podcast Peter Attia 2025-03-03 topics

#338 ‒ Peter's takeaways on aerobic exercise and VO2 max, insulin resistance, rising healthcare costs, treating children with autism and ADHD, and strength training | Quarterly Podcast Summary #4

Audio

Show notes

In this podcast summary episode, Peter summarizes his biggest takeaways from the last three months of guest interviews on the podcast. Peter shares key insights from his discussions on diverse topics such as aerobic efficiency and VO2 max with Olav Aleksander Bu; insulin resistance with Ralph DeFronzo; economics of the US healthcare system and cost-saving strategies with Saum Sutaria; diagnosis and treatment of autism, ADHD, and anxiety in children with Trenna Sutcliffe; and strength training with Mike Israetel. Additionally, Peter shares any personal behavioral adjustments or modifications to his patient care practices that have arisen from these fascinating discussions.

If you’re not a subscriber and listening on a podcast player, you’ll only be able to hear a preview of the AMA. If you’re a subscriber, you can now listen to this full episode on your private RSS feed or on our website at the episode #338 show notes page . If you are not a subscriber, you can learn more about the subscriber benefits here .

We discuss:

Overview of topics to be discussed [2:00];
Olav Aleksander Bu Pt.2 episode: metrics to track aerobic efficiency and insights about VO2 max, and the ability of increased carbohydrate consumption to boost performance [4:30];
The best practices for performing a VO2 max test, the differences between VO2 max training and all-out efforts, and the role of energy intake in endurance performance [14:45];
Ralph DeFronzo episode: the pathophysiology of insulin resistance and type 2 diabetes—how it impacts different organs, flaws in conventional diabetes treatment, and more [24:30];
Understanding type 2 diabetes beyond the traditional triumvirate of features: the “ominous octet” describes changes in other organs [31:45];
Pharmacological treatments for insulin resistance and type 2 diabetes [41:30];
The importance of early detection and intervention in insulin resistance [50:30];
Saum Sutaria episode: the economic and systemic drivers of high healthcare costs in the U.S. [54:00];
Reducing health care costs: redefining health insurance, lowering drug prices while maintaining innovation, leveraging AI for efficiency, and more [1:07:15];
Trenna Sutcliffe episode: insights on autism, ADHD, and anxiety in children—definitions and diagnosis [1:11:45];
Exploring the rising prevalence of autism spectrum disorder [1:17:15];
Trenna’s views on caring for children with autism [1:21:15];
Misconceptions around vaccines and autism [1:26:00];
Mike Israetel episode: insights about strength training, minimum effective dose, troubleshooting plateaus, tips for beginners, and more [1:28:15];
Topics Peter is interested in exploring in future podcasts [1:40:15]; and
More.

The Peter Attia Drive

338 ‒ Peter’s takeaways on aerobic exercise and VO2 max, insulin resistance, rising healthcare costs, treating children with autism and ADHD, and strength training | Podcast Summary #4

Show Notes

Notes from intro :
Welcome to another special AMA episode, this is the 4th installment of the podcast summary
In this conversation, Peter discusses what he learned from the last quarter of interviews What he thinks were some of the most important insights Things that have resulted in a change in his work and behavior
Please note, listening to this episode does not remotely constitute a substitute for listening to the actual episodes
His hope that listening to this episode helps you think about refining what you might have learned there, and if something you hear in one of these summaries is of interest and you missed the original episode, you’ll go back and listen to it
Today, we cover interviews with: Olav Aleksander Bu, Ralph DeFronzo, Saum Sutaria, Trenna Sutcliffe, and Mike Israetel
Throughout these, we talk on various topics VO 2 max Power at VO 2 max Insulin resistance Metformin SGLT2 inhibitors GLP-1 agonists US healthcare costs Diagnosis and treatment approaches to autism ADHD and anxiety Zone 2 and resistance training Learning all the effects of anabolic steroids
That sounds like a hodgepodge of topics, but that’s because it’s pulled from all of these discussions
What he thinks were some of the most important insights
Things that have resulted in a change in his work and behavior
VO 2 max
Power at VO 2 max
Insulin resistance
Metformin
SGLT2 inhibitors
GLP-1 agonists
US healthcare costs
Diagnosis and treatment approaches to autism
ADHD and anxiety
Zone 2 and resistance training
Learning all the effects of anabolic steroids

Overview of topics to be discussed [2:00]

Purpose of the Episode :

Review past episodes from the quarter.
Highlight main takeaways, insights, and changes in behavior or practice based on the discussions.
Serve as an augmentation to the original episodes, not a replacement.
Many listeners use this summary episode to revisit key moments in prior episodes.

Topics Covered in the Past Quarter

Olav Aleksander Bu Pt.2 – Training, Performance, and VO2 Max Second appearance on the podcast. Attempted to make the episode less technical than the first, but ended up being equally or more technical. Breakdown of key insights and takeaways from Olav’s expertise.
Ralph DeFronzo – Insulin Resistance, Type 2 Diabetes, and Medications Discussion on Metformin, SGLT2 inhibitors, and other drugs. In-depth analysis of insulin resistance and diabetes management.
Saum Sutaria – U.S. Healthcare System Exploration of why U.S. healthcare costs nearly double per capita compared to other developed nations. Examination of inefficiencies and systemic issues in healthcare.
Trenna Sutcliffe – Autism, ADHD, and Anxiety Review of research and insights into neurodevelopmental conditions. Discussion of anxiety and its relationship to ADHD and autism.
Mike Israetel – Resistance Training Deep dive into resistance training principles. Impact on health, longevity, and overall well-being.
Second appearance on the podcast.
Attempted to make the episode less technical than the first, but ended up being equally or more technical.
Breakdown of key insights and takeaways from Olav’s expertise.
Discussion on Metformin, SGLT2 inhibitors, and other drugs.
In-depth analysis of insulin resistance and diabetes management.
Exploration of why U.S. healthcare costs nearly double per capita compared to other developed nations.
Examination of inefficiencies and systemic issues in healthcare.
Review of research and insights into neurodevelopmental conditions.
Discussion of anxiety and its relationship to ADHD and autism.
Deep dive into resistance training principles.
Impact on health, longevity, and overall well-being.

Peter’s Reflection on the Past Quarter

Volume of Notes Taken: This quarter had the most extensive set of notes Peter has ever taken for a quarterly review. The episodes with Ralph (diabetes) and Som (healthcare) stood out, with 3-4 times the usual amount of notes. Attempted to synthesize the insights but acknowledges the dense nature of these discussions.
Expectation for a Long Episode: Given the richness of the topics covered, this summary episode might be the longest quarterly review yet.
This quarter had the most extensive set of notes Peter has ever taken for a quarterly review.
The episodes with Ralph (diabetes) and Som (healthcare) stood out, with 3-4 times the usual amount of notes.
Attempted to synthesize the insights but acknowledges the dense nature of these discussions.
Given the richness of the topics covered, this summary episode might be the longest quarterly review yet.

Olav Aleksander Bu Pt.2 episode: metrics to track aerobic efficiency and insights about VO 2 max, and the ability of increased carbohydrate consumption to boost performance [4:30]

#331 ‒ Optimizing endurance performance: metrics, nutrition, lactate, and more insights from elite performers | Olav Aleksander Bu (Pt. 2 ) (January 13, 2025)

Peter’s biggest insights from the second episode with Olav

We talked beforehand about trying to make it a little less technical because the first one was pretty technical
Peter could not resist the tractor pull of trying to explain some technical things
Looking back, he was probably more technical
They started the discussion by explaining the difference between things like functional threshold power and critical power

Peter hesitates to bring these up now because he just doesn’t think they’re relevant unless you are a cyclist

⇒ The functional threshold power (or FTP) is the power that you can hold for one hour

One way to test it: get on a stationary bike and ride as hard as you can without blowing up for 1 hour
Whatever the average power is that you held is your FTP
When Peter was a cyclist, he would do this in a 20-minute test and would typically discount it by 10% Olaf suggested discounting by only 5% In Peter’s experience, 10% was necessary
Olaf suggested discounting by only 5%
In Peter’s experience, 10% was necessary

⇒ Critical power is very similar, but rather than it being the power you can hold for an hour, it’s the power you can hold for 30 to 40 minutes

The important distinction here is critical power is much easier to calculate
You can do it from a set of curves that are derived from 3-4 individual tests that are much shorter

Why is this relevant?

If you want to have other metrics beyond VO 2 max for higher end aerobic efficiency , you might want to know your FTP or critical power, and critical power is much easier to measure
These days Peter doesn’t spend a lot of time worrying about his FTP In fact, he doesn’t know what it is
In fact, he doesn’t know what it is

It made Peter realize maybe he ought to do a critical power test just so that he has another benchmark to be tracking

A personal insight about power at VO 2 max (PVO 2 max)

This is the relationship between VO 2 max and PVO 2 max (or vVO 2 max [which is velocity at VO 2 max])
VO 2 max is maximum ventilation, meaning maximum oxygen consumption It’s usually sustained for at least a minute when you’re doing the test It’s the highest amount of oxygen in liters per minute that can be consumed for a minute
This is achieved during a ramp exercise, almost exclusively done on a stationary bike or on a treadmill
And when you hit that VO 2 max, you will note the velocity If you’re on a treadmill, assuming you’re running flat Or if you’re on a bike, then your PVO 2 max is the power that you’ve achieved
Some have argued that vVO 2 max or PVO 2 max are actually more predictive of sports specific performance than just the number VO 2 max
Peter thinks there might be something to that because he shared his numbers with Olaf and the truth of it is he has always had a very low PV0 2 max to VO 2 max Stated another way, he has always had a VO 2 max that is higher than many people who are much better cyclists It wasn’t uncommon when he was training that his VO 2 max was 15 points higher than people who had a higher FTP than him
What Olaf suggested speaks to an inefficiency there Peter may be over-training aerobically and undertraining anaerobically That inefficiency means that he is actually requiring more energy across the board to put out more power
It’s usually sustained for at least a minute when you’re doing the test
It’s the highest amount of oxygen in liters per minute that can be consumed for a minute
If you’re on a treadmill, assuming you’re running flat
Or if you’re on a bike, then your PVO 2 max is the power that you’ve achieved
Stated another way, he has always had a VO 2 max that is higher than many people who are much better cyclists
It wasn’t uncommon when he was training that his VO 2 max was 15 points higher than people who had a higher FTP than him
Peter may be over-training aerobically and undertraining anaerobically
That inefficiency means that he is actually requiring more energy across the board to put out more power

Something interesting that comes from that is there might be an association between people like that and a lesser propensity to gain weight

Every time Peter has done a resting metabolic expenditure test (he’s actually done the doubly labeled water test and the metabolic chambers), he always seems to have a through-the-roof energy expenditure for his body weight

For the first time, that all came together

On the one hand he has an advantage perhaps in that he has a very high energy expenditure So relative to somebody else who eats as much as he does, he’s going to be leaner
The flip side of that is he is actually quite inefficient at utilizing energy
So relative to somebody else who eats as much as he does, he’s going to be leaner

Beyond training, is there anything a person can do to boost their VO 2 max

⇒ Beetroot concentrate is rich in nitrates

The body converts it to nitric oxide and that helps with vasodilation (opens up capillary beds)
Anything that impairs nitric oxide synthase is going to impede it And there are many things that do from homocysteine to insulin resistance
At the elite level, this doesn’t make much of a difference, but in amateur athletes, it can be about a 5% boost [Peter discussed evidence around benefits of nitrates from beetroot concentrate in a newsletter from 2022]
And there are many things that do from homocysteine to insulin resistance
[Peter discussed evidence around benefits of nitrates from beetroot concentrate in a newsletter from 2022]

A strategy that is more gamesmanship than anything else

⇒ As you approach failure in a VO 2 max test, do a few breath holds

That produces a significant boost in VO 2 max
It’s just a reactive overconsumption of oxygen (Peter doesn’t know if that means anything)

Some data that suggests that acetaminophen use can boost peak endurance performance by 1-2%

Kristian and Gustav (the 2 world-class triathletes Olav trains) are not using this
Olav raised an interesting point: while acetaminophen (or Tylenol) can obviously reduce the perception of pain (which has been one of the arguments for why it boosts performance), it may impair the athlete from giving off heat (from heat dissipation)
He questioned whether or not that would be a worthwhile trade-off
For Peter, that begs the desire to do an experiment and find out

What is the upper limit of carbohydrate consumption while doing cardio activity and racing ?

This is relevant to high performance athletes (a triathlete, cyclist, or ultra marathon runner)
If you do the math, energy expenditure of doing an Ironman is somewhere between 700-1000 kcal/hour (depending on your fitness level) Let’s say you’re going easy, you’d be at 600-700 kcal/hour
We don’t have that much stored glycogen You’ve got maybe 50, 100, 200 g of glycogen stored (if you have really big muscles)
That’s going to supply you for maybe 1.5 hours You’re going to run out pretty quickly
Thereafter, you have to meet your needs from body fat and intake of carbohydrates
The conventional thinking has always been that you can only consume about 60 g of glucose per hour And 60 grams of glucose, of course, is only about 240 kilocalories
This has always been one of Peter’s main arguments for why being fat adapted is very important because if you’re consuming that 60 of glucose and that’s giving you whatever it’s giving you, you have to get the balance from fat and you’re only going to do that if you’re heavily fat adapted So you get into this cycle
Let’s say you’re going easy, you’d be at 600-700 kcal/hour
You’ve got maybe 50, 100, 200 g of glycogen stored (if you have really big muscles)
You’re going to run out pretty quickly
And 60 grams of glucose, of course, is only about 240 kilocalories
So you get into this cycle

“ The biggest innovation in endurance sports like the Tour de France and Ironman over the past decade is the amount of glucose that these guys are able to consume .”‒ Peter Attia

Olav talked about numbers Peter had never heard before
Recently, Peter interviewed Tadej Pogačar , the greatest cyclist on the planet
His numbers blew Peter’s mind of the type of carbohydrate intake that he was tolerating: 150, 180 g/hour

Olaf said that they’re now pushing triathletes at the world-class level to 240 g/hour, and what that basically tells you is you can meet all of your glycolytic needs indefinitely through that

⇒ You have to train this system

These are athletes that consume gels and eat carbohydrates at a 12% mixture 12% mixture means 120 grams per liter 10% mixture is 100 grams per liter, etc.
Conventional wisdom is that our gastrointestinal system cannot tolerate more than a 5% mixture
Personally, when Peter used to do ultra-distance stuff, he had a hard time going above 5-6%
12% mixture means 120 grams per liter
10% mixture is 100 grams per liter, etc.

You can train your GI system to utilize a 12% carbohydrate mixture

How these guys are drinking 2 liters per hour of a 12% mixture (which would be 240 grams), Peter simply can’t fathom
But clearly, that’s what they’re doing or they’re doing some combination of consuming gels plus water that amounts to that mixture
When you look at the hyperbolic performance of endurance athletes today, it’s very quick and tempting to think they’re using drugs we haven’t figured out yet

Another explanation is that they literally figured out how to double the octane of the fuel

It’s like a car that went from racing at 70 octane to 140 octane Peter’s not even sure if there’s a 140, but you know what he’s getting at
Peter’s not even sure if there’s a 140, but you know what he’s getting at

The best practices for performing a VO2 max test, the differences between VO2 max training and all-out efforts, and the role of energy intake in endurance performance [14:45]

Protocol to make sure you’re giving it your best when you do a VO 2 max test

This is something Peter has started implementing with patients if they do a VO 2 max test
1 – You want to do a VO 2 max test at the time of day that you would normally train
2 – You want to be well rested, minimize traveling the day before, etc.
3 – The warmup should be: 6 minutes of very, very easy 6 minutes of Zone 2 3 minutes at threshold (or FTP) 2-3x at a 10-15 second burst at about what you expect your PVO 2 max is
4 – A relatively short rest of 10-15 minutes, get a drink, then do the test
6 minutes of very, very easy
6 minutes of Zone 2
3 minutes at threshold (or FTP) 2-3x at a 10-15 second burst at about what you expect your PVO 2 max is

This is a decent warmup

When Nick did VO 2 max testing, he doesn’t recall if they had him do any kind of warmup He kind of hit the treadmill and then started going
This has people go through a little bit of a workout before the test begins
He kind of hit the treadmill and then started going

Peter does most of his VO 2 max testing outdoors now using the VO2 Master device

He leaves the house and rides 10-15 minutes to the place where he does a hill repeat (that’s the warmup) There are a couple of short little climbs where he’ll do 30 seconds of relatively high power just to get up over a little pitch
He will do 2-3 full runs of the hill at escalating power before he’s truly going to hit his max
He’ll do a 4-5 minute up, maybe 85% of what his maximum power would be for that climb, come down for the same amount of rest period
Go up again at maybe 90% of what his maximum power would be, come back down, and then maybe he would go and give it The third one would be up there By the time he’s done it, he’s really warmed up
There are a couple of short little climbs where he’ll do 30 seconds of relatively high power just to get up over a little pitch
The third one would be up there
By the time he’s done it, he’s really warmed up

A patient did his VO 2 max test at a University facility

Peter was surprised what his number was ‒ it was lower than expected
The patient was told to get on the treadmill and just start cranking it
His VO 2 max was determined about 5 minutes into it

Peter’s takeaway, “ That’s a garbage protocol. You were not warmed up and ready to do that .”

Long-term preparation for a VO 2 max test

⇒ Peter has noticed that his performance takes a hit in the summer

He prefers to test VO 2 max in the spring, fall, or winter

In the month leading up to it, so the past month, have you changed your training at all? Are you doing anything in particular for it?

Not at all
This is just a data check

Just as a reminder for people, what does your VO 2 max training look like in a typical week? Is that 1 day a week?

Just 1 day a week
A typical week is 3 days of Zone 2 and 1 day a week of interval training Interval training at that specific 4, 5, upper limit 8-minute intervals
When we talk about VO 2 max training, Nick thinks back to one of the first video podcasts they ever did, with Alex Hutchinson [episode #151]
Interval training at that specific 4, 5, upper limit 8-minute intervals

A lot of times when people think of VO 2 max training (or Tabata training or going all-out), people lose sight of how hard that actually is

VO 2 max training hurts less than a true Tabata
That’s where people have a hard time understanding what all-out means

⇒ Technically, Peter doesn’t think the human body is capable of going all-out for more than 10 seconds

A Tabata is a 20 second effort followed by a 10 second rest repeated 8x
Even at the level of even a 20 second Tabata, there’s a governor that is self-regulating how hard you go
The reverse Tabata where you go 10 seconds all out, 20 seconds rest for 8 rounds, that’s about the closest thing that Peter thinks we’re capable of doing as a truly all-out

You will increase your VO 2 max doing that type of an exercise, but not nearly, not nearly as much as if you’re doing intervals in the 3-8 minute range

⇒ By definition, if you’re doing something for 3-8 minutes, you’re not going all-out

What you’re trying to do is go as hard as you can for that distance and for that time
It’s a different animal
Peter thinks it hurts more because it’s a lower level of peak pain, but it’s spread out over a longer period of time So the area under the pain curve is greater But it’s far from all-out and at any moment in time, the pain is not the same
So the area under the pain curve is greater
But it’s far from all-out and at any moment in time, the pain is not the same

A 4-minute interval

You’re not starting that going as hard as you can and trying to sustain it for 4 minutes; walk through how you think about the energy you put out spread across those 4 minutes

Technically, the power is constant throughout the 4 minutes Peter knows how many watts he wants to produce and the average wattage should over the 4 minutes
Let’s just say he wants to do 4 minutes at 300 watts Of course, you’re outdoors, so you don’t have complete control, it’s always jumping around He’s really watching the 3-second power tracing and the average power to keep it there
It should be really easy for the 1st minutes If you’re dying, you’ve set your target too high
Half-way in (2-2.5 min), Peter still feels pretty good His heart rate is going to be within 5 beats of what its maximum is
The “pain train” starts to leave the station at about 3-3.5 That’s when it really starts to feel miserable
That last minute is really difficult
Peter knows how many watts he wants to produce and the average wattage should over the 4 minutes
Of course, you’re outdoors, so you don’t have complete control, it’s always jumping around
He’s really watching the 3-second power tracing and the average power to keep it there
If you’re dying, you’ve set your target too high
His heart rate is going to be within 5 beats of what its maximum is
That’s when it really starts to feel miserable

If you’ve done this right, when you finish, you’re going to need 4-5 minutes of very, very easy pedaling to let your heart rate come back down before your repeat it

The goal is not to have killed yourself in the interval such that you you can’t do it again

⇒ What you’re trying to do is preserve that power across all the intervals

Follow-up on carbohydrate consumption explaining an increase in performance

How much of a difference do you think that makes?

For people like Peter and most people listening to this podcast, this is not something that should be on our radar He’s not doing a 10-hour endurance event again
If he’s exercising for 2 hours, that’s a long time ‒ and at 2 hours, he’s fine with just water He’s living off his own glycogen and whatever
It’s very difficult now to think about people competing at a world-class level in cycling and Ironman because what Olaf and many others have now argued is the problem of peak endurance is effectively an energetic problem
It’s basically a question of how much chemical energy in the form of food can you convert into electrical energy via the metabolism of food back into chemical energy in the form of ATP back into mechanical energy
He’s not doing a 10-hour endurance event again
He’s living off his own glycogen and whatever

It’s just an energy transfer problem, and more energy input means more energy output

The more logs you can put into the fire, the hotter the fire burns, the more steam it makes, the faster the wheel turns
What we’ve seen over the past decade is quite literally a more than doubling of the feedstock that goes into the furnace

Ralph DeFronzo episode: the pathophysiology of insulin resistance and type 2 diabetes—how it impacts different organs, flaws in conventional diabetes treatment, and more [24:30]

#337 – Insulin resistance masterclass: The full body impact of metabolic dysfunction and prevention, diagnosis, and treatment | Ralph DeFronzo, M.D. (February 24, 2025)

This episode might set the record for the most notes Peter has ever taken in a podcast He actually transcribed his notes twice because he’s trying to make them more condensed
This is one where you just have to listen to this podcast a couple of times
It’s more of a senior level class on the subject
He actually transcribed his notes twice because he’s trying to make them more condensed

Expectations for what was discussed

Focused more the pathophysiology, the organ level organization of type 2 diabetes
We didn’t talk about some of the important tools for addressing insulin resistance, such as diet, exercise, and sleep

“ The most mind-blowing was really around his approach to the pharmacologic management of this .”‒ Peter Attia

Peter is not a diabetologist
When he does have patients with type 2 diabetes, he tends to do the things that Ralph suggests we not do (he’ll explain shortly)_
Peter learned a great deal
They deliberately did not spend oodles of time redefining and explaining at the molecular level exactly what is happening in insulin resistance
Peter did a podcast with Ralph’s understudy, Gerald Schulman [episode #140] , a few years ago where they did a very deep dive on what’s happening at least in the muscle Which is one of the more critical organs and maybe one of the first organs to be affected
Which is one of the more critical organs and maybe one of the first organs to be affected

A brief summary of what’s going on at the molecular level in the muscle insulin resistance

Everybody can picture a little cell in the muscle, a myocyte
Insulin is secreted as a pro-peptide from the pancreas from a ꞵ-cell
This pro-insulin molecule gets broken down into C-peptide and insulin
Insulin makes its way onto the insulin receptor
The insulin receptor then signals something called IRS-1 inside the cell, which then activates a pathway

The PI3K/AKT pathway , which then basically activates a system called PPARγ , that then brings something called the GLUT4 transporter to the surface of the cell [see the asterisks in the figure below]

Figure 1. Insulin receptor (IR) signaling through the IRS-1/PI3K/AKT pathway. Image credit: Expert Reviews in Molecular Medicine 2012

What all that means

The GLUT4 transporter is like a straw Picture a straw that gets poked through the skin of a balloon, but without bursting it The balloon is the cell, the straw is the transporter Now anything on the outside can passively flow into the cell Assuming the gradient is right for that to happen (which it is) Glucose is outside the cell, glucose can now come into the cell
Picture a straw that gets poked through the skin of a balloon, but without bursting it
The balloon is the cell, the straw is the transporter
Now anything on the outside can passively flow into the cell Assuming the gradient is right for that to happen (which it is)
Glucose is outside the cell, glucose can now come into the cell
Assuming the gradient is right for that to happen (which it is)

That’s how insulin facilitates a transport of glucose into a muscle

An aside, and we didn’t talk about this on the podcast

⇒ GLUT4 translocation can also occur and does occur with aerobic exercise

We have what’s called insulin-mediated and non-insulin-mediated glucose uptake
Peter wrote about this quite a bit in Outlive , and he talked about the differences

This is a big part of why exercise is such an important part of glucose disposal, especially for people with diabetes

When a person is insulin resistant, what does that mean?

It means that when insulin binds to the insulin receptor, it kind of gets stuck
For the same level of insulin, it doesn’t kick off the IRS-1/PI3K/Akt/PPARγ pathway
And ever-increasing levels of insulin are needed to actually lead to the uptake in glucose

What is a normal level of insulin?

In a metabolically healthy person eating a well-balanced diet This is a person who’s not carbohydrate-restricted This is a person who’s getting 50 to 55% of their calories from carbohydrates, 15 to 20% from protein, and the rest from fats
This is a person who’s not carbohydrate-restricted
This is a person who’s getting 50 to 55% of their calories from carbohydrates, 15 to 20% from protein, and the rest from fats

Put somebody on a “normal diet,” if they’re insulin sensitive, they are going to be making a grand total of probably less than 35 units of insulin per day

⇒ If you are on exogenous insulin and it’s 40, 50, 60, 200 units of insulin a day, you have to ask yourself, “ What is going on? Why is it that I’m now requiring even more insulin than I would’ve ever made at a healthy state? ”

The answer is because the muscle and the liver, as we’ll talk about, are very insulin resistant

How is all of this measured?

We spent time discussing a technique called an euglycemic clamp (developed by Ralph and his colleagues)
This is not a test you will have ever done
You don’t go to your doctor and get an euglycemic clamp
Those of us who read the literature all the time are familiar with this, and it’s something that is only done in the hospital for a research setting

How the euglycemic clamp technique works

You take a subject and put an IV in each arm
Euglycemic means constant level of glucose
So, in one arm, they start infusing glucose
In the other arm, they start infusing insulin
They are adjusting the glucose intake,t(he concentration of glucose) to hold the serum glucose level constant So if the person shows up and their blood glucose is 100 mg/dL, they will infuse insulin and glucose such that they keep the glucose at 100 mg/dL
And they are raising the insulin dose as they go
So if the person shows up and their blood glucose is 100 mg/dL, they will infuse insulin and glucose such that they keep the glucose at 100 mg/dL

⇒ What they’re looking to see is how much glucose do they need to infuse

The more glucose you need to infuse, the more insulin-sensitive you are

This is where we got into what is happening to each set of cells in the body in response to insulin
For Peter, this was the most eye-opening

The standard definition of type 2 diabetes (the triumvirate)

⇒ What is happening in the liver, in the muscle, in the pancreas (in type 2 diabetes)

The muscle is getting resistant to insulin
When that happens, more and more insulin is required to dispose of the same amount of glucose
The pancreas is getting fatigued, so the ꞵ-cell is having a harder and harder time getting insulin out, and eventually, it begins to fatigue
Then the liver itself is also becoming resistant to the effects of insulin

What should the liver do when insulin goes up?

It should stop making glucose
If insulin is up, it means glucose is up in the periphery, and the last thing the liver should do is undergo gluconeogenesis (making glucose) or even letting glucose into the circulation (hepatic glucose output)

Understanding type 2 diabetes beyond the traditional triumvirate of features: the “ominous octet” describes changes in other organs [31:45]

All of that is pretty logical, but what Ralph has done is taken that and expanded that from what he refers to as the triumvirate to the “ ominous octet ”
In addition to those things that we’ve talked about, let’s add a few more things

The muscle is insulin resistant

Which means that that pathway, that continuum of insulin to IRS-1 to AKT/PI3K/PPARγ, that system is not working
An aside: one of the drivers of that is intramyocellular lipids Inside a muscle, as fat accumulates that is stagnant (that is not being oxidized), that’s going to impair fat oxidation
Inside a muscle, as fat accumulates that is stagnant (that is not being oxidized), that’s going to impair fat oxidation

We talked about what’s happening in the pancreas

The ꞵ-cell is fatiguing And so we are seeing eventually less insulin coming out
And so we are seeing eventually less insulin coming out

We talked about the liver becoming resistant to insulin

That means that even in the presence of high insulin and high glucose, the liver doesn’t care It keeps pumping out glucose, and it keeps making glucose
It keeps pumping out glucose, and it keeps making glucose

We talked about the fat cell

This is interesting, and this is the area where Peter was always confused
There’s a transition
Initially, the fat cell is your friend when you have insulin resistance because what it’s doing is protecting you from the excess glucose It’s taking the excess glucose in as stored fatty acid, which is great because it’s not toxic Remember, while we might aesthetically not love the idea of having big, plump fat cells full of fat, it’s very safe That’s the way to store excess energy
However, over time, that insulin resistance actually leads to something paradoxical, which is increased lipolysis Lipolysis is when the fat cell begins to release fat, and this fat, which gets broken down into free fatty acids, actually impairs insulin secretion
It’s taking the excess glucose in as stored fatty acid, which is great because it’s not toxic
Remember, while we might aesthetically not love the idea of having big, plump fat cells full of fat, it’s very safe
That’s the way to store excess energy
Lipolysis is when the fat cell begins to release fat, and this fat, which gets broken down into free fatty acids, actually impairs insulin secretion

⇒ Free fatty acids impairs insulin secretion by the ꞵ-cells, by disrupting the insulin signal transduction pathway

We also talked about what’s going on with the GI system and the gut hormones

People are going to be very familiar with the idea of GLP-1s and GIP

The GI system is really important, and it has become much more important in the past 5 years as GLP-1 class drugs have become very popular

These are referred to as incretins and the incretin effect
Discussed in the very first podcast [ episode #184 ] Peter did on GLP-1 receptor agonists (4-5 years ago) Talked about the difference in glycemic and insulin response to an oral versus intravenous glucose bolus If you gave somebody 100 grams of oral glucose versus 100 grams of intravenous glucose, you get a totally different response It’s because of GLP-1 and GIP
Talked about the difference in glycemic and insulin response to an oral versus intravenous glucose bolus
If you gave somebody 100 grams of oral glucose versus 100 grams of intravenous glucose, you get a totally different response
It’s because of GLP-1 and GIP

⇒ Ralph pointed out something Peter was not aware of: under normal, healthy circumstances, GLP-1 and GIP account for 70% of insulin release

In other words, 70% of the ꞵ-cell’s response in terms of giving insulin is in response to GLP and GIP Normally, an oral glucose challenge (which is basically anytime you’re getting glucose in your system, since none of us walk around getting intravenous glucose) produces the release of GIP and GLP-1, and they go and tell the pancreas to make some insulin
And while the mechanism is still unclear, the beta cells begin to become deaf to the sound or signal of GLP-1 and to GIP
If that signal becomes weaker, you get less insulin coming out
So it’s a double whammy: at the time when you need more insulin, you’re actually making less
Normally, an oral glucose challenge (which is basically anytime you’re getting glucose in your system, since none of us walk around getting intravenous glucose) produces the release of GIP and GLP-1, and they go and tell the pancreas to make some insulin

The α-cells in the pancreas

We talked about the ꞵ-cells that make insulin, but now we’re talking about the α-cells of the pancreas that make glucagon

⇒ The α-cells make more glucagon in the presence of insulin resistance, which actually promotes gluconeogenesis and it increases hepatic glucose output

So it further exacerbates the insulin resistance that we see in the liver normally

The kidney

We’ve talked about the kidney on previous podcasts [ AMA #49 starting at 42:30], and it’s a pretty amazing in that it’s a tiny organ It’s a couple % of the body weight, but it gets about 20-25% of the cardiac output It’s the most important organ for filtering [the blood]
What happens is, all the glucose that enters your circulation is going to the kidney
It gets filtered out of the kidney and then reabsorbed
We have these transporters called SGLT2 and SGLT1 that were discovered by Raply and his colleagues, and they’re responsible for most glucose absorption
It’s a couple % of the body weight, but it gets about 20-25% of the cardiac output
It’s the most important organ for filtering [the blood]

⇒ SGLT2 is responsible for 90% of glucose reabsorption (which is about 180 g per day of filtration)

If you have type 2 diabetes, you would like to see these things go down You would want to see the SGLT2 transporter fail so that you reabsorb less
But what happens is the exact opposite
You would want to see the SGLT2 transporter fail so that you reabsorb less

In type 2 diabetes, the SGLT2 transporter is upregulated, and a patient with diabetes paradoxically absorbs even more glucose

Everything about this system just doesn’t make sense fundamentally
It’s like the body initially does the right thing and then completely flips
Of course, this explains why we have drugs that are highly effective, called SGLT2 inhibitors

The final organ we introduced to this “ominous octet” is the brain

Here our friend GLP-1 revisits because now the individual in the presence of insulin resistance is having neurotransmitter dysfunction

⇒ Insulin resistant individuals are having GLP-1 resistance, amylin resistance, leptin resistance

All of these things, taken together, mean that the brain is no longer responding to satiety signals

That means we eat more than we should We eat even when we don’t metabolically need to eat
This is a very important point to understand because we can very easily look at an individual who’s overweight and say, “ Why the hell are they eating? ” Well, they’re eating because they’re hungry Even though you may think they have 100 lbs. of fat (that should feed them forever) If they are insulin resistant, the brain doesn’t appreciate the storage form of energy
We eat even when we don’t metabolically need to eat
Well, they’re eating because they’re hungry
Even though you may think they have 100 lbs. of fat (that should feed them forever)
If they are insulin resistant, the brain doesn’t appreciate the storage form of energy

When you take all of this together, you start to realize there’s a lot going on here

Peter adds, “ One of the most important takeaways from this podcast for any clinician is if you have eight organs that are involved in something, you have to be pretty optimistic if you think just one drug is going to be successful. ”

Peter says this in the context of we’ve tried everything else

We’ve tried increasing exercise
We’ve improved sleep
We’ve tried non-pharmacologic interventions, caloric restriction
But the individual is still diabetic or pre-diabetic

One more point, before getting into the pharmacology side of this

⇒ It’s well-understood and well-documented that insulin resistance exacerbates significantly cardiovascular disease

Peter was unaware of the following point: the Akt/PI3K/PPARγ pathway, that is so critical in insulin resistance, is also critical in the regulation of nitric oxide synthase, which regulates nitric oxide production , which improves vascular health

One reason that we see a significant increase in cardiovascular mortality with type 2 diabetes (and even frank insulin resistance) is the inhibition of the nitric oxide pathway, which inhibits vasodilation, thereby allowing vasoconstriction to increase the risk of vascular disease

Peter’s summary on understanding insulin resistance in these 8 tissues

Peter has said a lot here, but this is not a substitute for listening to the podcast
Maybe there are a lot of people listening to this that are like, “ I don’t care. Just tell me what to do. ”
You have to go through that to understand why some of the pharmacological stuff makes sense
Going through that, you realize you’ve got a problem in the muscle where it’s developed insulin resistance because of intramuscular lipid
You’ve got a pancreas that fatiguing and not listening to the gut
You’ve got the brain that’s resistant to the gut
You’ve got the liver that is paradoxically making more glucose than it should be When it should be shutting it off, it’s making it
You’ve got a kidney that’s reabsorbing more of it [glucose] than it should
It’s like everything is broken
If you’re going to tackle that pharmacologically, you’re going to need to combine therapies
When it should be shutting it off, it’s making it

“ You need to tackle the foe by addressing more than one leg of the stool at a time .”‒ Peter Attia

Pharmacological treatments for insulin resistance and type 2 diabetes [41:30]

One of the more important drugs in this discussion is metformin

⇒ Metformin works by impairing hepatic glucose output, but it does not increase insulin sensitivity

A very important thing that Peter learned in this podcast

In the past, Peter has raised concerns personally about metformin, “ Maybe we shouldn’t be using this except in people with diabetes who are significantly insulin resistant who can’t or won’t exercise .” Because it mucks around with the mitochondria, impairing the electron transport chain
Because it mucks around with the mitochondria, impairing the electron transport chain

Ralph pointed out that metformin can only get into a cell through an organic cation transport that is only found in the liver

Ralph sent some papers that showed this [he emailed the slide below]

Figure 2. PET studies of labeled metformin show its uptake by the liver and kidneys . Image credit: 2015 ADA meeting abstract #128-LB

That’s quite interesting because it means that when we see that slight elevation in lactate (both resting and produced lactate) in patients taking metformin, that’s hepatic production of lactate By the way, that might not be a bad thing at all We know lactate is a preferred fuel [discussed in episode #312 with George A. Brooks]
By the way, that might not be a bad thing at all
We know lactate is a preferred fuel [discussed in episode #312 with George A. Brooks]

Peter explains, “ A big takeaway for me was that these organic cation transporters , they’re called OCT-1 (which are found in the liver) and OCT-2 (found in the kidney), means that basically those are the only two cells that are taking in metformin. ”

Metformin is taken in by the liver, where it exerts its effect on minimizing hepatic glucose output
And then it’s excreted by the kidney, and that’s the end of it
Peter has looked through the tracer studies
Ralph argues that the weight loss that is seen in metformin is mostly due to the GI issues, nausea, and slight reduction in appetite (not due to insulin sensitivity) The average weight loss is about 6 lbs.
The average weight loss is about 6 lbs.

The biggest surprise by far was pioglitazone

Pioglitazone (brand name Actos) is a drug Peter has never prescribed Because nobody seems to prescribe it anymore
It is an insulin-sensitizing drug, and it does a whole bunch of good things
But it does 2 things that nobody seems to like, and for that reason, it has largely fallen out of favor That doesn’t mean there aren’t people listening who take Actos or pioglitazone Instead, it’s no longer considered a 1st-line or 2nd-line or even a 3rd-line drug
Because nobody seems to prescribe it anymore
That doesn’t mean there aren’t people listening who take Actos or pioglitazone
Instead, it’s no longer considered a 1st-line or 2nd-line or even a 3rd-line drug

The good things it does

It’s a very inexpensive pill
The first thing it does is lower triglycerides
It lowers blood pressure
It lowers HbA1c
It raises HDL-cholesterol
It reduces hepatic glucose production
Generally, these are all good things

The 2 things people don’t seem to like

1 – People seem to get a little fatter
Ralph argues that what it’s doing is redistributing fat from the places where you don’t want it (in the liver and in the muscle and around the viscera) to the places you do want it (in the subcutaneous space)
So we go from being an insulin-resistant phenotype to an insulin-sensitive phenotype
2 – The other thing is it causes vasodilation, and if you have a patient who’s on the cusp of heart failure , that’s a really bad thing

But it might be that the weight gain and the vasodilation are just as much a part of pioglitazone actually fixing the PPARγ mechanism that we talked about

Remember, that’s what’s driving nitric oxide synthase
That’s what’s leading to the production of more nitric oxide
So it could be that while pioglitazone does all of these good things, the 2 not-so-good things it’s doing are masquerading as bad things, but they’re actually redistributions Meaning, as we see more nitric oxide production, we see more vasodilation That, in the short run, manifests itself as more fluid retention and, obviously, the redistribution of fat
Meaning, as we see more nitric oxide production, we see more vasodilation
That, in the short run, manifests itself as more fluid retention and, obviously, the redistribution of fat

⇒ What Ralph maintains is that you don’t ever want to use pioglitazone in isolation and that what you want to do is use pioglitazone in combination with at least 1 (if not 2) other drugs that will address its shortcomings

Some of those other drugs to use with pioglitazone

1 – The SGLT2 inhibitor
Now that you understand what the kidney is doing, especially in the insulin-resistant patient (or the diabetic patient) where the SGLT2 transporter is actually doing more glucose reabsorption, it clearly makes sense why you would want a drug that inhibits that
This will get more glucose out and with it more fluid
2 – The GLP-1 agonists
We are now on our 4th generation of GLP-1 agonists with tirzepatide Prior to that, semaglutide Prior to that, liraglutide Prior to that, exenatide was the first one
Peter has no experience with exenatide
The first GLP-1 agonist he ever prescribed was in 2014, and that was liraglutide It was effective but not that effective, and certainly not for weight loss (which is what he was trying to prescribe it for)
Prior to that, semaglutide
Prior to that, liraglutide
Prior to that, exenatide was the first one
It was effective but not that effective, and certainly not for weight loss (which is what he was trying to prescribe it for)

What Ralph argues is that no patient should ever be on these drugs in isolation and that what you really want to do is be on triple therapy

⇒ Ralph’s go-to first-line triple therapy is metformin, pioglitazone, and exenatide

The reason he uses exenatide is that it’s far less expensive, and it also doesn’t have the weight loss effects if an individual is not overweight

Other options for triple therapy

If a person has very favorable insurance, you might instead go with metformin, pioglitazone, and a new GLP-1 agonist
Or you might go with metformin, pioglitazone
If the diabetes is quite severe, you would add a SGLT2 inhibitor along with, or plus or minus, a GLP-1 agonist, including a more modern one

The Qatar Study

Because Peter’s mind was being so blown by this, Ralph shared a study called the Qatar Study Performed in the country Qatar
The Qatar Study looked at about 220 people who were very poorly managed type 2 diabetics
They were on metformin and sulfonylurea , but they were failing therapy Sulfonylurea is a class of drug we also talked about on the podcast
Their HbA1c was 10% on average 10% is very, very poorly managed type 2 diabetes These are people whose lifespan is dramatically shortened, not to mention their quality of life When you have a HbA1c of 10, you’re going to be going blind You’re going to be getting amputations You can’t have erections Everything is broken in your body
These patients who were on metformin and sulfonylurea but clearly not responding to therapy were randomized to either a combination therapy of just pioglitazone and exenatide Not even staying on the metformin or intensive insulin therapy
You may think that both metformin and intensive insulin therapy should lower your glucose, but only the former is addressing the problem
Performed in the country Qatar
Sulfonylurea is a class of drug we also talked about on the podcast
10% is very, very poorly managed type 2 diabetes
These are people whose lifespan is dramatically shortened, not to mention their quality of life
When you have a HbA1c of 10, you’re going to be going blind You’re going to be getting amputations You can’t have erections Everything is broken in your body
You’re going to be getting amputations
You can’t have erections
Everything is broken in your body
Not even staying on the metformin or intensive insulin therapy

How each drug acts to address a problem with insulin resistance and type 2 diabetes

Pioglitazone is attacking the insulin resistance
Exenatide is amplifying GLP-1 to overcome the GLP resistance
Insulin therapy is not going to fix any of that It’s just going to hammer more insulin into the system
It’s just going to hammer more insulin into the system

Results from the Qatar study of combination therapy

The combination therapy group at 1 year ended up with a HbA1c of 6.1% ± 0.1%
Whereas the intensive insulin therapy group was still actually diabetic at 1 year at 7.1% There’s many reasons for that When you’re using a hammer-like insulin, you also have to be wary about hypoglycemia It certainly would’ve been possible for those patients to have had a lower glucose level (a lower HbA1c level), but you would’ve compromised their safety in other ways
There’s many reasons for that
When you’re using a hammer-like insulin, you also have to be wary about hypoglycemia
It certainly would’ve been possible for those patients to have had a lower glucose level (a lower HbA1c level), but you would’ve compromised their safety in other ways

⇒ All of this is to say that if those studies are run long enough, the group receiving combination therapy is also going to see a reduction in their risk of diabetic comorbidities, whereas the intensive insulin therapy group will not

We know this from study after study that when you use intensive insulin therapy, you correct the microvascular deficits (i.e., the deficits that accrue to high levels of glucose), but you do not correct the macrovascular deficits You do not reduce the risk of cerebrovascular disease and the risk of coronary artery disease
You do not reduce the risk of cerebrovascular disease and the risk of coronary artery disease

Peter’s final thoughts about the episode with Ralph DeFronzo

Even before this podcast came out, he sent the unedited recording of it to his entire clinical team and said, “ Listen, guys, maybe we’re taking a little too long to move to pharmacotherapy for recalcitrant insulin resistance. ”

Mainstays of therapy for managing hyperinsulinemia, insulin resistance, and even type 2 diabetes

⇒ Even though we didn’t discuss it on the podcast, Peter still believes that the primary mainstays of therapy should be exercise, caloric restriction, and sleep

He no longer feels it is ethical to sit around and wait 1-2 years for a patient to get better when we have a suite of drugs that can be used to augment what they’re trying to do from a lifestyle perspective

The importance of early detection and intervention in insulin resistance [50:30]

Follow-up question: opinion on using metformin in non-insulin resistant folks for geroprotection

You talked about muscle, and updated thoughts [on metformin not getting into muscle]

It doesn’t really change his thought on that
If anything, this information makes it harder for Peter to imagine that metformin is geroprotective (it’s still TBD)
If metformin is geroprotective, wouldn’t that imply it needs to be able to get into all sorts of other cells as well?
Or do we believe that all of its geroprotector is coming through its impact on the liver? (that’s possible)

Peter doesn’t think we have enough information either way to say that

Follow-up on the importance of not letting yourself become insulin resistant

Not only is there complexity with how you treat it
You can see how many things being insulin resistant affects
You should be even more aggressive in making sure you don’t get to an insulin-resistant state because of how many negative consequences there are Even though if you would have asked Peter 6 months ago, he would’ve told you he’s already a 10 out of 10 in terms of how aggressive he is on this
Even though if you would have asked Peter 6 months ago, he would’ve told you he’s already a 10 out of 10 in terms of how aggressive he is on this

Peter was talking to a patient this morning

He’s in his mid-30s (young and healthy) and he ostensibly looks amazing, but his HbA1c was 5.8% on his last blood draw And it’s always been a little bit elevated, but, HbA1c is often falsely elevated
They did a CGM on him, and Peter didn’t like the results His average blood glucose over the 10 days was 115 mg/dL This is actually congruent with the HbA1c of 5.8%
That puts him in the category of pre-diabetic, and even though his fasting glucose was 90 and his fasting insulin was 6 (which are both great numbers)
And it’s always been a little bit elevated, but, HbA1c is often falsely elevated
His average blood glucose over the 10 days was 115 mg/dL
This is actually congruent with the HbA1c of 5.8%

It’s clear that, postprandially, he’s having issues

Again, that’s being captured in the A1C and in the CGM
Most people wouldn’t even dream of getting worked up about this, but we talked about it today, and Peter said, “ Look, let’s do another CGM tracing in a couple of weeks, and if things continue to be this way, I want to get even more aggressive on what we’re doing with you from an exercise-nutrition standpoint .”
And although Peter didn’t bring up pharmacology today, he thinks we’re still one step removed from that

⇒ The bottom line is Peter is not going to let that guy walk around with an average glucose of 115 because pretty soon he will start to develop postprandial and fasting hyperinsulinemia, and then he’s on a devastating cascade

This guy is only 30 or 35, so you have to play the long game here ‒ not letting someone get to that point

Saum Sutaria episode: the economic and systemic drivers of high healthcare costs in the U.S. [54:00]

#327 – Choices, costs, and challenges in US healthcare: insurance intricacies, drug pricing, economic impacts, and potential reforms | Saum Sutaria, M.D. (December 2, 2024)

The episode with Saum on the US healthcare system was insanely technical, insanely interesting
Clearly a lot of eyes are on it right now, to say the least
Peter has known Saum for a long time and has been interested in this topic for a long time Saum is not only an incredibly close friend, but also among the most important mentors Peter has ever had
Saum is a physician, trained as a cardiologist, but left medicine to join McKinsey 25, 26 years ago
He was the guy that recruited Peter when he left medicine to come to McKinsey That’s where Saum was running McKinsey’s North American healthcare practice, at the time
Even though Peter spent most of his time at McKinsey working in financial services, he did spend about a quarter of his time in healthcare and all of it was with Saum and it was working in the healthcare system
Even after Peter left the firm, he and Saum have remained very close
Saum’s career has been remarkable
Peter and Saum always talk about this stuff
Saum is the CEO of Tenet, which is the second largest hospital system in the United States
Peter said to him, “ Look, I realize it’s hard for CEOs of public companies to do podcasts .” But he agreed to do it
Saum is not only an incredibly close friend, but also among the most important mentors Peter has ever had
That’s where Saum was running McKinsey’s North American healthcare practice, at the time

⇒ This episode was just his views, not the company’s views

Why we pay 2x per capita of what every other country pays for healthcare and we can’t deliver even top-20 life expectancy

This podcast did a deep dive into this, and Peter is not going to be able to summarize it all here
If you find this topic even remotely interesting, just pick up your phone and go to this episode and listen to it

Peter’s main takeaways

It’s important to just have some scale on all of these things

The US federal budget each year is about $5 trillion
We typically overspend that by about $2 trillion, which is what a deficit means
So the US budget ends up being actually closer to about $7 trillion per year (that’s what the US government spends every year) And obviously that $2 trillion of excess is being added to a pile of debt that now sits at about $35 trillion
Our GDP (gross domestic product) is probably in the neighborhood of about $28-30 trillion at this time
Total US health care spending: right now we’re sitting at $4-4.5 trillion a year It varies a little bit from year to year, and we saw a bit of an uptick in COVID
And obviously that $2 trillion of excess is being added to a pile of debt that now sits at about $35 trillion
It varies a little bit from year to year, and we saw a bit of an uptick in COVID

Breakdown of the $4+ trillion/year of US healthcare spending

About $1 trillion of that is insurance premiums and out of pocket spending by consumers
About another $1 trillion of that is the employer’s spending to insure their employees
About $2+ trillion of that is the government spending through something called CMS (the Center for Medicaid and Medicare Services) This is both state and federal spending
This is both state and federal spending

Where does that money go?

15% off the top (of $4.5 trillion) goes to administration Payment adjudication and reconciliation is a very big deal This is very unique to the US because we are the only country that doesn’t have a single payer system (like the UK, Canada, Australia, or most of Europe) The movement of money in our system is something that doesn’t exist in any other system, and that costs an enormous administrative burden
Payment adjudication and reconciliation is a very big deal
This is very unique to the US because we are the only country that doesn’t have a single payer system (like the UK, Canada, Australia, or most of Europe)
The movement of money in our system is something that doesn’t exist in any other system, and that costs an enormous administrative burden

The easiest way to think of the remaining 85% is to actually just break it into thirds

⅓ of the remainder is drugs and medical devices Much more of that cost comes from drugs than medical devices (we’re going to come back to that)
⅓ is capital investment in hospital facilities and infrastructure (the actual buildings)
⅓ is the medical office, clinic We’re really mostly talking about the wages
Much more of that cost comes from drugs than medical devices (we’re going to come back to that)
We’re really mostly talking about the wages

The easiest way to think about it is we’ve got the administrative spending about moving money, wages are very high in the US, and we have a lot of infrastructure

As an aside, we talked about the Hill-Burton Act of 1946 , which was really an act that wanted to ensure that Americans didn’t have to travel very far to get medical care

⇒ This is a bit of a death by 1,000 cuts problem

When you live in a country as big as the United States, it’s very different than if you’re creating healthcare policy in the UK where people are on top of each other
The US has a huge rural population relative to other places potentially and at a greater distance from major cities, you just have to incur much more of a capital base and infrastructure
Drugs and drug prices: it’s no secret that we pay more for drugs than anybody else

Healthcare is about 18% of GDP ‒ that’s staggering, both in absolute dollars, nobody’s within the same zip code

Keep in mind, we were 5% of GDP in 1960, 6% of GDP in 1970, and we’re 18% of GDP today
In absolute dollars : in 1970, it was $74 billion
In the year 2000 (this is 24 years ago), it was $1.4 trillion
Today it’s $4.5 trillion

“ The thing that should give anybody chest pain if they’re listening to this is the rate of healthcare inflation is exceeding the real rate of GDP inflation and the nominal rate of GDP inflation. This is very concerning. ”‒ Peter Attia

If someone wanted to reduce healthcare spending by $3-4 trillion dollars in the next 5 years, how would they do it?

Saum took a framework that Peter has talked about before, but added something to it that makes it better

⇒ American consumerism demands and it expects choice, speed, quality as the things that matter; it’s not optimizing for cost

The reason for it is (like most healthcare systems) we don’t bear the cost
Individuals only make up 15% of the total healthcare costs Out of our pockets we pay very little
In the 1950s, Americans paid 50% of healthcare out of pocket, but the total dollars were so much smaller on a relative basis to their budget
Out of our pockets we pay very little

If we demand choice and speed and quality and we’re not optimizing for cost, we’re never going to lower that

The question is, do we really want to go from $4 trillion to $3 trillion in spending? And if so, what would we give up?

Would we give up choice? Speed? Quality?
If you need an MRI, do you want to be able to get it within the week or do you want to wait 6 months like your neighbors to the north and our European neighbors might have to wait? Because if you want to get it in a week, you need more MRI machines than if you’re willing to wait 6 months You’re going to need more radiologists, more technicians
You have to think through this
Because if you want to get it in a week, you need more MRI machines than if you’re willing to wait 6 months
You’re going to need more radiologists, more technicians

⇒ Most other countries throttle supply on healthcare

How other countries spend a fraction of what we spend on healthcare

Basically by limiting the number of facilities and machines and the infrastructure
Coupled with having the power to negotiate drug prices
Coupled with having a single payer system that doesn’t have any administrative friction of cost
Coupled with not dealing with PBMs (we’ll come back to this)

But you have to give something up; and the question is, how much of that are we willing to give up?

“ The good news is I actually don’t think we have to give up a whole heck of a lot, and I think we could actually reduce the cost of healthcare significantly .”‒ Peter Attia

How we could significantly reduce the cost of healthcare

He doesn’t know that we could get it down to $3 trillion
But we could absolutely get healthcare cost to grow at a much slower rate than GDP is growing

Go back to the 3 buckets [that constitute ⅓ of spending after administrative costs]

1 – The hardest bucket to constrain is wages because that’s where you really do need to keep up with inflation We’re just not going to ask people to all of a sudden overnight start working for less money Doctors, nurses, technicians, their standard of living shouldn’t go down in an effort to cap the cost of healthcare
That is going to be the bucket where cost is going to continue to grow at what it’s growing
We’re just not going to ask people to all of a sudden overnight start working for less money
Doctors, nurses, technicians, their standard of living shouldn’t go down in an effort to cap the cost of healthcare

⇒ That bucket can start to come down and we use AI do more of the jobs (AI efficiency)

But on a per individual basis, we’re not going to see cost cutting there
2 – On the infrastructure side, there is an opportunity to reduce cost by basically reducing the capital footprint of marginal assets
This is not a huge opportunity because if you cut down too much of the infrastructure, you start to lose choice and speed
Americans have, for right or wrong (Peter included) decided we don’t want that If you need an MRI, you want to be able to get it in a day or two; you don’t want to wait 6 months like Peter’s family does up in Canada That’s the trade-off that we have to think about
3 – On the drug pricing side, there’s an enormous opportunity
If you need an MRI, you want to be able to get it in a day or two; you don’t want to wait 6 months like Peter’s family does up in Canada
That’s the trade-off that we have to think about

2 things would need to be true and there’s no reason either of these need not be true

1 – The United States government (to begin with) and ultimately private payers need to be able to negotiate reasonably with pharma companies Peter’s not going to get into the history and the politics of why that’s not the case, but that’s a fixable problem and that’s a very bipartisan issue
Peter’s not going to get into the history and the politics of why that’s not the case, but that’s a fixable problem and that’s a very bipartisan issue

⇒ As it stands today, the United States is truly subsidizing the rest of the world from a pharma pricing standpoint

We develop all these drugs and everybody gets them for cheap and we don’t
2 – The PBMs (pharmacy benefit management entities) need to go away
They’re opaque Peter will probably end up doing a dedicated podcast on this because he really wants to make sure every single listener to this podcast understands how perverse the incentive structures are around PBMs and how value destructive they are
They operate in the shadows of healthcare, most people are completely unaware of what they are, and as a result are completely unaware of how destructive they are
Peter will probably end up doing a dedicated podcast on this because he really wants to make sure every single listener to this podcast understands how perverse the incentive structures are around PBMs and how value destructive they are

If we could simply get rid of PBMs and basically allow AI to help benefits managers within companies do the work themselves, we can get rid of that shell game

US life expectancy is horrible relative to other developed nations

Which is that much more upsetting in light of how much we spend
Saum made an interesting point, which is it’s mostly because of how much life expectancy is dragged down for people below 60

⇒ Where the US is really failing is that our mortality for people under 60 is really bad

It’s largely due to prenatal care
It’s largely due to higher infant mortality.
It’s largely due to higher drug use and substance abuse
Higher injuries and homicides
Most of those things end up killing people below the age of 60, and that’s a huge drag down on life expectancy

If you look at people between the ages of 60-75, it actually flips and we actually end up having a very high life expectancy

Once you make it out of those other causes of death

⇒ Between the ages of 60-75 is probably where our medical system thrives

It’s like eking out incremental life via medicine 2.0 once the low-hanging fruit of some of these other public health measures are not addressed
Peter could talk about this stuff forever, but it’s just better to go back and listen to the podcast

Reducing health care costs: redefining health insurance, lowering drug prices while maintaining innovation, leveraging AI for efficiency, and more [1:07:15]

Health insurance isn’t necessarily insurance

It’s almost like a discount card

This is an interesting way to think about it that’s not necessarily always talked about
Of all the things that you have insurance for ‒ you buy insurance for unexpected things You have umbrella insurance, car insurance, property insurance, life insurance You don’t buy car insurance to take care of our oil change We buy car insurance so that if we get into an accident, someone pays the catastrophic cost
That’s not actually what health insurance is at all
You have umbrella insurance, car insurance, property insurance, life insurance
You don’t buy car insurance to take care of our oil change
We buy car insurance so that if we get into an accident, someone pays the catastrophic cost

⇒ Health insurance is a discount card: it’s basically a tool that allows group purchasing power for services that are to be expected

There’s all sorts of people, policymakers and non, are thinking about ways to sort of rethink the model of healthcare delivery

Another obvious point Peter failed to mention: if people were healthier, healthcare costs would be lower

Americans are not very healthy
Not just mortality, but talk about morbidity Americans have more obesity, more type 2 diabetes, and more of these things means higher costs as well
Americans have more obesity, more type 2 diabetes, and more of these things means higher costs as well

Where the US could improve

What would it take to be healthier?
What would it take to have AI reduce all the administrative friction in payment?
What would it take to have AI improve efficiency in the workforce?
What would it take to legally be permitted to negotiate with pharma companies and to eliminate the PBMs?

You add all that up, and there’s a world in which we only have to spend 12% of GDP on healthcare

And that’s still at the upper end of what a developed nation spends, but that’s infinitely more sustainable than the path we’re on now

If we used price controls to lower the cost of drugs, how would that affect innovation?

Peter doesn’t think that more competitive drug pricing in the United States to mirror the rest of the world should hamper innovation
There are always going to be innovative people
There are always going to be innovative small companies that are always going to be pushing the envelope of what can be done
Yes, AI is already making an impact at part of that It’s certainly making an impact around on the molecular side of drug discovery [discussed in episode #33 with Mike Israetel, Ph.D. ]
Peter thinks this is a narrative thrown out by people whose interest is in preserving a certain economic structure But he doesn’t see any evidence to it
Most innovation these days is done by smaller companies that are taking the risk, and that’s venture backed
It’s certainly making an impact around on the molecular side of drug discovery [discussed in episode #33 with Mike Israetel, Ph.D. ]
But he doesn’t see any evidence to it

Peter would be willing to take the bet that this will not stifle innovation

He thinks innovation is only accelerating, not slowing down

Do you think money spent by pharmaceutical companies on advertising has anything to do with lowering healthcare costs?

Peter doesn’t know how much of that would factor into healthcare costs

⇒ He absolutely thinks pharmaceutical advertising should be abolished

He sees no reason for it whatsoever
He thinks it’s manipulative
Advertising in general should be scrutinized a little harder

Peter would certainly agree with the idea that we just don’t need to be advertising drugs to either patients or doctors

We should allow people to do their own research and certainly not be told to go and ask your doctor about this and that and the other thing (he finds that to be distasteful)

Trenna Sutcliffe episode: insights on autism, ADHD, and anxiety in children—definitions and diagnosis [1:11:45]

#330 – Autism, ADHD, and Anxiety: Understanding the rise in autism and a multidisciplinary approach to diagnosis and treatment of each condition in children | Trenna Sutcliffe, M.D. (January 6, 2024)

Peter really enjoyed his discussion with Trenna
He wasn’t very familiar with what she refers to as a biopsychosocial model for treatment of children and adolescents It makes a lot of sense
It makes a lot of sense

⇒ Which is you want to be able to completely integrate the biological component, the mental health component, and the social component of a condition

And it’s that social component that bridges between mental health and education, which for kids is such an important thing

She’s largely talked about 3 things: autism, ADHD, and anxiety

But she always starts from a question of: is the trait a set of traits that manifest with the diagnosis of autism? A set of traits that manifest through ADHD? ADHD is a term that just gets thrown around nonstop, everyone’s got ADHD today and anxiety With anxiety?
A set of traits that manifest through ADHD? ADHD is a term that just gets thrown around nonstop, everyone’s got ADHD today and anxiety
With anxiety?
ADHD is a term that just gets thrown around nonstop, everyone’s got ADHD today and anxiety

⇒ She’s not interested in the label, she’s interested in the impairment

This is very helpful, especially for parents thinking, “ Hey, if I’m told that my kid has ADHD, should I be worried? ”

Look how a trait/ diagnosis is impacting 3 things

1 – How does it impact your child socially?
2 – How does it impact your child’s ability to communicate with other children, with adults?
3 – How does it impact your child’s ability to learn?

And if the answer is, “None of the above,” then it’s very difficult to call it a disorder

She digs in a little bit deeper

Are these traits impacting the interaction of the child with their peers?
Their ability to learn at school?
And even their own sense of self-esteem?

The system and the DSM-5

There was a change in the DSM-5 that took place about 10 years ago regarding autism spectrum disorder (now called ASD) ‒ this is a very broad spectrum that includes many things that were not included before

In the DMS-4, you had 3 diagnoses

1 – Autism was a disorder of social deficits, stereotyped or repetitive behaviors, restricted interests, and significant communication challenges that included but were not necessarily limited to non-verbal behavior
2 – You had something that was called pervasive developmental disorder not otherwise specified (or PDD-NOS) And it usually meant that a kid didn’t fit the threshold for autism or Asperger’s
3 – Asperger’s, which is no longer a term that is used, now someone that would’ve fit the diagnosis of Asperger’s is somewhere on the ASD spectrum Formerly, we specifically called out Asperger’s as a high-functioning form of autism that had autism-like symptoms, but normal language and normal or high IQ
These are kind of messy buckets Because you have Asperger’s, which is today what we think of as high-functioning autism You have autism, which was quite severe And then you have this PDD-NOS, which is just another way of saying something’s wrong, but it’s not otherwise specified, so we don’t know what to call it
And it usually meant that a kid didn’t fit the threshold for autism or Asperger’s
Formerly, we specifically called out Asperger’s as a high-functioning form of autism that had autism-like symptoms, but normal language and normal or high IQ
Because you have Asperger’s, which is today what we think of as high-functioning autism
You have autism, which was quite severe
And then you have this PDD-NOS, which is just another way of saying something’s wrong, but it’s not otherwise specified, so we don’t know what to call it

Right out of the gate the DSM-5 consolidated all 3 of these into 1 diagnosis called ASD (or Autism spectrum disorder)

They were divided into 3 levels based on the increasing severity
Kids at level one have a lot of skills, and many of the formerly Asperger’s children would be in level one A lot of times these kids are not diagnosed If they are, they’re not necessarily supported
A lot of times these kids are not diagnosed
If they are, they’re not necessarily supported

The definition today is a lot broader, but in particular it’s broader in 3 areas

1 – It pays more attention to social reciprocity
2 – It pays more attention to nonverbal communication skills, eye contact gestures
3 – It pays more attention to how children understand relationships

When people talk about the increase in the prevalence of autism, a part of that is due to a change in the diagnosis

Does that explain all of it? It certainly does not (and we talked about that)
But when you change the diagnostic criteria from autism (being a relatively narrow diagnosis) to now ASD (including a whole bunch of other things that weren’t there), you’re going to see it go up

Exploring the rising prevalence of autism spectrum disorder [1:17:15]

What is the heritability of ASD?

This also came up on 2 previous podcasts [episodes #191 & #268 ]
Heritability is a term that refers to how genetically encoded is it?
The way that heritability studies are done is by looking at identical twins that are raised separately It’s the closest we have to an experiment that can tell us: how do genes more than environment impact outcome? If you have identical twins that are raised in separate environments, that’s the best way we can understand the heritability of various conditions
For example, we understand the heritability of depression, the heritability of schizophrenia, the heritability of major depressive disorder, bipolar disorder, things of that nature
It’s the closest we have to an experiment that can tell us: how do genes more than environment impact outcome?
If you have identical twins that are raised in separate environments, that’s the best way we can understand the heritability of various conditions

⇒ Depending on this study you look at, the heritability of ASD is anywhere from 70% to the mid to high 90s

“ Here’s the kicker, it involves multiple genes. There is no autism gene. In fact, there is no commonly identified set of genes. ”‒ Peter Attia

There are simply many genes and what appears necessary is a confluence of factors

In biology this is often referred to as a hit model
So is it a two-hit model? A lot of hereditary cancers end up being two-hit models You have this gene that already predisposes you, and then you get one perturbation to the system and you get the cancer
Scientists don’t think that that’s what’s going on with autism
They think it’s a multiple hit gene and the timing of those hits matters
A lot of hereditary cancers end up being two-hit models
You have this gene that already predisposes you, and then you get one perturbation to the system and you get the cancer

What explains the increase in the incidence of autism beyond the diagnostic criteria change and the awareness of it and the increased screening for it?

⇒ Trenna cited the following: maternal stress, maternal diet, paternal age, and pollution

She believes that those 4 things in addition to the diagnostic changes, coupled with the increased awareness and the increased looking, are playing the biggest role
One of the things that was not here was vaccines We’ve covered this in great detail on previous podcasts [episodes #158 , #159 , and #317 ] and it’s important to understand that
We’ve covered this in great detail on previous podcasts [episodes #158 , #159 , and #317 ] and it’s important to understand that

When it comes to understanding the increase in autism, it seems there are 2 very polarizing views

One view is, it’s caused by vaccines
The other view is, this is all in your imagination
Peter finds that very problematic
Because while he completely agrees that vaccines are not driving autism that doesn’t mean something else in the environment isn’t triggering it Back to vaccines ‒ just on first principles it doesn’t even make sense when you look at the data, when you understand that autism develops long before children even get vaccines There’s so much evidence for why vaccines are not causing autism
Peter has no idea what in the environment may be triggering the increase in autism
He doesn’t think he’s qualified enough to suggest what role maternal stress, diet, paternal age, pollution, or any other environmental trigger plays
Back to vaccines ‒ just on first principles it doesn’t even make sense when you look at the data, when you understand that autism develops long before children even get vaccines
There’s so much evidence for why vaccines are not causing autism

Peter explains, “ The only reason I can comment so emphatically on the vaccine one is I’ve gone down that rabbit hole, and I’ve gone very deep down that rabbit hole, and I feel very comfortable asserting that. But it does bother me when the view is one or the other. It’s either vaccines or it’s nothing, and you’re just crazy to suggest that there’s an environmental thing going on here. ”

Trenna’s views on caring for children with autism [1:21:15]

She’s taken care of countless children like this

She said, “ If you’ve met one child with autism, you’ve met one child with autism. ”

⇒ All the care for kids with autism needs to be very tailored to them and their family’s needs

Diagnosis is important, but primarily what we want to do is get to treatment options
She wants to give tools to help the adults around the children
She wants to give structure for how to leverage a kid’s strengths and weaknesses
So we often hear parents of kids with autism their kid’s certain superpowers but they also their kryptonite
It’s very important to figure out how to manage that both at home and at school

The 3 A’s: ADS, ADHD, and anxiety

They involve different parts of the brain

ASD is in the frontal lobes, the amygdala, the cerebellum, and the temporal lobes
ADHD is mostly prefrontal cortex with executive function and the frontal lobes
Anxiety is mostly in the amygdala

⇒ The interesting and confounding and sometimes unfortunate thing is that ASD also has a high overlap with ADHD

It’s about a 40-70% overlap
And 40% of those kids on average have anxiety
So many have all three

This makes the diagnosis challenging, makes the treatment challenging

ADHD is also often accompanied by anxiety and it’s often accompanied by oppositional defiant disorder

Any parent listening can relate to a kid that sometimes comes across as quite oppositional defiant

⇒ Trenna asked an interesting question, a question Peter has been doing with his kids: When kids are being oppositional, what is the function it is serving?

Is it sensory overload?
Are they anxious about something?
Are they impulsive?
Are they embarrassed?
What’s going on?

Anxiety has many types

We talked briefly through these things

We talked also about the age at which kids can be diagnosed

Peter was surprised that the typical age for autism is 3-4, but she said if a person really knows what they’re looking for, 18 months would be when you would typically diagnose it
Amazingly, she said half of kids with autism are not diagnosed until they’re over 6 years of age That’s a little bit sad to hear, because that means that those kids have already entered school at the time of diagnosis, which means they didn’t get off to the right start necessarily
ADHD can be diagnosed at age 4-5 but 5-6 is more typical
Anxiety could be diagnosed at any point in time
That’s a little bit sad to hear, because that means that those kids have already entered school at the time of diagnosis, which means they didn’t get off to the right start necessarily

We talked about medications around this

This is an area where Peter just has no insight
He still finds it amazing that stimulants and amphetamines are the drug of choice here

Peter has studied this quite a bit and come to the conclusion that if the symptoms warrant it, and if your alternative is a child who’s not able to learn and not able to maintain relationships, then these drugs can be miracle drugs

Even though it seems counterintuitive and there’s clearly a cultural zeitgeist that suggests that parents shouldn’t be giving their children these drugs, Peter thinks that type of shaming is ridiculous
He can’t tell you the number of parents he knows whose kids have done so well, have experienced a complete 180, when they’ve been put on these various drugs

Types of medications used to treat ADHD

1 – The methylphenidate drugs: Ritalin, Focalin
2 – The amphetamine drugs: Adderall , Vyvanse

“ One of the things I took away from my discussion with her is, if your child looks like they’re going to benefit from pharmacology and you go down that route, don’t stick with the first drug that you try if you’re not happy with the outcome. There are so many different ways to do this. ”‒ Peter Attia

Just because a child is put on one of these drugs doesn’t mean that they’re going to be on that drug for life
Nor does it mean that they should tough it through that drug
3 – There’s also non-stimulants: Stratera increases norepinephrine level in the synapses

ASD really has no treatment

The only treatment pharmacologically would be for any of the symptoms that come up that are typically with ADHD or emotional regulation or anxiety

What do you look for in a provider?

Trenna said you want people who are very personalized in their care
People who are skilled in treating the parents and not just educating them (this is very important) It’s actually teaching them how to be better parents And then they’re very proactive as opposed to just open to helping the parent create a collaborative team between the school and the home
It’s actually teaching them how to be better parents
And then they’re very proactive as opposed to just open to helping the parent create a collaborative team between the school and the home

Misconceptions around vaccines and autism [1:26:00]

Do you think understanding where the increase in autism prevalence is coming from is an easily answerable question, or is that always going to be an area that causes confusion?

Peter thinks it is easy to answer if people could be honest
It’s very difficult to answer if there’s a constant shifting of goalposts
People can go back and listen to the Brian Deer podcast [ episode #158 ] where we go through the entire history of the controversy around MMR and autism Peter can’t tell you how many people he’s talked with who have said MMR vaccines cause autism, and he’s gone through that entire argument with them, which is, you can’t run from it And instead they come up with a new argument altogether the moment you’re done spending an hour talking through that And they’re like, “ Well, it’s got to be this thing… But what about this vaccine and what about this vaccine? ”
Peter wants to be clear, he’s not saying it makes sense to give children every vaccine according to every schedule He doesn’t want to get into it It’s not his area of expertise On a personal level, he did not choose to give his kids the Hep B vaccine 10 minutes after they were born or whatever was recommended Let’s give the things that seem like they make sense, and things that don’t make sense today, let’s postpone them for a year or two or whatever it is
He thinks there’s a middle ground where it’s very hard to exist because you feel attacked from both sides
Peter can’t tell you how many people he’s talked with who have said MMR vaccines cause autism, and he’s gone through that entire argument with them, which is, you can’t run from it
And instead they come up with a new argument altogether the moment you’re done spending an hour talking through that
And they’re like, “ Well, it’s got to be this thing… But what about this vaccine and what about this vaccine? ”
He doesn’t want to get into it
It’s not his area of expertise
On a personal level, he did not choose to give his kids the Hep B vaccine 10 minutes after they were born or whatever was recommended Let’s give the things that seem like they make sense, and things that don’t make sense today, let’s postpone them for a year or two or whatever it is
Let’s give the things that seem like they make sense, and things that don’t make sense today, let’s postpone them for a year or two or whatever it is

Peter adds, “ I really do pity the parents who are not armed with this information and are being misled by people who are suggesting that if their kids get the MMR vaccine or the polio vaccine, they’re increasing the risk of autism. ”

This is categorically untrue

It’s as demonstrated through the literature as virtually anything can be based on the complete and total abundance of negative epidemiologic data demonstrating any causality
So when we start to talk about MMR and polio and things like that, Peter gets particularly distressed

Mike Israetel episode: insights about strength training, minimum effective dose, troubleshooting plateaus, tips for beginners, and more [1:28:15]

#335 ‒ The science of resistance training, building muscle, and anabolic steroid use in bodybuilding | Mike Israetel, Ph.D. (February 9, 2025)

Really interesting episode about resistance training
Early on in the discussion, we started talking about the differences in intensity

Why is it that we can do lower intensity training (like cardio training) much more than we can do resistance training?

Basically he pointed out the obvious, which is that it’s really a difference in the metabolism of the muscle fiber When you’re doing cardio training, you’re mostly using type I muscle fibers When you’re doing resistance training, you’re mostly using type II muscle fibers, which are much more forceful
The oxidative stress of training type II fibers is much higher, and there’s a very nonlinear exponential increase in damage that occurs using muscle fibers under higher forces
When you’re doing cardio training, you’re mostly using type I muscle fibers
When you’re doing resistance training, you’re mostly using type II muscle fibers, which are much more forceful

All of those things say that to adapt type II fibers requires much less volume in total than to adapt a type I fiber

We talked a lot about the relationship between neural fatigue also being more of an issue when you’re lifting something really heavy versus even doing a VO 2 max type workout where you can actually tolerate much more volume because you’re still primarily using an oxidative system

Mike is probably one of the most transparent people about his use of anabolic steroids

Peter adds, “ It’s probably funny to watch my face during some of this episode, because again, it’s not that I haven’t heard of people taking such high doses before, but somehow Mike sitting there articulating it seemed that much more amusing to me. ”

Mike talked at length about his use of testosterone and other anabolic agents that are similar
His uses are quite significant
He said his maintenance low dose is in the 250-500 mg of testosterone per week Though at times he cycles to 2,000 mg/week
To put that in perspective, a typical medical replacement dose would be about 100-120 mg/week Maybe 140 mg/week ‒ that would be a lot
So if his maintenance dose is 2-4x that, and at times going up to 2,000 is pretty remarkable
Though at times he cycles to 2,000 mg/week
Maybe 140 mg/week ‒ that would be a lot

2 remarkable things about strength training

1 – What’s the minimum effective dose workout?
2 – If I’m in the gym and I’m training and part of my motivation is hypertrophy and it’s not happening, can you give me a sense of why?

For #1: Mike said that you could achieve 70-80% of your maximum potential if you did only 30 minutes twice a week

Spaced out by 3 days If you did a Monday, Thursday or Tuesday, Friday or something like that
If you did a Monday, Thursday or Tuesday, Friday or something like that

⇒ But this is only true if you’re willing to do a very, very intense and exhaustive workout

To be clear, most people (Peter includes himself) could not do that for very long
So this doesn’t apply to Peter and it doesn’t apply to most people, but it’s just interesting to note

What would that workout look like?

This workout would be a whole day body split twice Meaning you do the whole day body split on Tuesday, then on Friday
There are no breaks in the workout You just do opposing muscles back and forth, back and forth You’re basically just doing compound movements: rows, pull-ups, pull-downs, squats, deadlifts
Meaning you do the whole day body split on Tuesday, then on Friday
You just do opposing muscles back and forth, back and forth
You’re basically just doing compound movements: rows, pull-ups, pull-downs, squats, deadlifts

You’ll train much harder than you think is reasonable and be really exhausted if you do this

This type of a workout has its pluses and minuses You could say, “ Well, great, hey, I only need to work out an hour a day .” But the focus and attention that you need to bring to that workout is really hard
Peter prefers to work out more frequently, spend more time, and not have to work out that hard (sounds a little soft) For example, he works out 3x a week He does each body part once a week Each of those workouts ends up being about 90 minutes
You could say, “ Well, great, hey, I only need to work out an hour a day .”
But the focus and attention that you need to bring to that workout is really hard
For example, he works out 3x a week
He does each body part once a week
Each of those workouts ends up being about 90 minutes

For #2: I’m crushing it in the gym, but I’m not making any gains. What’s going on?

Mike used a nice framework which was there’s going to be “in the gym” reasons and “out of the gym” reasons

“In the gym” reasons include

1 – Exercise selection ‒ for example, if you want to grow your arms and you’re not doing a lot of triceps (they are more than ½ the upper arm)
2 – Technique ‒ if you’re trying to grow your pecs, but nothing you do is ever putting the pec under a deep stretch, you’re never going to fully appreciate the force at the open end of the range of the muscle
3 – Volume ‒ Mike argued that you need to do a minimum of 10 working sets per week That doesn’t include your warmups Per week per muscle If you’re below that, you’re probably going to have a hard time achieving hypertrophy
4 – Intensity ‒ you should not be at an intensity less than 2 reps in reserve This has been echoed by many other guests on this podcast: Lane Norton, Andy Galpin That means if you put the bar back and you could have done 3 or 4 reps, so that’s 3 or 4 reps in reserve (you didn’t go hard enough)
5 – A lack of consistency ‒ so even if you’re doing everything right but you can’t do it consistently, it’s going to be very difficult to make gains
That doesn’t include your warmups
Per week per muscle
If you’re below that, you’re probably going to have a hard time achieving hypertrophy
This has been echoed by many other guests on this podcast: Lane Norton, Andy Galpin
That means if you put the bar back and you could have done 3 or 4 reps, so that’s 3 or 4 reps in reserve (you didn’t go hard enough)

“Out of the gym” reasons include

1 – Rest ‒ it’s not just sufficient to rest between training days If you can’t train the same muscle 3x a week, it’s just not going to be sufficient So it’s the inter-week rest, but then it’s also, he figures about every 8 weeks you need a deload week where you really reduce your intensity And then maybe 1-2 weeks a year where you’re completely off
2 – Genetics ‒ which are not only out of the gym, but non-modifiable, plays a significant role here
3 – Age ‒ the older you are, all things equal, the harder it is to achieve hypertrophy This could be due to fewer type II muscle fibers, less capacity to recover, more stress, more inflammation
4 – Nutrition ‒ are you getting enough calories and enough protein?
5 – Lack of consistency ‒ you can be doing most of these things right most of the time, and that might not be efficient or sufficient
If you can’t train the same muscle 3x a week, it’s just not going to be sufficient
So it’s the inter-week rest, but then it’s also, he figures about every 8 weeks you need a deload week where you really reduce your intensity
And then maybe 1-2 weeks a year where you’re completely off
This could be due to fewer type II muscle fibers, less capacity to recover, more stress, more inflammation

How does someone truly know what their reps in reserve are?

You mentioned, if you stop and you really had 3-4 in the bank, do you really have to go to failure to understand what that is?
And then how often do you reset what your reps in reserve are? Do you go to failure, and then maybe a few times a year you do the same to see if you’ve made improvements?
Do you go to failure, and then maybe a few times a year you do the same to see if you’ve made improvements?

“ Everybody needs to know what it feels like to go to failure, and everybody therefore needs to get familiar with what does it feel like when I was one rep before that. ”‒ Peter Attia

And what it feels like when I was 2 reps before that, 3 reps before that
Peter’s 1 rep in reserve will vary quite a bit from week to week
Yesterday was a chest, tricep day, shoulder day and Peter shares, “ Man, for reasons I don’t know, I was doing 1-2 fewer reps for all of my presses yesterday. Swiss bar, dumbbell, incline, everything I did yesterday was just a little bit lower. But I didn’t have to go to failure to know that because I’m very familiar just through having done it so many times what it feels like when I’m about 1 rep away from failure. ”

Everybody needs to get familiar with failing

Obviously don’t pick an exercise that’s dangerous
You don’t want to figure it out on a squat or a barbell bench press

Get comfortable doing it either when you have spotters or when you can do it on an exercise where the consequences are not dire

You will quickly start to understand, when you’re on that second rep before failure, you will have the sense that you could do 1-2 more, and that’s about it

If someone’s new to exercise, is it worth them even knowing what that reps in reserve is? Or is that something where they need to just build up some muscle first?

It just depends on where you’re starting from
Because if you’re truly starting from a place of knowing nothing, having never done resistance training, it probably doesn’t matter It’s okay if you’re at 3-4 reps in reserve If you’re just doing a program that’s set up as 3 sets of 12 or 3 sets of 15, it’s not worth worrying about this, because you’re going to make amazing gains because you’re starting so far on the curve
It’s okay if you’re at 3-4 reps in reserve
If you’re just doing a program that’s set up as 3 sets of 12 or 3 sets of 15, it’s not worth worrying about this, because you’re going to make amazing gains because you’re starting so far on the curve

Peter wouldn’t have people stressing about this if they’re completely new to resistance training

Is there at any point, at a certain age you worry about also doing this? Is there potential harm if you hit failure?

Obviously it’s different if you have a spotter
Peter still has his older patients think about reps in reserve
It’s another way to be very deliberate when you’re training, and frankly, that’s more important as you get older Because we still have to try to do all we can to hold onto and preserve the type II muscle fibers
There’s lots of safe ways to do that They don’t have to be methods that put you at extreme risk
Because we still have to try to do all we can to hold onto and preserve the type II muscle fibers
They don’t have to be methods that put you at extreme risk

A lot of people understand that exercise is important, but time is a huge factor and hearing Mike say, “ Hey, all you need is 30 minutes twice a week, ” (just for the resistance training piece) is very appealing

The way he does it is no breaks ‒ it’s not an easy 30 minutes You’re not checking your phone, you’re not doing emails, texts in between
You’re not checking your phone, you’re not doing emails, texts in between

What advice would you have for someone who’s thinking to themselves, “I want to get in the gym, I don’t have much time”?

If the alternative is doing nothing, Peter would say, “ Absolutely do it ,” but he would build into it a little bit slowly
You would just hate to see somebody do that and then get injured along the way, or frankly just burn out due to the intensity of it
But for some people, that could be really effective, especially if they have a trainer
It might be a little challenging to do it on your own

This is one of those things where if you have an accountability partner, that kind of workout could be more bang for your buck

Topics Peter is interested in exploring in future podcasts [1:40:15]

Peter is really interested in the role of AI in education That might be something he’d like to do a podcast on ‒ to understand how to accelerate learning for kids
He’s really interested in healthcare policy stuff
He’s become a huge fan this podcast called Acquired David Rosenthal and Ben Gilbert do these super deep dives into interesting companies That’s Peter’s intellectual candy these days
That might be something he’d like to do a podcast on ‒ to understand how to accelerate learning for kids
David Rosenthal and Ben Gilbert do these super deep dives into interesting companies
That’s Peter’s intellectual candy these days

You want to talk about your renewed interest in chess right now?

Not renewed, Peter had never played chess until a couple weeks ago
Now, it’s the first thing we play in the morning, and it’s the last thing we play before bed

Selected Links / Related Material

Episode of The Drive with Olav Aleksander Bu (Pt.2) : #331 ‒ Optimizing endurance performance: metrics, nutrition, lactate, and more insights from elite performers | Olav Aleksander Bu (Pt. 2 ) (January 13, 2025) | [2:00]

Episode of The Drive with Tadej Pogačar : #318 ‒ Cycling phenom and Tour de France champion Tadej Pogačar reveals his training strategies, on-bike nutrition, and future aspirations (September 23, 2024) | [12:45]

Portable VO 2 analyzer : VO2 Master | [17:00]

Episode of The Drive with Alex Hutchinson : #151 – Alex Hutchinson, Ph.D.: Translating the science of endurance and extreme human performance (March 1, 2021) | [19:15]

Episode of The Drive with Ralph DeFronzo : #337 – Insulin resistance masterclass: The full body impact of metabolic dysfunction and prevention, diagnosis, and treatment | Ralph DeFronzo, M.D. (February 24, 2025) | [24:30]

Episode of The Drive with Gerald Shulman : #140 – Gerald Shulman, M.D., Ph.D.: A masterclass on insulin resistance—molecular mechanisms and clinical implications (December 7, 2020) | [26:00]

The “ominous octet” of diabetes : Banting Lecture. From the Triumvirate to the Ominous Octet: A New Paradigm for the Treatment of Type 2 Diabetes Mellitus | Diabetes (R DeFronzo 2009) | [32:15, 38:30]

First episode of The Drive about GLP-1 receptor agonists : #184 – AMA #29: GLP-1 Agonists – The Future of Treating Obesity? (November 15, 2021) | [35:00]

Episode of The Drive discussing the kidney : #262 – AMA #49: Heart rate recovery, strength training, rucking, kidney function, and brain health (July 17, 2023) | [36:45]

Qatar study : [47:45]

Combination Therapy With Exenatide Plus Pioglitazone Versus Basal/Bolus Insulin in Patients With Poorly Controlled Type 2 Diabetes on Sulfonylurea Plus Metformin: The Qatar Study | Diabetes Care (M Abdul-Ghani et al. 2017)
Combination therapy with pioglitazone/exenatide improves beta-cell function and produces superior glycaemic control compared with basal/bolus insulin in poorly controlled type 2 diabetes: A 3-year follow-up of the Qatar study | Diabetes, Obesity & Metabolism (T Abdul-Ghani et al. 2020)
Type 2 diabetes subgroups and response to glucose-lowering therapy: Results from the EDICT and Qatar studies | Diabetes, Obesity & Metabolism (T Abdul-Ghani et al. 2022)

Episode of The Drive with Saum Sutaria : #327 – Choices, costs, and challenges in US healthcare: insurance intricacies, drug pricing, economic impacts, and potential reforms | Saum Sutaria, M.D. (December 2, 2024) | [54:15]

Episode of The Drive with Trenna Sutcliffe : #330 – Autism, ADHD, and Anxiety: Understanding the rise in autism and a multidisciplinary approach to diagnosis and treatment of each condition in children | Trenna Sutcliffe, M.D. (January 6, 2024) | [1:12:15]

Episodes of The Drive that discuss the heritability of autism : | [1:17:30]

Episodes of The Drive that debunk the perceived link between vaccines and autism : [1:19:45]

Episode of The Drive with Brian Deer : #158 – Brian Deer: A tale of scientific fraud—exposing Andrew Wakefield and the origin of the belief that vaccines cause autism (April 19, 2021) | [1:27:00]

Episode of The Drive with Mike Israetel : #335 ‒ The science of resistance training, building muscle, and anabolic steroid use in bodybuilding | Mike Israetel, Ph.D. (February 9, 2025) | [1:28:30] Podcast Peter is a huge fan of : Acquired: Every company has a story (2024) | [1:41:30]

People Mentioned

Transcript

Show transcript