Hacker News Reader

Comment by waterheater | Hacker News Reader

A shift in understanding the metabolism has been gradually occurring since the 1990s with the isolation of leptin.

What is leptin? Discovered in 1994, leptin is a hormone that signals the availability of energy reserves in the body. "High leptin levels are interpreted by the brain that energy reserves are high, whereas low leptin levels indicate that energy reserves are low", the latter causing the body to enter a resource-minimizing starvation mode [1]. More recently, leptin has also been shown to regulate "neuroendocrine and immune function and [play] a role in development" [2].

Why is leptin so important to obesity? "Most obesity occurs in the presence of increased leptin levels", which suggests a form of "leptin resistance" in obese individuals [2].

Leptin changes our understanding of body fat, also called adipose tissue. We generally think of body fat as simply stored energy reserves, so fat can be burned by decreasing our caloric intake (diet) and increasing our energy usage (exercise), forcing our body to dip into its reserves. This approach is generally referred to the calorie-in/calorie-out (CICO) model. While true to an extent, this perspective is too simplistic, given one key fact: leptin is primarily produced in adipose tissue [1]. This fact means body fat is actually an endocrine system organ.

Consider this: an obese individual has elevated leptin levels. A decrease in adipose tissue causes a decrease in their leptin production. Due to leptin resistance, low leptin levels causes the body to enter starvation mode, and hardcore dieting causes leptin levels to drop further, which makes weight loss even more difficult. What they mean by "metabolic adaptation" may be understood as "leptin adaptation", though I'm just speculating there.

It's strange that so many articles talk about weight loss and "metabolism adaptation" without mentioning the hormone signalling energy availability. With leptin, we clearly see the body has an energy availability feedback loop, and fighting the intrinsic dynamics of a feedback loop rarely leads to good outcomes.

On this vein, I want to highlight slothtrap's comments:

>The more frequent the dieting and steeper the caloric deficit, the worse metabolic adaptation gets, and the longer it takes to return to normal.

>I wonder if those who sustained higher levels of exercise were actually in a steeper caloric deficit.

>perhaps it's fair to say that if you're going to exercise, you best sustain it.

These comments align with present understanding of leptin dynamics. A steep caloric deficit pushes the obese body into starvation mode, so net energy expenditure from high levels of exercise will be decreased. Weight loss will be maintained once the body adjusts its leptin production and usage, which, as of now, can only be accomplished with habit.

It's also worth mentioning here the hormone ghrelin, which was identified in 1999 and is nicknamed the "hunger hormone" [3].

[1] https://en.wikipedia.org/wiki/Leptin

[2] DOI: 10.1146/annurev.physiol.62.1.413

[3] https://en.wikipedia.org/wiki/Ghrelin

View on HN

> This approach is generally referred to the calorie-in/calorie-out (CICO) model. While true to an extent, this perspective is too simplistic, given one key fact: leptin is primarily produced in adipose tissue [1]. This fact means body fat is actually an endocrine system organ.

This part I think needs more explanation. The reason calories-in/calories-out model is too simplistic (I would call it almost useless if not completely wrong) is because of mitochondrial uncoupling proteins [1] that control how much energy in the protein gradient gets converted to ATP and how much is lost to waste heat. These uncoupling proteins are upregulated by leptin so when your leptin levels fall, your body literally changes how efficient its metabolism is to do more work with fewer calories. Body fat isn't just an endocrine organ, it might be the primary metabolic controller with a nasty catch 22 feedback loop - which would help explain why a certain minimum percentage of body fat is required or else a person enter organ failure.

That's why there are some people who don't seem to gain much weight despite eating a lot of food: their bodies are naturally less efficient and lose a lot more energy to thermogenesis. Without taking that metabolic adaptation into account, CICO is nothing but harmful dogma that needs to die. Otherwise people are fighting what might be one of the most important biological adaptations humans have.

[1] https://en.wikipedia.org/wiki/Uncoupling_protein

CICO doesn’t work across individuals for reason you outline (just because I lose weight with a certain level of calories, doesn’t mean you will, etc.)

But for a single individual, working to establish your base metabolic rate by counting every calorie and weighing yourself regularly, you can always find a way to be successful with the CICO model. If you establish a baseline that your metabolic rate is (say) 2000 calories a day, then eating 1600 calories a day will cause you to lose roughly a half-pound to a pound per week. It does not mean that every individual’s baseline metabolic rate is the same.

As you also outline, your base metabolic rate also changes as your weight does, so re-baselining by continually measuring caloric intake and your weight to recalibrate is always going to be necessary.

Given the above CICO almost seems tautological: Measure the weight loss you actually observe at a given calorie level, and that tells you… how much weight you will lose at that given calorie level. So it is almost useless as you describe.

BUT, the main reason I’m still a proponent of CICO is that it simplifies dieting away from thinking about what you eat, and instead focuses on how much you eat. If I establish that I need to eat 1800 calories a day, I can still have a Big Mac meal and drink some beers. But I may not be able to eat much else that day. That alone makes it useful for me. What isn’t useful to me, are diets that essentially limit the type of food you eat and tell you to just go nuts and eat however much you want: I will often find myself hating the foods I’m restricted to, and the quantity I eat may end up being such that I gain weight anyway.

To me, CICO is simply a technique that says “actually measure how much you’re eating”, no more and no less. That is what makes it a useful tool.

This is the dogma that I'm talking about.

> If you establish a baseline that your metabolic rate is (say) 2000 calories a day, then eating 1600 calories a day will cause you to lose roughly a half-pound to a pound per week.

At this point it's pretty safe to say that's likely untrue at the population level. This is a cutting edge area of research but the UCP1 adaptations seems to happen really fast, for some people on the order of weeks. On the order of months for the vast majority of the population. It doesn't take long for the basal rate to fall to whatever your intake is because the system evolved to naturally have a lot of room to spare. The human body always adapts and if I were a gambling man, I'd say this is largely why most people fail at dieting: after a month or two the results stop and they have to further decrease their intake which interferes with other pathways and tanks motivation.

If it works for you, that's great! You're one of the lucky ones, maybe because you've hit your floor for uncoupling proteins. But given just how hard dieting is for most people and how bad the obesity epidemic has gotten, it's probably not useful for the majority of people. A far more holistic approach that tackles macro- and micronutrient deficiency, motivation, lifestyle, and not just net calories is required.

> It doesn't take long for the basal rate to fall to whatever your intake is because the system evolved to naturally have a lot of room to spare

All this is accounted for in this sentence of my post:

> your base metabolic rate also changes as your weight does, so re-baselining by continually measuring caloric intake and your weight to recalibrate is always going to be necessary

If you’re actually measuring yourself at the scale to get the ground truth of your actual metabolic rate, and you’re doing this every single day, then there isn’t a single thing that can change this: UCP1 adaptations, fat viruses, leptin changes, hormones, stress… all of this is covered by the actual weight on the scale. If I ate X calories and gained Y lbs, I need to adjust my calories by Z, no matter the reason I gained those pounds.

You can argue that such a diet is too hard, and that for some people the adjustment to your calories is going to leave you essentially starving, but it doesn’t change the truth of the matter that it will work if you do it right. Now, whether it’s worth it for people, and whether the process is going to leave you utterly miserable or even extremely unhealthy as a result, I’m making no claim on.

Uhh... Isn't the whole thing pointless because of your last sentence?

Empirically it's really quite rare that eating less than your caloric expenditure is going to make you unhealthy for obese people. It's a rare caveat, not ground for blanket dismissal.

Feeling hungry or being unhappy with such a diet is more common, but it's generally also fixable, and it doesn't defeat the point either.

No, I’m saying I’m making no claim on that because I very much don’t want to get into the (very) subjective and context-dependent argument about whether a diet is a good idea for someone or not. My claim is much narrower: CICO works if you follow it. Whether it’s right for you, I can’t say.

(For my part, CICO worked amazingly well for me, and I was able to follow it without much trouble. But that’s just me.)

Now, whether it’s worth it for people, and whether the process is going to leave you utterly miserable or even extremely unhealthy as a result, I’m making no claim on.

then CICO is not practical for weight loss. the "CO" becomes so small that you're screwed

The cope.

What normal human reduces its CO so that you wont lose weight? Are you talking about someone in a coma?

You will lose weight, in fact if you cut to 0 for long enough you are guaranteed to lose all soft tissue weight. Can I recommend you the “0 calories till you die” diet? The issue is that:

- many people can cut from a diet of 2200 calories per day to 1800 calories per day and pretty quickly and sustainably lose weight. This kind of calorie reduction is very manageable, and anyone who has lost weight this way starts saying “it’s just CICO man, so easy, just do it”

- meanwhile, in order to lose weight, some (probably not huge but very real) amount of people who maintain at 2200 calories need to drop to 1000 calories per day, or even lower, consistently, for months, in order to lose weight at a 0.5 lb/wk. This is miserable, not sustainable, and probably a net negative on your life (damage not worth the weight loss) unless you are morbidly obese.

The people in category 1 yell “you’re just coping, it’s not hard, just CICO” at group 2. Meanwhile they have never had to reduce calories in that way and have no idea how unsustainable it is. They are so sure their experience is universal, that they usually utterly deny the other groups existence, and smugly go about the world assuming people struggling with weight are just weak.

I get what you are saying. And you put forward a nice argument. The problem is that the ANTI-CICO-Crowd takes the stance that "CICO WONT WORK BECAUSE MY CO IS REDUCED HAH TAKE THAT CICO-FUCKTARDS"

Which is stupid.

If you want to loose weight and maintain a reasonable BMI during the long run you need 2 things.

1) A healthy diet, if you are in America this requires a bit of planning, if you live in Europe its easier. Limit the portion size (and if you are fat you don't know what a normal portion size is, look to you skinnier friends for inspiration).

2) Get your heart rate up a couple of times a week. If you are fat this might be the _walk_ to the store. If you are just weak weight training is great. And for 99% you most likely should aim to increase the amount of condition training, brisk walks, joggins, biking, chopping wood whatever.

Diet and exercise are indeed the keys to weight loss.

By focusing exclusively on CICO, one dismisses the fact that people dislike feeling hungry. To say it's simply an issue of willpower dismisses the importance role of hormones in driving behavior. When it comes to weight loss, some people are dominated by their hormonal responses, while others have more regulated responses.

I think I've found a good analogy: advocating CICO as primary strategy to lose weight is like saying all you need to do to complete a road track is to drive forward and stay on the track. The track shape analogizes to the unique dynamics of a person's hormonal feedback loops. Completing the race analogizes to hitting a weight loss goal, while going for fastest lap time is like a doing a 30-day weight loss challenge.

There's roughly 8 billion unique tracks in the world. Some tracks may be a drag strip, others may be Nurburgring-style circuits, and an unlucky few get the Baja 1000. Virtually everyone rides a normal car with gas, brakes, and a steering wheel, though performance characteristics do vary. At all times, you are the only driver on the track. Completing a drag strip track is trivial: just drive straight. Some tracks (like a NASCAR circuit) have banked turns which allow for high speeds with minimal focus on turning or braking. Many tracks, however, require different strategies. As an example, you MUST brake at turns on the Nurburgring because failure to break WILL cause you to leave the track, potentially causing a spinout or even a crash. If a racer aims for a specific lap time, they'll need to maintain a minimum average speed across the lap, which requires learning optimal driving, turning, and braking strategies specific to that course.

I think you can mine this analogy for more insight, but the primary point is: CICO is in all weight loss strategies, but because of personal hormone dynamics, not all weight loss strategies can be CICO.

If CICO isn't practical, what is? If you're someone whose base metabolic rate plummets so much when dieting that you have to basically starve just to lose weight, you're probably screwed no matter what. Why single out CICO as not being practical for these people?

it works until you run against the limit where it's not possible to cut more calories but are still overweight. I think this is mediated by genes. Even willpower is genetic.

Renaissance Periodization just did a video on calorie cycling https://www.youtube.com/watch?v=QwZfryrVddY (the video is titled carb cycling, which he dismissed in the first few minutes and he moves on to calorie cycling) and in it he discusses some of the benefits and disadvantages of calorie cycling vs just eating at a deficit.

The tl;dr is follow your body (e.g. not sleeping well, exhausted all the time, eat more for a bit), but try starting out around 2:1 for eating a fair bit under vs slightly above maintenance. (e.g. diet 6 days, eat a bit extra for 3 days, repeat).

In another video he also mentions that after weight loss, that it is a good idea to slowly increase calories until the scale doesn't move, and then eat at that level for a few weeks to reset the body's hunger signaling. (I am dramatically simplifying there).

> The human body always adapts and if I were a gambling man, I'd say this is largely why most people fail at dieting: after a month or two the results stop and they have to further decrease their intake which interferes with other pathways and tanks motivation.

This is why "cheat days" or even "cheat weeks" work. People are on a strict diet, weight loss stalls, they go on vacation and eat at a surplus for a bit, and then they find that after vacation they are able to resume losing weight.

Likewise a big thing in Keto awhile back (not sure now) was cycling off Keto for the weekend, which unless a person is very careful, ends up with people eating at least a little bit to "excess".

Plenty of people do "lazy keto" and eat a big Sunday family meal and so long as the excess calories on Sunday don't exceed the sum total of deficits from the other 6 days, they'll be A-OK, and now it turns out they may actually be better off.

I highly recommend this article from the MacroFactor team on this: https://macrofactorapp.com/refeeds-diet-breaks/

Basically their view is that refeeding (aka cycling) probably doesn't do much physiologically speaking, but there may be advantages to adherence when people use diet breaks (typically longer than a cycle or refeed).

> As shown in Figure 9, results indicated that “intermittent energy restriction interventions,” which include relatively long diet breaks, led to a fairly negligible weight loss advantage that was not statistically significant.

The problems I am seeing with all these studies is that the participates in the control groups (just regular dieting) are still losing weight throughout the study.

Instead, do a study where you have a bunch of people who are calorie restricted, and when the scale stops moving, have half the participants do a re-feed before trying to cut again, and the other half keep soldiering on as is.

Or to put it another way, we all know people plateau with diets. Running a study that where participants do not hit a plateau, and instead having science just go "oh well! Sorry can't help you!" aren't of much use.

(It is funny because the author of that article starts out talking about plateaus, and then runs a bunch of studies that.... don't encounter plateaus.)

I've met people, who are grossly obese, who after going from 400lb to 300lb, find that they aren't losing weight at even 2500 kcals a day. So they drop it to 2000kcal. 25 pounds later the scale stops, and they try 1500kcal, feel like shit, and they give up. Math says at 250lb they should be shedding tons of weight at 1500kcal, but they aren't.

I want to see a study that investigates that, instead of "oh someone lost 10kg, woohoo!"

10-15kg is a lot of weight to lose, but almost any dietary intervention will pull of a 10-15kg weight loss (which is what all the studies looked at show!)

> Theoretically, reverse dieting should be most effective in scenarios where metabolic adaptation is most extreme, and to my knowledge, the Minnesota Starvation Experiment reports some of the most dramatic metabolic adaptation ever observed in the scientific literature.

Except it was taking people at healthy weights and having them drop to an unhealthy low weight, and then having them regain weight.

The situation we are faced with now is people at unhealthy high weights, with all the hormonal problems that entails, losing a bunch of weight, and messing up their hormones in a different way, and now the goal is to stabilize their body at that new lower weight.

> As previously discussed, the available research hints at a number of subjective and behavioral improvements related to hunger, satiety, diet adherence, mood, and perceived energy level, in addition to improved performance

All of those "subjective" feelings have basis on the body's physiological systems. With better measurements, the underlying cause of those "subjective feelings" can be found.

Edit: FWIW I think the extreme weights we are seeing in developed societies have multiple underlying causes, and that pure dietary interventions are not the solution for everyone. From gut microbiome to environmental pollutants, to the # of ads for fast food on TV, IMHO the problem is multi-causal.

I'd say this is largely why most people fail at dieting: after a month or two the results stop and they have to further decrease their intake which interferes with other pathways and tanks motivation.

Yes this is it. People with shit genes simply do not burn enough calories , so these people need to diet to absurdly low levels of calories to merely not be overweight, way lower than predicted by formulas/tables. Which is not sustainable , hence regain.

This is all so obviously refutable if you look at the fact that this affects the USA so much more than other countries and even there if you go back three generations, peolr with essentially the same genes just at fewer calories and - surprise - were less fat.

It's not the calories, it's the carbs (aka sugar). The modern Westernized diet is carb-heavy, and carbs are the main macronutrient contributing to obesity (see the clear increase in obesity following the creation of the food pyramid).

Many Pacific Island nations also saw massive obesity rates. The reason was carbs, not calories. Soda contributes to obesity because of sugar, not the calories.

This person didn’t say “everyone in America”, they said “people with shit genes”. Are you assuming this is the same set of people?

Amazingly, some people are different. Who knew. We shouldn’t gaslight them into thinking they aren’t different because they aren’t like the aggregate, and we get to feel so morally superior if every individual is like the aggregate.

I found CICO very helpful in losing weight (from 120kg to 90kg).

It was a relatively easy metric to track, and using it alone meant that I could still eat all the things I wanted, though I needed to balance what I wanted to eat with portion size.

CICO also meant I started out slow, making small cuts. So it wasn't bad in the beginning. Over time I had to find ways to adapt so I wouldn't go feeling hungry (mostly increase protein and fiber ratio significantly), but I never needed big changes.

On the flip side CICO also meant that if I messed up one day it wasn't so bad. It just moved the goal post one day or two further away. So it prevented me from feeling like there was no point and giving up, instead getting back into it the next day.

By following CICO I had a very linear weight loss, which didn't feel like a restriction or such. I estimated my CO and aimed for roughly 0.5kg per week and that's almost exactly what I managed.

Not being "prevented" from certain foods due to diet X was a big factor in making this bearable.

CICO is very nice if you put it into the proper context, which is,

1. The scale lies. Liar liar pants on fire it lies. (We could go into this in detail, my wife never seems convinced, but it is to my mind a basic fact of the physics of the system that the scale lies in multiple orthogonal ways.)

2. So you want multiple different probes into the same phenomenon so that you get a better insight into what is going on.

3. So, track calories as a second piece of data to try and check whether the scale is lying to you or not.

4. As part of that, they give you greater insight into the choices that you make. For instance you realize that it is functionally impossible to jog off a Krispy Kreme donut but very easy to replace half a meal with it. Or you discover that you can cut fruit juice for water with no dissatisfaction.

Regrettably this is not peoples’ usual relationship with food and the scale, which is exemplified by the Cheat Day phenomenon. The calories are viewed not as a separate data stream but as a morality judgment, which is then turned off for brief periods because constant judgment is hard on us, and these periods contain a lot of very relevant data such that if you turn the metric off selectively for them, the metric is no longer useful.

> But for a single individual, working to establish your base metabolic rate by counting every calorie and weighing yourself regularly, you can always find a way to be successful with the CICO model. If you establish a baseline that your metabolic rate is (say) 2000 calories a day, then eating 1600 calories a day will cause you to lose roughly a half-pound to a pound per week. It does not mean that every individual’s baseline metabolic rate is the same.

Anecdote: I did this over the past year and a half, weighing myself weekly and tracking all calories, including a period of weight loss and gain to get some good data points, and starting around 6-8 months in the linear regression stabilized: I found I burn on average 2200 calories per day even though I'm in my 30s and mostly sedentary. I've since found it to be pretty accurate.

Knowledge of leptin changes this because what you eat does actually matter. The goal should be intermittent fasting, where you eat little-to-no carbs about five consecutive days/week and then plenty of carbs two consecutive days/week. This approach ensures leptin levels in the obese body remain high enough to prevent onset of starvation mode while minimizing glucose levels to promote catabolic action on body fat [1].

The bottom line is: you're welcome to eat your Big Mac and beers, just only two days out of the week.

Also, stress has an important role to play on weight loss. Stress causes your adrenal glands to release cortisol. Cortisol increases glucose levels through gluconeogenesis by making glucose from non-carbohydrate sources, such as proteins from muscles.

[1] https://en.wikipedia.org/wiki/Catabolism

> Knowledge of leptin changes this because what you eat does actually matter. The goal should be intermittent fasting, where you eat little-to-no carbs about five consecutive days/week and then plenty of carbs two consecutive days/week. This approach ensures leptin levels in the obese body remain high enough to prevent onset of starvation mode while minimizing glucose levels to promote catabolic action on body fat [1].

I hate to post the same video in the same thread twice, but this video dives into carb cycling, and explains that actually it is calorie cycling that makes the difference. https://www.youtube.com/watch?v=QwZfryrVddY

FWIW, and I'm a big keto proponent, the goal should be whatever works. Keto is super effective for some people, and completely useless for others. Research has started to figure out why some people respond to keto better than others, but low carb fails really hard for a certain % of the population, works obscenely well for another % of the population, and works as well as any other dieting system for the majority of the population.

As someone for whom keto works really well (I was on keto for 5 years, training 12hrs+ a week in the gym!), I totally spent time promoting low carb to everyone, until I tried to help some people for whom it just didn't work at all.

It sounds like any other of these diets that tell you when and how to eat but not how much. I’m pretty sure I could gain weight when doing everything you describe, if I eat really terribly during the days I eat “plenty of carbs”.

The key to CICO is that you’re actually measuring not just the calories you eat, but the weight you gain/lose, and are continually calibrating. If my diet jacks up my leptin levels such that I don’t lose weight, that will show up on the scale, and I have to adjust my calories down to compensate. With this system I am guaranteed to lose weight so long as I don’t cheat. This is trivially true: the calibration could technically push you all the way to zero calories per day (which will never happen in practice, obviously.)

Case in point, I lost 50 lbs with this approach, eating McDonald’s nearly every single day. I kept it off for roughly 8 years. (I since became a dad and have let myself go tremendously, gaining ~30lbs back, but I will cross that bridge soon.)

>If my diet jacks up my leptin levels such that I don’t lose weight, that will show up on the scale, and I have to adjust my calories down to compensate.

>With this system I am guaranteed to lose weight so long as I don’t cheat.

Unfortunately, that's just not true. Here's some great quotes from the book "The Obesity Code" (which I've just recently found and find an excellent book):

>The abrupt increase in obesity began exactly with the officially sanctioned move toward a low-fat, high-carbohydrate diet.

>A 30 percent reduction in caloric intake results in a 30 percent decrease in caloric expenditure.

>Total energy expenditure is the sum of basal metabolic rate, thermogenic effect of food, nonexercise activity thermogenesis, excess post-exercise oxygen consumption and exercise.

>Since hormones control both Calories In and Calories Out, obesity is a hormonal, not a caloric, disorder.

>If hormones regulate fat growth, then obesity is a hormonal, not a caloric disorder.

>Sugar will increase the blood glucose level and provoke an insulin response from the pancreas. Olive oil will not...[cause] significant increase in blood glucose or insulin. The two different foods evoke vastly different metabolic and hormonal responses."

All of this is to say that it's not enough to simply adjust calories. Yes, calories matter to a point, and everyone knows that starvation (e.g., 0 calories for an extended period of time) is effective in the long-term, but starvation will also harm your body, and no reasonably-minded person can suggest it as a viable approach to weight loss.

The best available scientific evidence shows that continuing to decrease caloric intake can make weight loss more difficult for an obese person.

whoosh

That was the sound of you zooming right past my argument and pretending I was saying something I wasn’t.

I’m not saying CICO is a good idea. I’m not saying it’s healthy. I’m not saying it’s easy. I’m not saying it will have a high chance of success vis a vis people actually being able to stick with it. I’m merely saying IF you do it right, THEN you will lose weight. This is trivially and tautologically true (to the point of CICO being useless to most people!), but it’s extremely tiring having to explain it over and over that it still works despite all of this.

To put it another way: there are two operative definitions of “works” for a diet:

1. Will it work if you do it exactly as directed?

2. Is a typical dieter likely to have success by attempting this diet?

CICO fails miserably at 2. It passes with flying colors at 1. Its best not to conflate the two definitions in the middle of an argument because you end up talking right past one another.

This is something I've said repeatedly in other places and this is a certain segment of people who can't understand it.

CICO is, especially at the population level, terrible dieting advice. We have decades of experience and oodles of research showing that, as advice it does not work.

However, it is, as you point out, basically tautologically true/correct. For CICO to _not_ be correct would essentially be breaking physics and generating mass/energy from nothing.

Yes, your body can change it's energy expenditure in response to dieting, but this is accounted for in the CO part of CICO. If you consume fewer calories than your body expends (in all the myriad of ways that your body can/does expend calories), then you _must_ lose weight.

The fact that CICO is true does not make it good advice but the fact that it's bad advice doesn't make it false.

I think of it this way: CICO is an axiom we have to assume in order to figure out why a diet isn't working.

E.g. if your input is 1200 calories and you work out, but you aren't losing weight? Maybe your NEAT is way down (or you have a thyroid issue). You are super active but can't lose weight? Maybe you aren't tracking food calories accurately, or you need to eat more satiating foods.

CICO isn't a strategy, it's part of the landscape and your diet is a strategy to navigate that landscape.

But no matter what your particular strategy, it is always going to involve eating less hyper-pallitable foods and getting more activity. That may not be all, but you cannot go from obesity to a healthy weight without those things.

CICO makes me pay attention.

Pay attention to what I eat and how often I exercise.

I rarely actually count, but I do lower calories when I look at a label and say OMG. Like, who knew a shake was 2000 calories? Hmm, fries only 500. I just switch and cut my calories by 1500. I’m so good, I deserve these fries! Yum.

Well if CICO is not a good idea, not really healthy not really applicable to most people, then why bring it up when we are talking about solution to obesity ?

The main problem with CICO is that it is confusing the description of a process with an intervention designed to modify said process. Both things a very different.

> I’m merely saying IF you do it right, THEN you will lose weight... > that it still works despite all of this...

That's the first part, the "works" include the IF... You mixing efficacy and effectiveness is here.

> 1. Will it work if you do it exactly as directed?

Try to apply CICO to water and hydration level : You can modulate your hydration level by driking more or less. BUT, you can go above a certain level or below a certain level independent of how much you drink.

The problem with CICO's is you fail to realize that CI and CO are not the control variables and you might be getting the causality wrong.

Much in the same way that you can change the temperature of a room by changing the value on a thermometer... You can't change the body set point and fat disposition by "just" changing CI/CO.

> Well if CICO is not a good idea, not really healthy not really applicable to most people, then why bring it up when we are talking about solution to obesity ?

You may want to read my first post again. I spend nearly the entire post agreeing about the problems with CICO. But in my last paragraph I explain that I still like it and that it works for me, and explain why. YMMV.

> That's the first part, the "works" include the IF... You mixing efficacy and effectiveness is here.

Right, that’s called propositional logic. It’s perfectly valid to make “CICO works” predicated upon whether you follow it correctly. This isn’t as trivial it sounds: there are diets for which this is not true!. I can totally stick to a low carb diet to the letter and still gain weight, because if you don’t include a calorie limit, you can still go way over your calorie expenditure even with low carbs. As trivial as my “definition 1” is for “works”, it’s still a bar that many diets don’t pass.

Well the point i am trying to make is that your explanation on why it work for you is probably wrong, and even more the advice you generate from your theory can be dangerous and counter productive.

> Right, that’s called propositional logic. It’s perfectly valid to make “CICO works” predicated upon whether you follow it correctly.

Well no. And that really the core of the issue. I see that you didn't answer to the example i gave you.

But if we want to be pedantic :

1 - There is no definition of "works" in propositional logic.

2 - Here what i understand from your logic deductions :

  A - CICO can be rephrase as : For a given individual , for every time duration t : delta_weight(t) = CI(t) - CO(t)


  B - What you are saying is  : for a given individual , for every time duration t : delta_weight(t) = CI(t) - CO(t), THEREFORE, for every delta_weight, there is a time duration T, and CI(t) and CO(t) such that delta_weight = CI(t) and CO(t);

And my point is that B is false (empirically), and does NOT follow from A (logically). And to repeat myself you can try this game with room temperature and thermometer readings.

Your understanding of CICO (or at least the point I am making about it) is wrong in your “B” paragraph.

My argument boils down to the following:

For every individual, at any given time, there is a metabolic rate M for which, if their calorie intake was larger than that, they’d gain weight, and if their calorie intake was smaller, they’d lose weight. Crucially, M changes over time.

But going CICO properly means using the scale to measure M, every day, by diffing your expected value of your weight loss given how much you’re eaten, with the actual value on the scale. You now have an updated value for M to use in calorie counting.

Since M is determined by empirical observations of your weight loss/gain in response to the calories you have consumed, you will by definition lose weight if you consume less than M calories. This is the argument that CICO must work by definition. It’s a logical contradiction for it not to work.

> Your understanding of CICO (or at least the point I am making about it) is wrong in your “B” paragraph.

Just to clarify B is not cico, it's the inference that you seems to make from CICO.

> For every individual, at any given time, there is a metabolic rate M for which, if their calorie intake was larger than that, they’d gain weight, and if their calorie intake was smaller, they’d lose weight. Crucially, M changes over time

This is "not" the definition of CICO, and again does not follow from CICO. CICO is fundamentally a thermodynamic statement about the observed state of a system. You are making predictive statement about a possible future state of the system.

Again, if you want to believe that statement that's fine. The point i am making is just it's does not follow from cico, no have i seen anything empirically that validate that statement.

> Since M is determined by empirical observations of your weight loss/gain in response to the calories you have consumed, you will by definition lose weight if you consume less than M calories.

sure this derived from your "definition" of CICO... but your definition is not thermodynamically derived...

> This is the argument that CICO must work by definition

the premise (the definition of the argument) is flawed.

> It’s a logical contradiction for it not to work.

Not really, CICO not working (as in this case) can be just the result of sound logic based on bad premises.

You focus on carbs specifically, but IF is about all calories being eliminated for periods, and it doesn’t necessarily mean you eat plenty on the break days (if you take break days at all - some people exist on 16:8 or 20:4 or whatever perpetually). I think you have something important here but the wording is ambiguous and I am wondering if you can clarify your thinking?

Are you saying only carbs matter with respect to leptin?

Thanks for asking. I'll first explain the process (because I want to highlight why IF matters), and then I'll answer your question directly. It's going to take a little bit to properly elaborate, but I believe it's worth your time to read.

We first need to choose which of the following macronutrients contribute most to weight gain: protein, fat, or carbohydrates. Proteins (or, rather, amino acids) are not known for contributing to body fat in any significant manner. Fat was once thought to be the main contributor when calories--not carbohydrates--became accepted as fattening. However, modern science shows fat has little significant contribution to body fat and that the real contributor to body fat are carbohydrates. The discovery of leptin about 30 years ago finally shows why.

Carbs are sugars. Sugars, such as glucose, are the main source of fuel in the body, as they are eventually synthesized into ATP. Blood glucose levels are regulated by the hormone insulin. Elevated blood glucose levels increases insulin production, whereas low blood glucose decreases insulin production (with glucagon helping stabilize blood glucose levels). High insulin levels promotes anabolism, whereas low insulin levels promotes catabolism. Creating body fat is an anabolic process, whereas burning body fat is a catabolic process. Putting it all together, eating carbs inhibits the essential catabolic processes for burning fat.

While entirely cutting out carbs should be the answer, it's not. Many obese people report failure to lose weight on ketogenic diets. Why? The hormone leptin. See, leptin levels increase with insulin levels. Leptin signals the availability of energy reserves in the body. Adipose tissue (that is, body fat) is the primary producer of leptin. Importantly, leptin levels vary exponentially with body fat. The obese individual's body is so oversaturated with leptin, the body becomes resistant to leptin.

Decreased leptin levels will cause body fat to be burned. On a ketogenic diet, catabolic processes initially cause a reduction in body fat, which then decreases leptin production. Because the obese body has leptin resistance, the leptin production decrease is interpreted by the brain as a severe drop in available energy reserves, causing total energy expenditure to be immediately reduced and placing the body in "starvation mode." At this point, eliminating body fat actually becomes more difficult, as excess carbs become anabolized into body fat due to decreased energy expenditures.

So, to finally get to your question, how do carbs matter with respect to leptin? To lose weight over months and years, an obese person needs to eat carbs to (1) keep their leptin levels stable over the course of each week to prevent the body from entering starvation mode while (2) maximizing the efficacy of catabolic processes. As the body gradually recovers from its leptin resistance (which will likely take time), the body will shed fat without triggering the energy-minimization mode.

In this paradigm, your total energy expenditure determines how fast fat will be burned. According to the book "The Obesity Code", "total energy expenditure is the sum of basal metabolic rate, thermogenic effect of food, nonexercise activity thermogenesis, excess post-exercise oxygen consumption [(EPOC),] and exercise". Of these five factors, EPOC is mostly likely underutilized. A simple 10-minute HIIT routine every other day will regularly drive EPOC, which causes ATP synthesis. When an person isn't eating carbs, the body dips into fat stores to acquire the necessary glucose for ATP synthesis. Note that with this approach, exercise isn't strictly necessary but becomes extremely beneficial to maximize the efficacy of catabolic processes.

Intermittent fasting should work as long as the specific protocol doesn't decrease leptin levels such that the obese body enters starvation mode. In my opinion, the most workable approach for people is to try their best to completely eliminate carbs Monday-Friday, eat all the carbs they want on the weekends, and do the bare minimum of a 10-minutes HIIT routine every other day.

> Intermittent fasting should work as long as the specific protocol doesn't decrease leptin levels such that the obese body enters starvation mode. In my opinion, the most workable approach for people is to try their best to completely eliminate carbs Monday-Friday, eat all the carbs they want on the weekends, and do the bare minimum of a 10-minutes HIIT routine every other day.

This is just calorie cycling with extra steps.

It doesn't honestly matter if it is 5 days on, 2 days off, or 2 months on, 2 weeks off. It is up to a person's individual biology. When the scale stops moving, up calorie intake to the extent that the scale doesn't move up more than a smidge, wait a bit, then drop calories again.

TKD, CKD, intermittent fasting, dirty keto, keto with a once a month weekend pizza binge, whatever, almost any combo of "eat less for awhile, then eat normal for little bit, then eat less again".

>This is just calorie cycling with extra steps.

That's almost right, but it's carb cycling, not calorie cycling. The difference does matter.

>TKD, CKD, intermittent fasting, dirty keto, keto with a once a month weekend pizza binge, whatever, almost any combo of "eat less for awhile, then eat normal for little bit, then eat less again".

I believe that the best approaches, for a given period, will be 2/3 of the time on minimal-to-no carbs, then 1/3 on carbs (which, for seven days, rounds to 5-off / 2-on). The leptin information also suggests that an obese person shouldn't be off carbs for long periods of time, and one week is a nice time frame for people (go low-carb during the work week, have carbs on the weekend).

CICO simplifies dieting so much that 2000 calories from drinking cola are equal to 2000 calories from eating a stew of chickpeas, beans and nuts. The human body is not a furnace literally burning fuel, it's a complex chemical reactor in a tight feedback loop with what it actually processes.

I mean, it’s not that far off. Take this guy who ate only twinkies/nutty bars/donuts/doritos for a month (with some veggies added in for fiber/ruffage):

https://www.cnn.com/2010/HEALTH/11/08/twinkie.diet.professor...

> His body mass index went from 28.8, considered overweight, to 24.9, which is normal. He now weighs 174 pounds.

> But you might expect other indicators of health would have suffered. Not so.

> Haub's "bad" cholesterol, or LDL, dropped 20 percent and his "good" cholesterol, or HDL, increased by 20 percent. He reduced the level of triglycerides, which are a form of fat, by 39 percent.

I mean, yeah, of course if you eat less than your body needs, you will lose weight, nobody is denying that. You can lose weight drinking only coca cola every day. Now, how _easy_ that would be, how much hunger one would feel, the effect on health, mental health, energy, what the rebound is after the diet is done etc., that's a completely different discussion. And another discussion is how much the weight loss is sustainable/maintainable in the long term.

So yeah, in a vacuum, with a perfectly spherical cow, it's not that far off.

Genes play a big role. Some people have a faster (I think the more appropriate word is inefficient) metabolism, and can lose weight or stay thin despite eating more calories compared to someone with a slower/efficient metabolism, controlling for body size, age, and other factors. There was a thread on Reddit a while back where formerly or current obese people shared their old diets, and many people were commenting that it did not seem like much food. Even though people are generally bad at self-reporting calorie intake, that's called having a crap metabolism.

Imagine two individuals who both both need 2,500 calories/day to feel satiated, but the second individual stores anything above 1,500/day as fat, and a second individual stores anything above 3,000 as fat. The first person will constantly be fighting his body to not overeat and will feel hungry constantly. The second individual can eat to fullness and not gain weight.

Some men , particularly the formerly obese, have such slow metabolisms that anything stored above 1,200-1,700/day is stored as fat despite weighing 170 pounds or so, from reading personal accounts here and on Reddit. This is also why bariatric surgery patients regain weight and become obese again despite only eating around 1,200-1,500 calories/day. These people basically got short-changed when it comes to metabolism genes. If there is a famine, they will outlive everyone. But in abundance they become fat very easily. This does not answer why there are more obese people today compared to in the past, but there is still probably a genetic component nonetheless.

And on the other extreme are people who can be very slim despite eating a normal amount of calories or only slightly less than 2,000. These people are much rarer, as it's disadvantageous from an evolutionary perspective to have an inefficient metabolism, but they exist in the same way very tall or short people exist by virtue of phenotypic variance.

I am skeptical about reports of people who don't seem to gain much weight despite reading a lot of food. Most of those reports are merely anecdotes, or at best observational studies, and we know that most people are terrible at accurately tracking their caloric intake. More rigorous studies that involved putting subjects into locked environments and carefully measuring their intake have consistently found that the CICO model is accurate.

However, as a practical matter for people who want to lose adipose tissue there are likely more effective approaches than CICO.

Anecdote: throughout my life, on many occasions, I've gone on multi-month binges of over-consuming massive quantities (1500+ cals) of desserts on a daily basis, while retaining a sedentary lifestyle. I always seem to weigh the same weight (skinny).

I realized I've been doing this over the past month. I weighed myself yesterday. Same weight.

Clearly something is different about my body, at least when it comes to sugar calories. It just doesn't seem to readily convert it to fat.

Is it really that hard to believe that some people's bodies actually behave a little more rationally than most?

You need more than vibes to know how many calories you're eating. Unless you have Cronometer / calorie tracker receipts, I would dismiss your calorie estimations and so should you.

It really is that hard to believe. Have you ever had a resting metabolic rate test? Have you ever gone several weeks carefully tracking everything you consume using an accurate food scale? I am skeptical that even during binge periods you were really taking in as many calories as you claim.

That said, there are limits to what the human digestive system can absorb from a single meal. Binge eaters do sometimes exceed that limit and just poop out the excess without storing it as fat.

Lol. We exist.

I got the taste for raspberry ripple ice cream in my mid-20s, and ate 10 litres a week - that's 2000+ kcal a day just in dessert.

That's not binging. I put that on top of 500g dry weight of pasta with cheese and sauce, and kept it up for months. Plus breakfast, lunch, and several beers. And packets of chocolate biscuits. I was covering 10km a week or less, no marathons. No tapeworms, either.

You may be right that I was shitting a lot of it out; the horrible shits are what turned me off the ice cream habit. Nonetheless, I ate well over 4000 kcal every day, for years, and never put on weight. 6000 was typical. And most of that was potatoes and wheat, which are very bio-available carbs. I've always been a black hole for food.

I suspect I radiate it as heat. Exes have told me I'm hot - in a caloric sense. I feel the cold more, now that I eat less.

To the overweight, if it helps at all, know that I have other struggles.

There is one individual Michael Rae who is able to do just this. He stays at 115 pounds despite eating 1,900 calories/day with a BMI of 15.6, which is insane. Most men would probably need to eat 1000-1500/day to get to that lean, as the infamous Minnesota starvation study showed, and even then many of those participants did not get that thin.

Here is a profile of him, among other individuals who are part of his dieting cult/group https://nymag.com/news/features/23169/

Michael’s regimen of 1,913 calories a day is exactly that: 1,913 calories every single day, 30 percent of them derived from fat, 30 percent from protein, and 40 percent from carbohydrates. Cooking for him is the same elaborate exercise in dietary Sudoku it is for all CR die-hards, only more so.

There is no evolutionary justification to have such an wasteful metabolism (for all of history until now it's a a disadvantage), so people like him are rare, but they exist.

Another is ‎Roy Walford who got to a similar weight during Biosphere2 on a similar 1900/day diet when the artificial habitat began running out of food.

There's no evidence that Michael Rae has an unusual or wasteful metabolism. 1900 kcal/day really isn't very much. With his size and input level (CI), a moderate activity level (CO) would keep him in energy balance.

Note that there is no reliable evidence that caloric restriction to the point of having an underweight BMI will actually extend human lifespans. The experiments have all been done on smaller lab animals living in safe little cages where they are protected from injuries and infectious diseases. Under nourished humans are at severe risk of sarcopenia as they age.

https://peterattiamd.com/outlive/

The Minnesota starvation study showed that the formulas fail when extreme weight loss occurs. So if someone starts with a BMI of 22-24 or so and then tries to get to 15.6, will burn way fewer calories than predicted by the calculators/formulas. 1900 is not much but a BMI of 15.6 is pretty low. He did not start that low, so I wonder how much he was eating before he embarked on this experiment. As for extending lifespan, agree, that the data is iffy on that.

That's not insane at all. At ~183cm, his BMR at 135lb would be 1600 cal/day using a standard formula, so he would only need to expend 300 calories through activity in order to breakeven, which is something many people do with just their daily routine walking around. There's nothing even slightly unusual here.

The key is that very small changes in calorie intake, over a long time, have a large impact on equilibrium weight, and that height factors a lot more into BMR than people think.

A BMI of 16 is very rare. It's unlikely it was always that low for him. Assuming someone starts with a BMI of 22-24 and then diets to get it to 16, this would likely would req. fewer than 1900 even if the calculator says so, due to metabolic adaption and the breakdown of the calculations at extremes .

Of course it would take less than 1900 to get there. But it would take around 1900 to stay there, according to our models of basal metabolic rates. And it turns out that this person stays there at 1900, so our conclusion is that the BMR model we're using fits this data point.

It's possible that for some people at such extremes commonly available models would break down, but clearly for this person, they haven't. And while BMIs of 16 are relatively rare, so are 6 foot tall males that eat an average of 1900 calories for years.

Ultimately if you're trying to use this case to show that the models are inaccurate, that's not what this case does, it's actually remarkably close to what is predicted and if anything would be better used as evidence in their favour.

no, Rae is truly exceptional. Other calorie-restrictors eat far less and weigh more ,even controlling for height

https://www.webmd.com/diet/features/restrict-calories-revive...

"I eat the same thing twice a day, seven days a week, 365 days a year," says Pomerleau, of his diet of 1,200 calories a day. "It's a lot of veggies with olive oil-based salad dressing, a lot of mixed fruits, almonds and hazelnuts, and flaxseed oil for omega-3 fatty acids."

https://archive.is/e345S#selection-4235.216-4239.443

.Bob Cavanaugh, a 54-year-old landscaper from Morehead, N.C., has trimmed his intake to two meals a day, totaling 1,500 calories. Breakfast consists of one cup of quick oats, two tablespoons of toasted wheat germ, one cup of skim milk and blueberries. For dinner, he eats vegetables, fruit and a small portion of fish. "I'm hoping to see my great, great grandchildren," he says. Mr. Cavanaugh's diet may sound extreme, particularly since at 5-feet 9-inches tall and 158 pounds, he isn't overweight. But over the next several years, if the monkey results hold up, they could represent a major shift in how we view food and nutrition. The U.S. Department of Agriculture says the average sedentary woman should eat 1,600 calories a day and the average man 2,200 -- benchmarks already significantly overshot by most Americans.

and a third guy https://pbs.twimg.com/media/EW8oScgXsAAVqeg.jpg

A bmi of 21, nowhere close to 16 despite eating fewer calories (1700)

Bob Cavanaugh's case is not relevant, as he has relatively recently made these changes, and therefore isn't at equilibirium. Same goes for your third guy, the BMI is from 2004/2009, while his caloric intake is from 2019 and was after a 5 year taper (and was therefore higher in 2009).

Your first example. Pomerlau, weighed only 121 pounds - that's a BMI of 18 - and lost 5kg/year, which means that if he did eat 1200 calories, he almost certainly wasn't at equilibrium yet, either - it would take a few more years.

None of those examples really are outliers.

They have been some studies where they showed that NEAT (Non-exercise activity thermogenesis) varies widely acros individual and even with the same individual at different time. I think variation as wide as 1500 have been observed. Enough to explain the people who eat a lot but don't gain weight .

That's me, and I have yet to discover any upper limit - I eat less now because I'm bored of chewing.

CICO works quite well at the extremes. It's effectiveness varies, but it is directionally correct.

If someone is sitting at a 35+ BMI, then their caloric intake is so out of whack, that CICO helps ground them. Similarly, an anorexic who starves gets a clear directive for increasing weight by increasing calories.

CICO also promotes calorie general calorie comparison. It helps you understand that a milkshake has the same weight impact as an average Mediterranean lunch bowl.

Now the absolute impact on both foods will vary from person to person. But comparatively, it helps ground the unhealthiness of foods like sugar rich shakes and icecream.

The key insight from leptin is that the carbohydrates in the milkshake actually causes weight gain, not the calories. Yes, you do need to know how many calories your specific body needs to properly operate. For example, taller people require more calories than shorter people. However, obesity is a hormonal issue stemming from leptin resistance, not a caloric intake issue.

> However, low leptin levels causes the body to enter starvation mode, and hardcore dieting causes leptin levels to drop further.

What about fasting? Dr Jason Fung wrote a book about fasting that IIRC says that it does not cause starvation mode.

Just looked up his book "The Obesity Code"; not sure if that's the one you're thinking of.

Dr. Fung says obesity is a basically a hormonal issue, not a caloric issue. He discusses a "causality test": "If we inject this hormone into people, they must gain weight". According to Dr. Fung, leptin and ghrelin don't pass this test, but insulin and cortisol do. Still, leptin and insulin appear to have a close relationship. "Insulin promotes fat storage. Leptin reduces fat storage. High levels of insulin should naturally act as an antagonist to leptin. However, the precise mechanisms by which insulin inhibits leptin are yet unknown." He also says that "too much insulin causes obesity" and that "obesity is a state of leptin resistance." (Sorry for that mini book review; I needed to understand what he was saying.)

On the fasting point, he's explicitly supportive not of regular fasting but intermittent fasting: "Diets must be intermittent, not steady...we cannot fast all the time. It won't work."

Based on my fast read of his work and my own understanding, intermittent fasting keeps leptin levels high enough that the body doesn't kick into calorie restricting mode. One approach which seems to accomplish this is eating close-to-zero carbs five days/week (say, during the week) and then eating lots of carbs two days/week (say, on the weekend). The burst of carbs will cause leptin levels to spike since the body now has tons of available energy. Over the week, the leptin levels will remain elevated, but since you're minimizing carbs, the body will draw energy from adipose tissue.

Then why are so many people who are insulin resistant overweight? Yet people with type 1 diabetes tend to be thin. I don't think the connection between insulin and weight is well known, assuming one even exists. It's a basic CICO problem. Whether there is insulin resistance or sensitivity does not change this much. IF his theories actually worked, they would work on general population at fixing obesity, but there is scant evidence it does. Like Tabues and others, he built a career on some theory that falls short when applied to real people in controlled setting.

According to this article [1], many people with type 1 diabetes are also overweight. In fact, the article cites a research study showing that, for those with type 1 diabetes, higher blood glucose levels corresponded to higher weight gain. Clearly, this weight gain isn't CICO-related but is actually related carbohydrates.

As far as I understand, people with type 2 diabetes (insulin resistance) have excess glucose in their bloodstream, which becomes easily stored as body fat. Type 2 diabetes seems particularly caused by eating excessive amounts of carbs over a long period of time.

In both cases, excessive carbs leads to obesity. The type of calories matters more for obese people trying to lose weight than for healthy people maintaining their weight.

[1] https://www.dbl-diabetes.com/all-about-diabetes/food/type-1-...

>Yet people with type 1 diabetes tend to be thin.

People with type 1 diabetes need to be conscious of what they are eating in a way the rest of us don't. I'm not sure you can draw much inference from them.

Agree, straying from thermodynamics is wrong. Human fat holds around 8000Kcal per Kg. Stop eating altogether. Every the 8000th Kcal more or less you would have lost 1Kg of fat.

There’s the issue of not being able to decide what gets burnt. I’ve read that under certain circumstances, the body can actually burn the muscles faster than it burns the fat. Which obviously makes exercise harder.

When losing weight, about 75% fat and 25% muscle is lost, with a ton of variables being in there (exercise routine, how much muscle you have, what your diet looks like, etc).

Lots of arguments about what diets maintain the most muscle (e.g. fasting people say they keep the most muscle, and as much as I love fasting, plenty of studies say that is probably not true.)

Why would the body touch anything else when it has a preferential energy storage?

because the body is not a perfect machine. If it were, all this discussion would be moot

It's not always going to be fat.

The dramatic weight loss induced by GLP-1's demonstrates that.

We're seeing a disproportionate amount of lean muscle mass loss.

The problem is that it is overly simplistic. It is similar to saying "running a marathon is just a thermodynamic problem, expend 2500 calories in a single direction and you can do a marathon." Sure, that's true, but that's not the hard part. Existing at a caloric delta long enough to meet your goal is the problem!

My reply to him addresses this. Intermittent fasting works better than starvation.

Reply on news.ycombinator.com