Metabolic flexibility: how well your body switches between fat and carbohydrate

Metabolic flexibility is your body's ability to switch cleanly between burning fat and burning carbohydrate as the situation demands: fat when you are fasting or moving gently, carbohydrate when you have just eaten or are working hard. A flexible metabolism reads the fuel on offer and oxidises it; an inflexible one gets stuck, oxidising fat poorly when fasted and failing to switch fully to glucose after a meal. That stuckness, first described in human muscle by David Kelley and Bret Goodpaster, tracks closely with insulin resistance, obesity and type 2 diabetes, which is why it has become a useful lens on metabolic health, as long as you ignore the fad framing around it.

What metabolic flexibility actually means

The phrase entered the literature through a specific human experiment. In 1999, David Kelley, Bret Goodpaster and colleagues measured fuel use across the leg (mostly skeletal muscle) in lean and obese people, both after an overnight fast and during insulin infusion. Lean muscle did the expected thing: it leaned on fat during fasting, then switched smartly toward glucose when insulin rose. Obese, insulin-resistant muscle did neither well. It oxidised relatively less fat when fasted and switched less toward glucose under insulin. Kelley later named the healthy version metabolic flexibility and the impaired version metabolic inflexibility.¹

It helps to be precise about the two everyday transitions involved. The fasting-to-fed switch is about responding to insulin: a meal raises glucose and insulin, and flexible tissue suppresses fat burning and ramps up glucose uptake, oxidation and storage. The rest-to-exercise switch is about responding to demand: gentle movement is fuelled largely by fat, while hard effort recruits more carbohydrate. A genuinely flexible metabolism handles both gear changes; an inflexible one slips the clutch on one or both.²

How it is measured: RER and indirect calorimetry

You cannot see fuel switching directly, so researchers infer it from breath. Indirect calorimetry measures the oxygen you consume and the carbon dioxide you produce, and the ratio of the two, the respiratory exchange ratio (RER, also called respiratory quotient), reveals the fuel mix. Burning carbohydrate produces more carbon dioxide per unit of oxygen than burning fat, so RER rises toward 1.0 on carbohydrate and falls toward 0.7 on fat.³

Flexibility is then expressed as a change in RER, not a single value. Two challenges are standard. The first is the fasting-to-fed (insulin) switch: measure RER fasted, then again during a hyperinsulinaemic-euglycaemic clamp or after an oral glucose load, and a larger rise (a bigger delta-RER) means a more flexible response. The second is the rest-to-exercise switch, where RER is tracked as intensity climbs. Jose Galgani, Cedric Moro and Eric Ravussin set out the framework for interpreting these deltas, and stressed that flexibility must be judged against how much glucose was actually disposed of, otherwise you can mistake poor glucose uptake for good fat burning.³ This is laboratory work; no wearable measures it.

Why inflexibility tracks with insulin resistance

The link runs through an old idea, refined. In the 1960s Philip Randle proposed a glucose-fatty-acid cycle: when fat is plentiful and being oxidised, it competitively suppresses glucose oxidation. The original reading was that excess fat oxidation caused insulin resistance. When Kelley and Mandarino reexamined human muscle in 2000, they found something subtler. Insulin-resistant muscle did not over-oxidise fat in the fasted state; if anything it under-oxidised it, while leaning relatively more on glucose at rest, and then failed to ramp glucose oxidation up properly once insulin arrived. The defect was not too much fat burning but a loss of the ability to switch on cue, in either direction.⁴

That reframing matters because it ties flexibility to the same machinery as insulin resistance: the handling of fat inside the muscle cell. When fatty acids arrive faster than mitochondria can fully oxidise them, partially processed lipid species (diacylglycerols, ceramides) accumulate and interfere with insulin signalling. So inflexibility and insulin resistance are not merely correlated; they are two readings of the same underlying problem, fuel arriving faster than it can be cleanly used. We unpack the insulin side of this in detail in our guide to the metabolic health library and specifically in insulin resistance: early signs and reversal.²

One honest nuance from the most recent evidence. A 2025 systematic review and meta-analysis by Hansen and colleagues pooled 65 studies and analysed 985 people (256 lean, 497 with overweight or obesity, 232 with type 2 diabetes). Lean people were clearly more flexible, with a larger insulin-stimulated rise in RER (delta-RER about 0.10 versus 0.07). But once the analysis accounted for how much glucose each group actually disposed of, the gap between the overweight and diabetes groups largely disappeared. Their conclusion: metabolic inflexibility is driven more by body-fat and weight status than by a type 2 diabetes diagnosis as such.⁶

Mitochondria and muscle: where the switching happens

Skeletal muscle is the main stage because it is the body's largest sink for glucose after a meal and a major site of fat oxidation at rest. Within muscle, the switching is done by mitochondria, the organelles that burn both fuels. Two things govern how well they switch: how many you have (mitochondrial content or density) and how well each one works (oxidative capacity and the enzymes that route fat versus glucose into the furnace).²

Endurance-trained people sit at the flexible end. Their muscle carries more mitochondria and more of the transport proteins that ferry fatty acids into them, so they oxidise fat well at low intensities and still switch hard to carbohydrate when they sprint. People with metabolic syndrome sit at the other end: lower mitochondrial oxidative capacity, an earlier and steeper switch from fat to carbohydrate as exercise intensity rises, and rising blood lactate at workloads a fit person would barely notice. Inigo San-Millan and George Brooks used exactly these signals, fat oxidation and blood lactate during graded exercise, to grade flexibility across professional cyclists, recreational exercisers and people with metabolic syndrome.⁷

It is worth being careful about cause and direction here. Trained muscle is more flexible, but we cannot cleanly separate how much is training versus genetics versus simply not carrying excess ectopic fat. The fair summary is that mitochondrial quantity and quality are central to fuel switching, and that they are among the few features of metabolism you can meaningfully change with behaviour.² We cover the organelle itself in our piece on mitochondrial health and energy.

What may improve it: fitness, fasting, low-carb, Zone 2

Here is the evidence, sorted by how well it holds up, and stripped of the "train your body to burn fat" overpromising.

Exercise and weight loss: the strongest case

The best human evidence is for exercise combined with fat loss. Goodpaster, Katsiaras and Kelley put 25 obese adults without diabetes through 16 weeks of moderate physical activity plus modest calorie reduction, with fuel use and glucose disposal measured by indirect calorimetry and a clamp. Fasting fat oxidation rose and insulin-stimulated glucose disposal improved together, with the two changes correlated, which is close to a textbook demonstration of flexibility improving.⁵ Separate work shows endurance training expands mitochondrial content and fat-transport capacity, the structural basis for that shift.⁸

Zone 2 training: plausible, mechanistically sensible

Zone 2 is steady aerobic work at an intensity where you could still hold a conversation, roughly where lactate is produced and cleared at the same rate. Mechanistically it is well suited to flexibility: it is largely fat-fuelled, and done consistently it drives mitochondrial biogenesis, which should widen the range over which you can oxidise fat. The human flexibility-specific trial data are thinner than the enthusiasm suggests, so treat Zone 2 as a sound, low-risk way to build the relevant machinery rather than a proven flexibility cure. We go deeper, with the honest caveats, in our Zone 2 and VO2 max guide.⁷

Fasting and low-carbohydrate periods: real but easy to over-read

Fasting and low-carbohydrate eating reliably push RER down and increase fat oxidation, because you have simply given the body less carbohydrate to burn. That is a substrate-availability effect, and it is not the same as proving your mitochondria have become better switchers. A genuinely keto-adapted person oxidises a lot of fat, but may also show a blunted ability to switch back to carbohydrate quickly, which is flexibility in one direction, not both.² The defensible reading: occasional fasting or lower-carbohydrate periods can be a useful stimulus and, mainly by helping people eat less and lose fat, can improve the underlying picture, but the fall in RER on its own is not the prize.³

The honest limits of "metabolism" claims and CGMs

Two cautions, because this is where marketing outruns the science.

First, consumer "metabolism" and "fat-burning" claims rarely measure metabolic flexibility at all. Real measurement needs a metabolic cart and usually a clamp or glucose challenge. Supplements, teas and apps that promise to "boost metabolism" or "switch on fat-burning" are not assessing the delta-RER that researchers actually use, and "burning more fat at rest" is not a health outcome in itself: across a full day a flexible metabolism burns whatever you supply, and body fat is still governed by total energy balance.²

Second, continuous glucose monitors (CGMs) are increasingly sold to people without diabetes as a window onto metabolic health, and they are at best an indirect, noisy proxy. A flat glucose curve is loosely consistent with good flexibility, but CGMs do not measure fat oxidation or fuel switching, and in people without diabetes the spikes you see often sit within a normal physiological range. In 2024 the US regulator cleared over-the-counter CGMs for people without diabetes, but how to interpret the readings in that group remains poorly defined, and 2025 work from Mass General Brigham found these devices can misrepresent true glucose control in non-diabetic users, sometimes flagging "highs" that overstate any problem.⁹ A CGM can be a useful behavioural nudge; it is not a flexibility test. Our take on glucose, mood and CGMs lives in the insights library.

Practical levers that hold up

The interventions with the best evidence are unglamorous and overlap heavily with everything that improves insulin sensitivity.

• Build and use muscle. Muscle is the main glucose sink and a major fat-oxidising tissue; more of it, used regularly, widens your fuel-switching range. Combine resistance training with regular aerobic work.⁵
• Train aerobically, often. Regular Zone 2-style sessions build mitochondrial content, the structural basis of flexibility, and even single sessions improve insulin action for a day or more.⁷
• Stop grazing. Leave clear gaps between meals and avoid late-night eating so the fasting-to-fed switch can operate across its full range.²
• Lose excess ectopic fat. Since inflexibility tracks weight and body-fat status more than diabetes per se, reducing visceral and liver fat is one of the most direct levers.⁶
• Protect sleep. Short sleep worsens insulin sensitivity in controlled studies, and the same machinery underlies flexibility.²

If you want to turn this into a concrete plan, our start here guide sequences these changes, and the stack builder can map your situation to candidate interventions without overpromising.

What to ask your GP

What to do next

This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.