Type 2 diabetes is not a glucose problem but a whole-body glucose overflow problem: cells are full, not deaf — so the correct treatment is to empty the suitcase (fasting, SGLT2 inhibitors) not jam more insulin in.
2
Jason Fung reversed T2D in patients with 20+ years of disease and large insulin doses using 24-hour fasts 3×/week, getting them off all insulin within 5–18 days — what he calls 'Medical Bariatrics' without surgery.
3
Hyperinsulinemia is a growth signal as much as a metabolic one: high chronic insulin drives cardiovascular proliferation, PCOS, NAFLD, and 13 obesity-related cancers via the same PI3K/mTOR pathway bodybuilders exploit.
4
Attia and Fung disagree on entry strategy: Fung leads severe T2D patients with 7-day water fasts cold turkey; Attia pre-loads a week of ketosis (0.5–1 mmol BHB) before entering any extended fast to ease the glycogen-depletion transition.
Protocols
Concrete recipes — what, when, how much, and why
7 items
24-hour fast 3×/week (OMAD protocol) for T2D reversal
WhatEat one meal per day, three days per week. The eating window covers the remaining four days normally. Combined with loose low-carbohydrate guidance but no strict carb counting.
WhenFor type 2 diabetics on insulin or multiple oral agents who have failed conventional management. Fung also starts here for patients who find longer fasts psychologically intimidating.
Dose24-hour fast (one meal a day) on Monday, Wednesday, Friday (or equivalent). Continue until insulin can be stepped down; then maintain or transition to 16:8 for maintenance.
For whomType 2 diabetics on oral medications or insulin, especially those with elevated ALT, visible abdominal obesity, or long disease duration. Not appropriate for type 1 diabetics without close monitoring.
WhyAfter 24 hours without food, liver glycogen is depleted, forcing the body to mobilize visceral fat. Hepatic fat (the root of insulin resistance) drops first, then pancreatic fat (the root of beta-cell failure). Diabetes reverses before significant body weight is lost.
CaveatsMust adjust diabetes medications in parallel — sulfonylureas and insulin must be reduced or stopped before fasting begins to avoid hypoglycemia. Requires close blood glucose monitoring. Not appropriate for T1D or LADA.
Fung's case series published in BMJ Case Reports documented three patients with 20–25 years of T2D on large insulin doses who were freed of insulin within 5–18 days on this protocol. Years later they remained non-diabetic. He distinguishes the mechanism from bariatric surgery: both work by rapidly depleting hepatic and pancreatic ectopic fat, restoring insulin sensitivity and beta-cell function before significant subcutaneous fat is lost. Roy Taylor's MRI data from the Counterpoint study provides the imaging evidence for this sequence.
Mechanism
After 24 hours, liver glycogen runs out and the body shifts to fatty acid oxidation. The liver — the primary site of ectopic fat in T2D — is the first organ to have its fat mobilized. As hepatic fat drops, insulin resistance (overflow paradigm: the suitcase empties) decreases. As pancreatic fat drops over weeks, first-phase insulin response recovers.
We put on 24 hours of fasting 3 times a week which is like a one meal a day right... We got off all their insulin like between 5 and 18 days. It was ridiculous how quickly they responded.
Also said
“After the first 24 hours glycogen runs out, so now you're going into a period where your body is going to have to metabolize fat for energy. The first place it's going to start pulling it out of is going to be the internal organs — the liver is gonna be the first place.”— The physiological sequence that explains why T2D reverses before body weight falls.
Individualized fasting escalation ladder (mild to severe T2D)
WhatChoose a fasting intensity matched to disease severity, age, and urgency: (1) 16:8 time-restricted eating for pre-diabetics or mild cases; (2) 24h fast 3×/week (OMAD) for moderate T2D on oral agents; (3) 36-hour fast 3×/week for severe or urgent cases (e.g., non-healing ulcers); (4) 7-day+ water-only fast for the most urgent cases in monitored hospital settings.
WhenSelect the tier based on urgency and severity: pre-diabetes → tier 1; T2D on oral agents → tiers 1–2; T2D on insulin → tiers 2–3; imminent limb loss or renal crisis → tiers 3–4.
DoseTier 1: 14–16 hour daily fast. Tier 2: 24-hour fast 3×/week. Tier 3: 36-hour fast 3×/week. Tier 4: 7+ day continuous water fast (hospital-based, close monitoring).
For whomTier 1: anyone; Tier 2: T2D on oral agents with supportive family/team; Tier 3: T2D on insulin with urgency (non-healing ulcer, impending amputation); Tier 4: hospitalized severe T2D cases under direct medical supervision.
WhyAll four tiers share the same mechanism (insulin lowering, ectopic fat mobilization) but differ in speed and aggressiveness. The 16:8 is described as a return to the 1970s baseline when 14-hour daily fasting was normal; full 7-day fasting is 'Medical Bariatrics' — the same mechanism as bariatric surgery but non-surgical.
CaveatsTiers 3–4 require stopping or massively reducing sulfonylureas and insulin before the fast begins. Phosphatidylserine (evening) and magnesium supplementation reduce the sympathetic surge and cramping respectively. Sleep disruption is expected and should be warned about in advance.
Fung treated a patient in his forties with a non-healing diabetic foot ulcer for over a year (close to osteomyelitis). He started with a 7-day inpatient water fast then transitioned to 36-hour fasts 3×/week. The ulcer healed in approximately 3 weeks. Two years later the patient was classified as non-diabetic with HbA1c of 5.9. Fung notes this level of aggression is rare — it was appropriate because of the immediate risk of limb loss — and that most patients can succeed with Tiers 1–2.
We use all of them so it's predominantly clinical. We will individualize. The things we take into account: one, how severe is it? How old the patient is and how willing they are to do it.
Ketogenic pre-loading before extended fasts (Attia protocol)
WhatSpend one week in nutritional ketosis (strict carbohydrate restriction, high fat, adequate protein) before entering any fast of 3+ days. This raises blood BHB to 0.5–1.0 mmol before fasting begins.
WhenBefore any planned 3-day or longer fast, especially for metabolically healthy individuals doing fasting for longevity rather than disease reversal.
Dose7 days of ketogenic eating precede the fast. Target BHB: 0.5–1.0 mmol on day of fast initiation. Attia measures with a blood ketone meter.
For whomAlready-healthy individuals doing periodic extended fasts for longevity or metabolic benefits. Not primarily for acute T2D reversal (where urgency may preclude pre-loading).
WhyA metabolically fat-adapted person (RQ approaching 0.7) enters a fast with much less glycogen-depletion distress. The transition to full fat oxidation is nearly seamless. In contrast, a glycogen-dependent person (RQ ~1.0) experiences severe hunger, irritability, and fatigue during the first 3–5 days as the body struggles to make the metabolic switch.
CaveatsAttia acknowledges even with this preparation he is still 'quite hungry at least once' during every seven-day fast. Pre-loading does not eliminate hunger — it makes the transition tolerable and smooth rather than brutal.
Attia describes the contrast starkly: his severe T2D patients (like the foot-ulcer case) are 'obligate glycogen consumers' — their respiratory quotient is essentially 1, meaning they have not oxidized a fatty acid 'since Guns and Roses were popular.' For those patients, asking them to fast cold turkey is a brutal metabolic shock. For metabolically flexible individuals, 1 week of ketosis effectively pre-empts the glycogen crash entirely.
Mechanism
Ketogenic eating for 7 days depletes liver glycogen, suppresses malonyl-CoA (removing CPT-1 inhibition), and up-regulates hepatic beta-oxidation enzymes. By day 7, the body is already running primarily on fat. Starting the fast from this state means the 'metabolic switch' has already occurred — the fast merely continues the same substrate utilization.
Personal experience
Attia: 'I usually spend a week in ketosis before the fast which means I'm showing up in a fast with a ketone level of somewhere between 0.5 and 1 millimolar of beta-hydroxybutyrate. I sailed generally pretty easily into these fasts.'
I usually spend a week in ketosis before the fast which means I'm showing up in a fast with a ketone level of somewhere between 0.5 and 1 millimolar of beta-hydroxybutyrate. I sailed generally pretty easily into these fasts.
Electrolyte and supplement protocol during extended fasting
WhatDuring multi-day fasts: (1) Magnesium supplementation daily — T2D patients are often already depleted; (2) Bone broth (with liberal salt) for dizziness or symptomatic sodium depletion — technically breaks a pure water fast but prevents the fast from being abandoned; (3) Phosphatidylserine in the evening to blunt the sympathetic surge that causes insomnia; (4) Optional: GABA supplement for sleep.
WhenStarting from day 2 of any fast longer than 24 hours. Bone broth used reactively when dizziness or significant cramping occurs.
DoseMagnesium: standard supplemental dose daily throughout the fast. Bone broth: as needed (1–2 cups). Phosphatidylserine: evening dose. GABA: bedtime.
For whomAnyone fasting for more than 36 hours, especially T2D patients with baseline magnesium deficiency or a history of cramps and sleep disruption.
WhyThe kidneys can defend sodium well in healthy individuals but T2D patients are often magnesium-depleted at baseline and cannot manage the supplementation themselves. The fasting sympathetic surge (noradrenaline elevation) is real and predictable; warning patients in advance and providing tools to manage it (phosphatidylserine, GABA) dramatically improves adherence.
CaveatsSodium depletion in truly healthy kidneys is rare; the inter-Salt study shows some populations survive on near-zero salt. Routine sodium supplementation is not necessary. Bone broth introduces calories and amino acids — technically breaking the fast — but is acceptable to prevent abandonment.
Fung is explicit that his clinical practice does not include proactive salt loading — he disagrees with the popular advice to add salt to water during fasts. The kidneys of healthy individuals can reabsorb all necessary sodium. He does supplement magnesium reactively. Attia adds phosphatidylserine specifically for the adrenal glands — it blunts cortisol and noradrenaline — then layers in standard sleep supplements (GABA, potentially glycine or L-theanine) which produce what he describes as some of the best sleep he gets, once in ketosis.
Magnesium is a bit more difficult because a lot of type 2 diabetics are depleted of magnesium to start, so we'll often recommend magnesium supplements, and that's where a lot of the cramps kick in. Some people do get a little dizzy and so on — so that's when we'll use the bone broth, where you can put a decent amount of salt in.
Also said
“We recommend that patients take phosphatidylserine with the fast in the evening and that sort of calms down the adrenal glands and then even just an oral over-the-counter GABA... that coupled with some other sleep supplements actually tends to help people sleep.”— Attia's specific addition to Fung's protocol — managing the sympathetic surge pharmacologically to improve sleep compliance.
Pre-fasting patient preparation: warn all side-effects in advance
WhatBefore starting any patient on a fasting protocol, explicitly pre-brief them on: expected hunger (present but manageable even on day 7); sleep disruption (sympathetic system activation); headaches (first 2–3 days, then resolves); muscle cramps (magnesium depletion); diarrhea or constipation; blood sugar monitoring requirements; medication adjustment plan.
WhenAt the clinical appointment before initiating the fast — not on the first day of the fast.
DoseA single briefing session covering all expected side-effects. The key message is: these symptoms are expected, temporary, and do not indicate the fast is harming you.
For whomAll patients initiating any fasting protocol of 36 hours or longer.
WhyForewarned patients interpret symptoms as expected milestones rather than emergencies. When the headache arrives on day 2 and then resolves on day 3 exactly as predicted, patient trust and confidence increase. The most common reason patients abandon fasting is psychological — fear of unknown symptoms, not the symptoms themselves.
Fung describes this as similar to the book 'What to Expect When You're Expecting' — it does not make the contractions less painful, but knowing they are normal and temporary makes them dramatically more manageable. He emphasizes that the physical side of fasting is 'super easy to deal with' compared to the psychological side, and that most of his effort in his clinical practice goes into creating the right social/psychological environment for fasting success.
We warn them ahead of time of all the potential problems. It doesn't make the symptoms any better but it makes dealing with it a lot better. So we tell them hey look, you can have headaches, those will go away. You can have cramps, this is what to do. You can have diarrhea, this is what to do. Sleep is — you can expect this.
16:8 time-restricted eating as first-line fasting intervention
WhatRestrict all eating to an 8-hour window (Fung example: noon to 8 PM). No changes to diet composition required initially — just compress the eating window.
WhenAs an entry-level intervention for pre-diabetics, mildly insulin-resistant patients, or anyone transitioning from a ~15-hour daily eating window (what Satchin Panda's research documented as the modern default).
DoseDaily. 16 hours fasting / 8 hours eating. The 1970s baseline was approximately 14 hours fasting; 16:8 is a modest extension.
For whomAnyone wanting to improve metabolic health without major dietary overhaul. Particularly useful for culturally diverse patients who find carbohydrate restriction culturally incompatible.
WhyFung notes that compressing a person's existing diet into an 8-hour window produces clinical improvement even without dietary changes, because the previous feeding window had expanded to 14–15 hours of continuous eating. Time restriction allows insulin to fall for at least 16 hours per day, which creates the metabolic space for fat mobilization.
CaveatsResults are better if diet quality is also improved within the window, but 16:8 alone produces meaningful benefits over a 15-hour feeding window. Attia is skeptical that 16:8 alone provides large biological benefits — he thinks the primary value is psychological: proving to patients that not eating for 16 hours does not cause the sky to fall.
We do see people get better like we see people change a lot because... we did sort of one meal a day where we squished it even further and those people basically they didn't even change their diet that much.
Also said
“It's always about practicalities — is it easier? If you have people, I say two strategies: one is counting carbs, which I think can work, versus counting the number of hours that you don't eat. It's so much easier to tell people between 12 and 8 is your eating window.”— Fung's argument for time restriction over macronutrient counting as a simpler patient-facing tool.
Periodic 5-day fasting for healthy individuals (longevity maintenance)
WhatFor non-obese, non-diabetic, insulin-sensitive individuals: one 5-day fast per year, or periodic longer fasts calibrated to personal goals. Can be water-only or include bone broth and electrolytes.
WhenFor healthy people with no metabolic disease who want the longevity benefits of periodic deep ketosis, mTOR suppression, and potential immune reset (per Valter Longo's data on 7-day fasting).
Dose5 days per year minimum. Fung frames the risk-to-reward arithmetic: 5 days of fasting means missing 15 meals out of a yearly ~1,000 — a 1.5% caloric interruption with near-zero physiological risk.
For whomMetabolically healthy adults interested in longevity and chronic disease prevention. Pre-loading with ketosis (Attia protocol) is strongly recommended to make the 5 days tolerable.
WhyPeriodic deep insulin and mTOR suppression activates autophagy-like cellular recycling pathways that are impossible to reach during even the healthiest fed state. Fung cites Longo's work on 7-day fasts resetting immune function as a potential application even for autoimmune conditions where no drug evidence exists.
Fung presents this using a risk-benefit framework: missing 15 out of 1,000 meals yearly carries near-zero physiological risk in a healthy person. The potential benefits — mTOR suppression, insulin lowering, immune modulation — may be substantial even if unproven in RCTs. His framing is: 'Maybe there's no benefit; maybe there is. I can't tell you. But the risk is so low — why wouldn't you do it?' He contrasts this favorably with drugs that have number-needed-to-treat of 50, meaning 49 out of 50 patients receive all the side effects with no benefit.
In a year you will eat three meals a day — you'll eat a thousand meals. For five days you're gonna miss 15 meals out of a thousand. That's it. What is the risk of that? I'll tell you it's just about zero.
What's new
Personal practice updates, fresh positions, predictions
6 items
Insulin resistance is overflow, not a broken lock-and-key
~1 h 10 min
Fung's reframe: 'insulin resistance' is not a cell that stopped listening to insulin. It is a cell that is already full — a suitcase with no room left. The evidence is that insulin still works normally on fat tissue (driving obesity), still stimulates growth (PCOS, cancer), still makes VLDL — it only 'fails' at glucose uptake because the liver is already packed. The conventional lock-and-key model explains none of these paradoxes; the overflow model explains all of them.
Why this matters: Reframes three decades of T2D pharmacology as the wrong treatment for the wrong disease. If the suitcase is full, more insulin makes it worse — which is exactly what the ACCORD, ADVANCE, and VA-DT trials showed: aggressive glucose lowering with insulin produced no macro-vascular benefit and sometimes increased mortality.
Background
The lock-and-key / insulin resistance model dominated from the 1970s onward, leading to sulfonylureas → more insulin → more insulin as the standard progression for T2D.
Fung draws on Roy Taylor's Counterpoint and Direct studies (Newcastle, UK) to explain the mechanism: liver fat causes the overflow (insulin resistance), pancreatic fat causes beta-cell failure. Drain both by fasting and you reverse T2D. Lipodystrophy patients — who have no fat storage capacity at all — have paradoxically the worst insulin resistance ever measured, because all the glucose/fat that should be safely stored in subcutaneous adipose overflows directly into the liver. This validates the overflow model: obesity is not the disease, it is the body's protective response to the real disease, which is whole-body glucose excess.
I use an analogy like suppose you have a suitcase and you're putting your shirts in your suitcase which is fine... at some point your wife says here put these two shirts into your suitcase but you don't put them in — either the suitcase doesn't open or it's just full.
Also said
“The solution is to get rid of some of the stuff that's in your suitcase... The solution is to empty that cell of glucose. That's the whole solution, which is a totally different paradigm than what we've been taught.”— States the therapeutic implication of the overflow model directly — treatment must remove glucose, not force more in.
“This is why obesity itself is actually a protective mechanism against hyperinsulinemia, against too much sugar.”— The most heterodox claim: fat gain is the body's defensive response, not the root pathology.
SGLT2 inhibitors work because they empty the suitcase, not because they lower blood glucose
~1 h 25 min
SGLT2 inhibitors make patients pee out sugar — they barely move HbA1c (0.3–0.4 points) but reduce kidney disease risk by ~25% and cardiovascular events in ways no prior diabetes drug ever achieved. Fung's explanation: for the first time a drug is treating the actual pathology (whole-body glucose excess) rather than shuffling glucose from blood into a liver that is already full.
Why this matters: The clinical trials (EMPA-REG OUTCOME, CREDENCE etc.) showed end-organ protection that was completely disproportionate to glycemic improvement — a paradox that only makes sense under the overflow model.
Background
Prior diabetes drug classes (sulfonylureas, DPP-4 inhibitors, insulin) all work by amplifying the insulin signal — making the wife scream louder — which the overflow model predicts will worsen the underlying pathology over time.
Attia notes this was a key aha moment for him around 2008: if tight glucose control (via more insulin) did not improve macro-vascular outcomes in ACCORD/ADVANCE, and if SGLT2 inhibitors dramatically improved end-organ outcomes despite minimal glucose lowering, then blood glucose is not the disease — it is a downstream marker. The disease is whole-body glucose excess, and the correct treatment removes glucose from the system rather than redistributing it between compartments.
The sglt2 are a new class of drug and what they do is they actually make you pee out the sugar... every single one of these trials shows end-organ protection with almost no benefit to your blood sugar... because now you're actually treating the underlying problem.
24-hour fasting 3x/week reversed 20-year T2D in 5–18 days
~2 h 40 min
Fung's published case series (BMJ Case Reports): three patients with 20–25 years of T2D on large insulin doses were put on 24-hour fasts 3×/week (one meal a day) with low-carb guidance. All three got off insulin within 5–18 days. Years later they remained non-diabetic. He calls this 'Medical Bariatrics.'
Why this matters: The only insurance-recognized reversal of T2D was previously bariatric surgery. Fung's clinical series suggests fasting can achieve the same result non-surgically, with the same physiological mechanism: rapid reduction in hepatic fat and pancreatic fat.
Background
Bariatric surgery reverses T2D within 10 days pre-weight-loss, which nobody could explain until Taylor's counterpoint study showed the mechanism is rapid hepatic and pancreatic fat reduction. Fung's insight was that fasting does the same thing.
Fung explains the mechanism step by step: after 24 hours glycogen is depleted and the body is forced to pull fat from internal organs first — liver fat drops first (reversing insulin resistance), then pancreatic fat drops (restoring beta cell function). This matches Taylor's MRI data showing liver fat falling before meaningful weight loss. The diabetes resolves before the patient has lost significant subcutaneous fat, which is exactly what happens post-bariatric surgery too.
We put on 24 hours of fasting 3 times a week which is like a one meal a day right... We got off all their insulin like between 5 and 18 days. It was ridiculous how quickly they responded. And years later we still have people who are like non-diabetic.
Also said
“Fasting is what I call Medical Bariatrics. You get all the benefits of bariatric surgery without doing any surgery.”— Fung's framing of fasting as a non-surgical equivalent to bariatric surgery.
Hyperinsulinemia drives cancer risk via the same PI3K/mTOR growth pathway
~1 h 20 min
Insulin is a growth factor — that is why bodybuilders use it. Via PI3K, it activates the same downstream signaling (Akt, mTOR) that cancer cells need to proliferate. Thirteen cancers are classified as obesity-related; breast and colon cancer rates are dramatically elevated in T2D patients, even when controlling for obesity. Fung frames cardiovascular disease, Alzheimer's, and cancer as diseases of 'too much growth' — hyper-proliferation driven by chronically elevated insulin.
Why this matters: Connects metabolic disease to oncology and neurodegeneration through a single mechanistic framework. It also explains why rapamycin (mTOR inhibitor) is a candidate longevity drug — it targets the same hyper-growth axis.
Attia notes that a theorist predicted in 2006 that rapamycin would be the most important longevity drug, three years before the first animal data. The same logic applies to fasting: every time you fast, you suppress insulin and mTOR simultaneously, creating the metabolic conditions that are anti-proliferative and pro-longevity. Fung's framing — 'we shifted from infectious diseases to hyper-proliferative diseases but never changed our pharmaceutical paradigm' — explains why giving patients more drugs rarely treats chronic disease.
Insulin is a growth factor. It has huge implications for cancer... Breast cancer, colon cancer — very insulin sensitive, much higher rates in type 2 diabetes and obesity. So you're getting the cancer-causing effects of the insulin but yet you're saying people are insulin resistant.
Attia vs Fung: ketogenic pre-loading before extended fasts
~3 h 05 min
Attia and Fung disagree on entry protocol. Fung puts severe T2D patients directly into 7-day water-only fasts cold turkey when urgency demands it (e.g., a patient facing amputation). Attia never does this — he spends a week in nutritional ketosis first, arriving at the fast with BHB between 0.5–1 mmol, which he says makes the fast dramatically easier. The disagreement reflects different patient populations: Fung's are obligate glycogen burners; Attia's are already fat-adapted.
Why this matters: Practical calibration point for anyone attempting extended fasting: the entry state (metabolic flexibility vs glycogen dependency) determines how brutal the first 3–5 days feel.
Personal experience
Attia: 'I usually spend a week in ketosis before the fast, which means I'm showing up in a fast with a ketone level of somewhere between 0.5 and 1 millimolar of beta-hydroxybutyrate. I sailed generally pretty easily into these fasts.' Fung on his severe patients: 'There's no easy way. We just tell them this is like in a sock for a week, because you're not gonna feel that great.'
I usually spend a week in ketosis before the fast which means I'm showing up in a fast with a ketone level of somewhere between 0.5 and 1 millimolar of beta-hydroxybutyrate.
Fasting is culturally universal — the compliance barrier is social not physiological
~2 h 35 min
Fung notes that Ramadan demonstrates hundreds of millions of Muslims fast successfully for a month. Lent, Yom Kippur, Orthodox Christian fasts — every major religion has a fasting tradition. The compliance barrier with clinical fasting is not physical hunger (which he says is manageable) but social isolation: patients fast while surrounded by people who think they are doing something dangerous or extreme.
Why this matters: Reframes the compliance problem as an environment design issue rather than a willpower issue — the same insight from behavioral science applied to a metabolic intervention.
How do they all fast? Because people say 'Oh I could never fast.' But you know that literally hundreds of millions of Muslims fast. It's because the environment is supportive. If all your friends are doing it and all your family — it's not fun but it's not that difficult to actually do it.
Disclosed sponsorships3speaker disclosed
The Obesity Code by Jason Fung
Book Sponsored · disclosed
Fung's primary text arguing that obesity is a hormonal disease (hyperinsulinemia) rather than a caloric disease. The foundational read for the overflow paradigm discussed throughout the episode.
DisclosureFung is the author and the guest on this episode — direct promotion of his own book.
Attia describes it in the intro as one of the works that led listeners to demand this interview, citing it alongside The Diabetes Code and The Complete Guide to Fasting as the trilogy that brought clinical fasting into mainstream discussion.
He's the author of several books including The Obesity Code, The Diabetes Code, The Complete Guide to Fasting, and The Longevity Solution.
The clinical companion to The Obesity Code — focused specifically on T2D reversal using fasting and low-carbohydrate approaches. Contains the case series and protocol details discussed in the episode.
DisclosureFung is the author.
Fung notes he published these books rather than academic papers specifically because peer review would have killed the work, and because books reach patients directly, creating the cultural and social normalization that makes fasting compliance possible.
I wrote a couple of books about this of course and that's the point — if I wrote a couple of articles to get peer-reviewed, first of all the peer review would kill it, and then second of all you'd get no traction whatsoever.
Fung's clinical program in Toronto that has treated thousands of patients with fasting and low-carbohydrate protocols for T2D and obesity. The real-world evidence base for everything described in the episode.
DisclosureFung is the co-founder — direct promotion of his own clinical practice.
Fung emphasizes that the IDM program is not academic — it is clinical practice iterating in real time. He explicitly contrasts his clinical experience (zero patients stopped due to muscle loss concerns across six years and thousands of patients) with academic critics who have never treated patients with fasting.
In six years it's like thousands of patients — zero people had to stop because I was worried about muscle loss.
Lines worth pulling out — contrarian, specific, or perfectly phrased
6 items
I use an analogy like suppose you have a suitcase and you're putting your shirts in your suitcase which is fine... either the suitcase doesn't open or it's just full. There are two possibilities and they're completely different.
The core paradigm shift of the entire episode: whether T2D is a signaling defect (lock-and-key) or a storage overflow changes the entire treatment logic from 'force more insulin in' to 'empty the suitcase.'
Fasting is what I call Medical Bariatrics. You get all the benefits of bariatric surgery without doing any surgery.
Fung's one-line clinical framing that explains why fasting reverses T2D: it does the same thing as bariatric surgery (rapid visceral fat mobilization) non-invasively.
This is why obesity itself is actually a protective mechanism against hyperinsulinemia, against too much sugar.
The most heterodox statement in the episode — reframes fat accumulation as the body's defensive response to glucose excess rather than the root pathology.
Insulin resistance is caused by too much insulin — same as antibiotic resistance is caused by too much antibiotics. We treat insulin resistance with more insulin, which is what caused it in the first place.
Crystallizes why standard T2D treatment is self-defeating and why the insulin escalation ladder produces progressive worsening rather than improvement.
The risk of not doing something — the virtual certainty that you're going to in ten years go on dialysis — that's not an insignificant risk. It's actually a huge risk compared to doing fasting which has been done for thousands of years.
Attia and Fung's shared argument against the 'we don't have evidence for fasting' objection: the risk of doing nothing (progressive T2D, dialysis) is far larger than the risk of an intervention with thousands of years of human safety data.
In a year you will eat three meals a day — you'll eat a thousand meals. For five days you're gonna miss 15 meals out of a thousand. That's it. What is the risk of that? I'll tell you it's just about zero.
Fung's risk-framing for healthy-person periodic fasting — strips away the mystique and makes the cost-benefit calculation obvious.
Sign in to share feedback
Tell us if this brief hit the mark or missed it — feedback feeds back into the next iteration of the prompt.
Reading is free for everyone. A free account adds the personal layer: save protocols, follow experts, and see how the other experts weigh in on this same topic.
Educational summary of the cited expert source — not medical advice. Open the source recording linked above and consult a qualified physician before acting on any protocol.
