00:08:21.420
something like CGM in the case of non-diabetics, this framework to me is very helpful because at this
00:08:28.780
time, we don't have great RCTs to point to that say in people who are not yet diabetic, there is a
00:08:36.240
benefit to using CGM. So again, as you go through that, you ask yourself the question, what is the
00:08:41.740
risk of harm? And again, when we talk about CGM specifically, I think the risk of harm is very low.
00:08:48.720
If we were going to speculate what could be harmful about it, well, I think the most obvious thing that
00:08:53.320
comes to my mind is anxiety that it can stoke, right? It can create obsession in someone. And
00:09:00.040
certainly I can speak to that personally. I don't think personally I've found it harmful,
00:09:05.020
but I could absolutely understand it. And I frankly think we have some patients in whom
00:09:10.380
I've never recommended it. So for example, we have some patients who have a history of eating
00:09:15.200
disorders. These are patients I would not in any way, shape or form advocate the use of CGM.
00:09:21.620
I think it's yet another tool that can create a negative cycle around obsession. Is there any chance
00:09:29.220
of it doing good? Well, I mean, I think I've already made the case for that. I think there's ample
00:09:33.200
chance for it doing good on two fronts, right? The first front is what I call insight-based good,
00:09:40.740
which is teaching you what your carbohydrate tolerance is. That's what the tool is for.
00:09:47.280
And then secondly, what I would call behavioral good or behavior modification, which is effectively
00:09:53.380
a strapped on version of the Hawthorne effect. So when you're wearing a CGM, you're basically
00:09:59.420
utilizing a tool that is monitoring you. And there is no shortage of data to support the idea that when
00:10:05.940
people are asked to monitor food intake, they make changes in the right direction. So if I said to you,
00:10:13.040
Bob, I want you to record everything you eat for the next month and track it in a food diary,
00:10:20.320
are you going to make better food choices than you are making now? And the answer is unequivocally,
00:10:25.820
yes, you are going to do that. Which is also not to pile on, I guess this is the first time I'm
00:10:31.740
mentioning observational epidemiology so far, at least for this episode. But that's one of the
00:10:36.620
challenges right there with food frequency questionnaires is that they start asking them
00:10:41.320
to ask them what foods they're eating or they ask them what foods they ate. And oftentimes just that
00:10:46.820
by asking, they're changing their behavior. And so you're not really getting an accurate
00:10:50.860
representation of what they ate previously. Right. So that's a problem that plagues epidemiology,
00:10:56.580
but you can use that to your advantage. Right. So, you know, and we do that stuff clinically,
00:11:01.660
right? This is how you create accountability for patients. You say, look, we're going to check in
00:11:07.220
once a day and I just want you to tell me what you ate. And, and, and even if you provide no other
00:11:14.640
instruction, which is, oh, I want you to have this many grams of protein and this many grams of carbs
00:11:18.880
and this many grams of fat. No, no. Even if you don't go to that level, if you just say, I just
00:11:22.420
want you to tell me what you ate, that level of accountability immediately changes a person's
00:11:26.780
behavior. And that's an example of how you can use that to your benefit. I guess I'll get to the third
00:11:31.720
point, which is opportunity costs. So what's the opportunity cost of one of these devices? Well,
00:11:36.480
I think hands down the biggest opportunity cost is the economic cost. These things are not cheap.
00:11:42.200
And if you are not diabetic, you are not going to have your insurance company cover one of these
00:11:46.820
devices. So there are no shortage of companies out there that are repurposing and repackaging
00:11:53.620
CGM. So there are really three companies that make CGMs in the clinical grade. So there's Medtronic,
00:12:01.620
Abbott and Dexcom. And again, by way of full disclosure, I consult with Dexcom, not on their CGM
00:12:08.960
business, but actually on a part of their business that deals with other analytes. So other things that
00:12:14.340
you could measure. So I actually don't really interact much on the CGM side, but those three
00:12:19.640
companies make CGMs. And then there are lots of companies like Levels and Super Sapiens that people
00:12:25.700
have heard of who are plugging in those CGMs into their apps to help users with their goals, be it
00:12:32.380
weight loss or otherwise. Well, they're not cheap. I'm trying to remember what the cost is. The monthly,
00:12:38.720
I mean, I feel like the daily cost of CGM is about 10 bucks and there's probably cheaper ways to get
00:12:44.020
it if you're buying your CGM on eBay or if it's a little bit expired, but directionally speaking,
00:12:48.420
it's about a $10 a day habit. That adds up, right? But that's, you know, call it $3,500 a year. That's
00:12:54.520
a huge expense, assuming you need it every minute of every day. And I don't think you do. I think you
00:12:59.640
can gain a lot of insight using these things periodically. I don't think this is something you
00:13:03.560
need to be tethered to every minute of every day. There are some people like me who enjoy that.
00:13:08.200
Not going to lie. I've been wearing CGM for almost six years. It never gets old to me. I continue to
00:13:14.720
find insights that just provide value. And more than anything else, it's really the behavioral
00:13:20.840
tool. And that might make me a mental midget. I might just be a guy that is such a simple plebe
00:13:26.840
that having that little CGM on my arm is what keeps me away from the ice cream in the freezer,
00:13:31.900
in the cookies, in the pantry. But if you saw my freezer and you saw my pantry, you would certainly
00:13:36.140
understand why I stand to benefit from using CGM. I think you mentioned this in the article or the
00:13:42.980
Sunday post about how the percentages change with when you talk about how much insight you're getting,
00:13:49.660
say, in the first 30 days versus long-term, whether it's a motivational or a behavioral tool
00:13:54.620
and how those can shift. But I also want to mention that when I was at UVM, I was getting my undergrad and I
00:13:59.920
was studying nutrition that, to your point about ice cream in the freezer, I remember for one of our
00:14:06.300
courses that we were taking, we started doing a food log or a food diary. And I remember at the
00:14:10.160
time I was eating, I don't know why I got into this, but actually you probably will know why.
00:14:14.780
I was eating a pint of Ben and Jerry's chocolate chip cookie dough, just religiously.
00:14:19.220
Is a pint the one that's this big? Like the little one? Is that a pint?
00:14:23.260
That's the 16-ouncer. Yeah. But I think that's the only thing that Ben and Jerry's comes in.
00:14:29.240
Okay. Okay. And you were eating one of those a day?
00:14:31.240
I was eating one of those a day after dinner and it was just like a, yeah, whatever kind of habit.
00:14:37.200
Then I was doing the food log and I had to keep a food log. And that's how I kicked that. If you
00:14:42.780
want to call it a habit, that's how I kicked the Ben and Jerry's habit. I said, look, I'm going to
00:14:45.600
have to write, you know, for the, I think it's two days during the week and one day during the
00:14:49.660
weekend that I had to do it. And that's, I think, typically what they do for food diaries. And you
00:14:53.440
just log everything you eat, you know, real time over 24 hours for like a Monday and a Thursday and
00:14:58.420
a Saturday. And I remember I, I kicked that habit, but it just, it reminds me of this stuff that
00:15:03.980
some of these things are behavioral tools, maybe in disguise. But I think you mentioned something
00:15:09.600
like 90, 10 at the beginning with your. Yeah. Yeah. And that, that's sort of what I tell patients.
00:15:13.720
Cause look, a lot of our patients are like, Hey Peter, what do you think? I mean, do you,
00:15:17.080
do you think I should do this? And my answer for most of them is, yeah, I think you should.
00:15:20.040
I think everybody deserves a three month trial of CGM. Again, notwithstanding a handful of patients
00:15:25.000
who I think have contraindications. And I typically say, look, it's going to start out as about 90,
00:15:31.440
10, 90% of this is going to be insight. Like you're going to be going, Holy cow. I can't believe
00:15:38.360
fill in the blank. And only 10% of it is going to be changing your behavior through this Hawthorne effect.
00:15:45.000
I said, by the end of that 90 days, that's going to flip again. It depends on how much
00:15:50.500
insight you look to extract, but directionally within about three months, you're going to be
00:15:56.240
like, you know what? I've sort of figured out the effect of grapes and the difference between like
00:16:01.300
a grapefruit and a banana. Like I've, I've kind of got that dialed. And I've also figured out that
00:16:06.800
I can eat a ton of carbs after I work out. But if I eat a ton of carbs before bed,
00:16:12.140
totally different effect on my blood glucose and over the overnight and in the morning,
00:16:16.620
but the shift is you start to gamify it a little bit. So yeah, I guess that's sort of what I would
00:16:23.540
say on, on my framework for how to think about these things. Okay. So a related question to CGM.
00:16:31.060
I've got, actually, you just mentioned grapes. So this, this person says, my glucose spikes when I
00:16:34.740
eat some fruits, but not others. Do you know why this happens? And does it mean that I should
00:16:38.860
avoid fruits that spike my glucose? Thank you for listening to today's sneak peek AMA episode of
00:16:44.580
the drive. If you're interested in hearing the complete version of this AMA, you'll want to
00:16:48.940
become a member. We created the membership program to bring you more in-depth exclusive content without
00:16:54.700
relying on paid ads. Membership benefits are many and beyond the complete episodes of the AMA each
00:17:01.060
month. They include the following ridiculously comprehensive podcast show notes that detail
00:17:06.300
every topic, paper, person, and thing we discuss on each episode of the drive access to our private
00:17:12.740
podcast feed, the qualies, which were a super short podcast, typically less than five minutes
00:17:18.100
released every Tuesday through Friday, which highlight the best questions, topics, and tactics
00:17:22.920
discussed on previous episodes of the drive. This is particularly important for those of you who
00:17:28.420
haven't heard all of the back episodes becomes a great way to go back and filter and decide which ones
00:17:33.840
you want to listen to in detail, really steep discount codes for products I use and believe
00:17:38.560
in, but for which I don't get paid to endorse, and benefits that we continue to add over time.
00:17:44.280
If you want to learn more and access these member-only benefits, head over to
00:17:47.900
peteratiamd.com forward slash subscribe. Lastly, if you're already a member, but you're hearing this,
00:17:54.860
it means you haven't downloaded our member-only podcast feed where you can get the full access
00:17:59.080
to the AMA, and you don't have to listen to this. You can download that at peteratiamd.com
00:18:05.220
forward slash members. You can find me on Twitter, Instagram, and Facebook, all with the ID
00:18:11.660
peteratiamd. You can also leave us a review on Apple Podcasts or whatever podcast player you listen on.
00:18:18.560
This podcast is for general informational purposes only and does not constitute the practice of
00:18:23.500
medicine, nursing, or other professional healthcare services, including the giving of medical advice.
00:18:29.700
No doctor-patient relationship is formed. The use of this information and the materials linked to this
00:18:34.960
podcast is at the user's own risk. The content on this podcast is not intended to be a substitute for
00:18:41.380
professional medical advice, diagnosis, or treatment. Users should not disregard or delay in obtaining
00:18:48.400
medical advice from any medical condition they have, and they should seek the assistance of their
00:18:53.540
healthcare professionals for any such conditions. Finally, I take conflicts of interest very seriously.
00:19:00.040
For all of my disclosures and the companies I invest in or advise, please visit peteratiamd.com
00:19:07.000
forward slash about where I keep an up-to-date and active list of such companies.
00:19:18.400
For all of my disclosures, please visit peteratiamd.com forward slash about where I keep an up-to-date,