Conclusion: "Time-restricted eating, in the absence of other interventions, is not more effective in weight loss than eating throughout the day."
It is a pet peeve when a failure to reject the null hypothesis is reported (in confident language) as a confirmation of the null hypothesis. However, I do appreciate how clear the results section is in breaking down exactly what they did measure: both groups lost weight, and the time-restricted group lost more but not significantly more in the overall study.
I agree with you, especially considering news outlets will take this conclusion and report it in the absence of contextual data, and that it makes an incorrect claim.
As someone who's had Extensive experience with fasting, both intermittent and extended, I would hardly even consider this study 'Time Restricted Eating' much less fasting.
Indeed. This is just eating lunch at noon and dinner any time before 8pm with unlimited snacks allowed in between. That's not much of an intermittent fasting regimen unless you're eating all the time.
I don't want to move the goalposts here because I think this is worth studying and I'm happy to see any studies being done at all.
From my research, and I'm happy to dig up my sources:
The longer you go between meals the more pronounced the effect. 16:8 with ad libitum eating is what a lot of people kinda do naturally if they skip breakfast.
The longer you abstain the more pronounced metabolic changes you see. Human growth hormone levels triple over fasted period of, for instance, 5 days. Glucose levels fall off a cliff, and so do free fatty acids. [1] It's unclear that fasting for less than a day would yield similar effects.
I think the idea that if you don't eat, you lose weight, is pretty well documented so it feels like all we can really take away from this study is 16:8 with all the food you can eat doesn't qualify.
"don't eat, lose weight" is obvious of course. "eat the same amount of calories but with different timing, lose weight" is more interesting. Does having higher growth hormone levels result in burning more calories with the same amount of physical activity?
For many types of diet that focus not directly on calories but on the selection, timing or form of eating, the intent that you do not eat the same amount of calories, but rather that eating a "natural amount" according to your appetite within the "new habits" will result in you eating less calories without explicitly measuring and limiting these calories.
Based on what I've read so far, and again, I'm not an expert, just sort of an arm-chair enthusiast.
1. One major component to weight gain is insulin. Insulin triggers the uptake of blood glucose by your adipocytes, and prevents its release. This means when your insulin is high, your body stores free sugar as fat and prevents your body from releasing it for consumption.
2. Insulin follows a positive feedback cycle and can exhibit a runaway condition, broadly speaking, called insulin resistance. You eat more sugar, your body produces more insulin, your cells get used to the insulin, and your body has to produce more insulin to lower your blood sugar.
3. Eventually your body can't make more insulin. It's tapped out. So your doctor prescribes you, usually, a front-line treatment like Metformin -- itself fascinating in its own right as a life extension drug. Once that's not sufficient to control your blood sugar, you have to start start injecting insulin. This is broadly speaking Type II diabetes and, eventually, metabolic syndrome.
4. The idea with intermittent fasting, or adjusting the timing of your meals, is that your body does in fact lower its insulin level once you stop eating. That much we know. Over time, longer periods in this low-insulin state this will reduce your body's concept of what a "low" insulin level is, walking back that runaway condition.
5. In time, this might reverse Type II diabetes [1]. The linked writeup is a small study, n=3, however: "All three men were able to stop injecting themselves with insulin within a month of starting their fasting schedule. In one case this took only five days."
6. Once insulin levels are normal, you're back to normal, too.
HGH reduces body fat and increases lean muscle mass. [2] Further, when fasting, your metabolism is actually elevated, in part due to high levels of noradrenalin. Makes sense, if you're an ape in the jungle and you haven't eaten in a day or so, your body is primed to go out and get more food -- not to waste away.
So, lots of moving pieces, and we're not 100% sure yet. There appears to be smoke here, IMO, which is why I mentioned I'm just excited to see any studies at all.
Insulin is not the only hormone responsible for storing fat,
Acylation Stimulating Protein (ASP) is produced by adipocytes and also stimulates fat storage.
There's nothing really special about IF for fat loss, muscle gain, or for reducing insulin resistance. Improved insulin sensitivity is a natural result of fat loss.
There was a 10% weight loss, and their HbA1 levels (average blood sugar) returned to pre-diabetic levels (they could essentially stop insulin injections).
The increase in HGH levels that happens during a fast is not a net anabolic event, but a compensatory anti-starvation response that shifts fuel substrate use from glucose and amino acids toward lipolysis and ketone production. Part of the cause of GH release is reduced hepatic IGF-1 production, resulting in less negative feedback to the pituitary gland to produce GH. So during a fast you end up with more GH, but less IGF-1. IGF-1 is more directly active[0] in skeletal muscle in producing growth than HGH. Basically, the result is that while fasting-induced GH secretion does help counteract muscle protein breakdown, it does little to increase muscle protein synthesis when you're fasted and you remain in negative whole-body protein-balance. There is nothing anabolic about not eating, as your body has no fuel for new protein synthesis.
It's also only a matter of time before you start losing muscle. Muscle protein breakdown rates starts in between 15-30 hrs of fasting[1]. Fasting for 20 hrs decreases resting energy expenditure[2] and mTOR signaling, an important component of anabolic gene expression. This suggests that 20+ hour fasts may negatively affect you muscle mass and fat loss.
GH also drives insulin resistance[3], so more growth hormone isn't always desirable.
All in all, given some limited studies we have on IF, IF per se does not provide body recomposition advantages and does not alter energy expenditure, and fasts over 20 hours are even harmful. Any health benefits of IF are highly inconsistent and mostly limited to glucose and insulin dynamics. Being lean is much more important for our health than whether you do IF or not. Being muscular probably is too, because muscle mass improves insulin sensitivity. Combined with being lean, any benefits for insulin sensitivity of IF are arguably trivial for muscular, exercising individuals. I could give you a bunch more studies here, but I think this is enough for now :).
Having said all that, you can still use calorie cycling strategies to reduce caloric intake on "rest days" (or more precisely, outside of your anabolic windows) by using a PSMF (protein-sparing modified fast), and upping your caloric intake during anabolic windows (which could last up to 72 hours after training!).
re Insulin: I did say insulin was a major component, I agree of course it's not the only component.
> There's nothing really special about IF for fat loss, muscle gain, or for reducing insulin resistance. Improved insulin sensitivity is a natural result of fat loss.
I think it's too early to say that, which is why we need more studies. To your own point, there aren't enough. In no small part because there are so many different protocols and studies both mix-and-match protocols and also often use questionable ones. Sometimes "fasting" means eating a few hundred calories or a fasting-mimicking diet. Sometimes it means 16:8. Sometimes it means alternate-day. Sometimes it means 5 days on and 2 days off.
> It's also only a matter of time before you start losing muscle. Muscle protein breakdown rates starts in between 15-30 hrs of fasting[1].
For what it's worth, proteolysis does not imply the loss of muscle mass after re-feeding in short-term fasting. The working thesis is that while fasting your body attempts to spare as much muscle as possible (in part through HGH) and that protein breakdown slows dramatically [4].
After all, HGH increases fat oxidation and increases protein retention, so it would all make sense.
I didn't, and don't think that, fasting will increase muscle mass, but I have seen no evidence that it will, as you suggest, break down muscle in any meaningful way until far, far later in the starvation cycle. Muscle in particular is a fuel of last resort. Wouldn't make much sense otherwise. There will always be some turnover, of course.
> Resting Metabolic Rate
Studies show that during short-term fasting your metabolic rate stays pretty consistent, and increases over the first couple of days. A couple of studies pins it at up to +10% after 36-48 hours. [1,2] It doesn't actually drop until 60-96 hours into a fast. [3]
tl;dr: I guess my point is, it's too early to be dismissive. More data is needed.
> This suggests that 20+ hour fasts may negatively affect you muscle mass and fat loss.
I will say this particular line is kind of nuts, if you think about it. How can not eating anything at all slow down your fat loss when you have a caloric deficit of 2500kcal/day? That's why you have fat. That's why your body enters ketosis over the first 3 days. Why would your body burn your muscles (i.e. means of obtaining more food) when it's been storing stockpiles of energy in the form of fat? This right here, is a common myth.
There's only two tissues in your body that require glucose. Red blood cells and the brain. After 3 days 70% of the energy required by your brain is provided by ketones. The remaining glucose requirements are satisfied via gluconeogenesis from glycerol backbones of triglycerides, and yes, to a small extent, via protein breakdown. The entire rest of your energy requirements are satisfied via lipolysis, to the tune of 2/3 of a pound of fat per day. Or the equivalent of running 17 miles.
For a different point of view: I can't imagine being able to sustain restriction to 8 hours for more than a day. I've struggled with gaining weight my entire life, which is certainly not the norm, so I'm not really a central example of anything.
The point about growth hormone is interesting. Im not sure if I'd be able to eat enough Calories in that time to survive, but if I were able to fit them all in to 8 hours, do you think someone who is under weight but doing consistent weight training would actually gain weight?
I definitely plan to do something like a 16:8 ad libitum (or actually, if I have a minimum to hit, I guess it isn't ad libitum?) eventually, but I'm still working on just getting used to eating at a small surplus in general. Not the typical problem, haha.
Correct, it doesn't add to your muscle mass when fasting, of course, that wouldn't make sense. What it does is increase fat oxidation and prevent the breakdown of your existing muscle. The net of it is a better final body composition; specifically a better fat:muscle ratio.
There are physiological differences between caloric restriction and fasting approaches. Fasting naturally ends up in ketosis where you body has switched to using lipids as your primary fuel and hgh supports that process. This isn’t the case with normal caloric restriction diets generally speaking.
Continually eating keeps insulin levels up, suppresses hgh production and unless you’re very intentional about it, keeps you out of ketosis making your primary fuel glucose instead of fat. Your body also produces substantial amounts of noradrenalin while fasting.
Further compliance is another area where you may see differences. A lot of people find it much easier to eat nothing at all than to count and document each calorie. Certainly if ketosis if your goal where getting to that point takes few days and accidentally eating a few grams of sugar physically switches your brains primary fuel source from ketones to glucose.
Lastly the other big difference between continuous caloric restriction and fasting is that your metabolism will slow down when you’re on a CR diet to match your new food intake level over time. I can find some studies. However that is not true of fasting approaches.
The reality is weight loss via caloric restriction is incredibly hard. WeightWatchers for instance published a study that showed people on their program were less likely to lose weight than the general public. Caloric restriction diets generally simply do not work. You can always blame compliance or the participants but at the end of the day the goal of a weight loss program is to have compliance.
The reality is your body is in a very different mode when you eat nothing as compared to when you’re eating a little less.
It does not surprise me that weight watchers clients are less likely to lose weight - they are already looking for a miracle diet and miracle foods that don't have points to make them skinny.
Instead buy a kitchen scale, keep track of how much you eat in a journal for a few weeks to find your maintenance caloric intake, and drop it by 10% and add some physical activity like walking. If you still aren't losing weight you're probably eating calories you aren't tracking.
The problem with ketosis for people who struggle with diet control is that its easy to come out of,and then you're eating a high fat diet and not on ketosis.
Id rather a balanced diet that allows me to go out and eat with friends occasionally without worrying about ketosis.
Once you track and measure your normal food for a few weeks you get a better understanding of what 100grqms of pasta or 4 ounces of beef is, and it becomes easier to understand how much food you should be eating.
If the goal of a weight loss diet is compliance then weight loss should be fairly slow 1% a week without extreme caloric or food type restriction.
> If the goal of a weight loss diet is compliance then weight loss should be fairly slow 1% a week without extreme caloric or food type restriction.
Again there is no evidence caloric restriction diets actually lead to sustained weight loss, and a lot of evidence that they simply do not. Kind of mind blowing, right? In fact this meta-study showed people would actually be better off not going on a diet in the first place. [1]
Remember, your metabolism will slow down when you start restricting your caloric intake by a small quantity (while it increases when you do it by a large quantity) and you will start compensating for the food you don't eat with reduced desire for physical activity and cravings for food. You also start compensating for increased physical activity the same way. This is a natural evolutionary response: your body wants you to stockpile fat for a rainy day. The thesis for fasting is it tells your body it is that rainy day and it's time to dig into the larder.
If it was that easy, there simply wouldn't be fat people.
Caloric restriction diets don't do anything, and it's time we stopped pushing that narrative.
> Id rather a balanced diet that allows me to go out and eat with friends occasionally without worrying about ketosis.
And for what it's worth with intermittent fasting you're welcome to go to town when you're not fasting. That's one of the selling points.
Another thing people don't really understand about fasting is they assume you're going to be hungry. This isn't the case. After a few hours your hunger goes away, and it doesn't really come back. Not for a long time. Like, days. It's hardly a challenge to not eat. Certainly not with some practice.
I have tried keto, intermittent fasting (18/6), and Omad. Both were fine for short-term losses.
The thing that stuck was:
- finding my maintenance calories by journaling and meal prep (weighing food in advance). I found my maintenance calories to be about 2800 per day.
- reducing my caloric intake by about 250 calories net.
- replacing some of the fats and carbs in my diet with more protein (actually above the 1gr/lb lean body mass suggested but I find proteins more filling and delicious).
- walking for about 3-4 hours a week (super important). If I can't walk due to weather I use a stationary bike. I do relatively low intensity cardio this way that isn't hard on joints but I do it for an hour at a time.
- resistance training with the periodization strategy seriously 3-5 times a week (4-5 hours a week total), focusing on progressive overload and training everything.
- making it a permanent lifestyle decision. Not "I'm doing this for 3 months to lose weight" but finding an exercise routine, permanent diet, and foods that aren't calorie dense so that I can feel full and get the right amount of calories.
-adjusting my food intake by tracking a weekly rolling weight average and dropping calories if required (something that hasn't really happened since I picked up so much exercise)
I don't know what the studies say but I can tell you this is the only way that has been successful for me. Ive lost about 50 lbs, 6 pants sizes in a year and gained a lot of muscle as well.
I don't think you can permanently lose weight without committing to a new lifestyle. My old lifestyle made me fat. Temporarily dropping weight and returning to it will just yoyo my health.
Instead you have to find a new permanent lifestyle that is on the trajectory you need. That why weight watchers will almost never be a real weight loss program - everyone is there to look for zero point foods and eat cake. Perhaps by accident they will arrive there but not by following weight watchers
Anyone could implement my strategy on weight watchers or intermittent fasting but I found I could not sustain the exercise if I was fasted and I felt terrible. My way seems to be a good solution for me. I still eat bread, I occasionally have ice cream (although within my caloric deficit), i eat delicious foods that I enjoy eating but that aren't dense with calories and I don't see any reason to stop - this is just how I am from now on. Eventually it all becomes habit and the discipline required drops. I still hate cardio but doing sustained steady state makes it bearable and weight training is now something I look forward to.
I don't disagree with anything you said, of course, it's an anecdote. However, I think it's important to follow the studies, science and data and to look in aggregate.
I don't disagree with knowing what the science says, although often that seems to change from year to year. It's also important to know that I am not an aggregate, and like you said my anecdote is about what works for me, but what works for someone else is different, and what trends good in aggregate might not work for someone else for a variety of reasons (Compliance, etc).
Thats why the best diet and workout plan is the one you do consistently and commit to making your permanent diet/workout plan. Knowing the optimal science way is good, having a lifestyle you can live that achieves your goals is best.
This describes me most days and I'm certainly not intentionally fasting. I just don't really like to eat breakfast, so my first meal is usually lunch sometime around 12, and dinner is usually before 8. I might have a coffee in the morning though (although I suspect coffee on an empty stomach in the morning is making my anxiety worse, so perhaps making myself eat breakfast would be a good idea)
There doesn't seem to be any information regarding adherence to the protocol. Did the participants manage to keep to either strictly Intermittent Fasting or a stern 3-meals-per-day-and-nothing-else program? Considering that the participants were all "Overweight" or "Obese", implying that the participants have historical issues with portion/calorie-intake control, that seems like a pretty important point to ignore.
Certainly hard to measure, but Figure 2A certainly looks like an adherence chart.
Regardless, in RCTs we usually consider a 'intention to treat' analysis gold-standard. That means that since you are assigning an intervention in the real world, even if someone drops out, you still consider them part of the intervention group, because the same thing will happen in the real world. You can also do a 'per protocol' analysis, which attempts to measure the impact of the intervention to those that adhered, but not as important...
That's a fair point: if study subjects can't/won't adhere to the protocol why would we expect real world participants to do so? Either the study subjects adhered to the program and the results showed (little to) no improvement OR real people won't adhere to the program anyway and results will show (little to) no improvement.
I think the fact that the time-restricted eating group did slightly better than the consistent meal timing group is actually pretty strong evidence that intermittent fasting is effective. At least for me, it is much easier to only eat during a certain time period than to hold to a 3 meal a day schedule, so the fact that the groups performed similarly shows a lot of the value of intermittent fasting - same results with simpler rules.
Also I'm surprised that a 1% drop in body weight across ~100 participants gives statistically significant results - isn't that pretty well within daily fluctuation due to water intake? I would imagine most people who are trying to lose weight are aiming to lose more like 10% than 1%.
This is a logical fallacy. They didn’t test against “eat whenever you want” as a baseline, so you can’t draw any conclusions about effectiveness. Also, as you point out, there was no significant weight loss in either cohort, meaning NEITHER is effective for weight loss, though that may be against a baseline of weight gain (we don’t know without the zero-intervention control).
Sorry, I was using "significant" in the non-scientific sense of the word. They proved with statistical significance that people lost weight, but the amount they lost (less than 1 kilo) is not medically or personally significant.
My jaw dropped when I read `ad libitum`. The single biggest impact that restricting the number of meals had for me was to decrease my appetite and increase my resilience to a baseline level of hunger. Telling chronically overweight/obese people to eat as much as they want during this kind of a trial subverts a golden opportunity to accustom the body to actually eating less food.
this is something in clinical research called the dichotomy between pragmatic and explanatory trials. RCTs like this aren't useless even though they don't specifically answer the question YOU expect to have answered, which is essentially: "controlling caloric intake, what impact does restricted meal windows have on weight gain/loss/other measured outcomes?" This is an explanatory question--it seeks to isolate a variable (meal timing) and explain it's impact on people.
in contrast, the question the RCT is REALLY asking is essentially: "comparing ad libitum intermittent fasting (which is a very easy thing to suggest in a busy clinic) to a very expectably superior intervention like calorie tracking and restriction, does IF end up having some efficacy?"
it's a different question--more realistic of an intervention to provide and monitor (because you don't monitor it at all really...)--and so ends up being perfectly valuable if you interpret it for what it is, assuming the rest of your critical appraisal of the paper passes muster.
Anecdote: I tried 16:8 intermittent fasting for a month (it wasn’t hard considering the lockdown situation). I gained 4lbs (156lbs to 160lbs).
No claims are being made of mine being a proper experiment. Just sharing the observation. I generally don’t focus on weight. I did not do it to lose weight. My best guess is that my overall calorie intake went up as I fit all meals in that 8hr window.
I did this for years, and, it was the easiest way to cut calories and get quite lean, but really not great for life in general. Now I have a revolutionary new diet: “Eat when you are hungry. Don’t eat when you are not hungry.” Also choose things that digest slowly. I am not nearly as lean as I was on IF, but I also think I was losing muscle, which is a lot harder to gain back than the fat was to lose. Also I don’t have anybody to impress anymore, so, don’t really care. Side note, using IF as a social excuse against eating is a lot easier than “I don’t feel like it”.
These people were significantly overweight, from almost obese to morbidly obese(by BMI). From such a starting condition initial weight loss can be extremely rapid, yet after 12 weeks they lost only .94 kg?
This is a common source of misunderstanding, so I think it's worth clearing up:
Rapid initial weight-loss is usually just a loss of glycogen, the front-line starch form of energy storage in your body. For every gram of glycogen you clear out, 2-3 grams of water goes with it. As soon as you eat anything, all the water comes back instantly. That may be weight loss, but it's definitely not fat loss.
The metabolic pyramid roughly goes Alcohol > Sugar > Glycogen > Fat >>> Protein. Useful protein is extremely strongly conserved by your body, especially when fasting. Makes sense right, without your muscles, you wouldn't be able to hunt down and kill your next meal.
Fat loss when totally fasted happens basically at a rate of your BMR, which goes up a bit when fasted. This is due to an increase in noradrenalin in the fasted state. Your BMR is roughly proportional to your body weight.
A pound of fat contains 3500 calories. Your BMR is roughly ~2000 kcal per day for an average human, plus a bit more when fasted. This means when totally fasted you will lose roughly 2/3 of a pound of fat per day.
That rate remains pretty consistent throughout your weight loss regimen.
Over 12 weeks, they lost 2 pounds, or 7000kcal. That's a daily caloric deficit of 100kcal. Sounds like they weren't really on much of a diet.
I’m very familiar with this having recently lost exactly 2lbs per week for many weeks to hit a 35lb total. At least in my experience it really is much easier to lose weight initially. Not sure if there are studies on this or just anecdotes. When I was borderline obese, losing weight was simply a matter of being vaguely mindful about “eating better”. After I lost a decent amount that way losing more took deliberate effort and tracking calories. The people in the study didn’t even lose the weight you would expect from the effect of losing glycogen and water retention.
I suspect 16:8 with ad libitum eating is not going to trigger a reduction in glycogen, certainly not with 100kcal per day deficit. This feels more like a super light weight caloric restriction diet.
From my reading it sounds like calories eaten were not controlled between the two groups? Surely the study would be much more useful if the groups consumed the same number of calories per week?
If fasting results in wanting to eat less/more it's going to be completely unsurprising for that to result in more/less weight loss respectively, any weight loss due to the fasting itself, as opposed to standard calorie restriction is going to be difficult to extract from the data without a very large sample size.
Proponents of intermittent fasting seem to think it's more than just calorie restriction having an effect. I would think it would be useful to isolate why the treatment works.
Although I can also see why a clinical trial of a treatment would need to test the treatment as it would be applied in the real world, as opposed to a modified version intended to extract the why.
_When_ you eat your food has never directly related to weight loss, but by restricting your eating window, it is easier psychologically to not exceed your calorie limits for a given day than trying to stay within the calorie limit, but eating whenever you want.
I am curious if there is a study on if participants that use IF alongside calorie restrictive diet are able lose weight faster than people that don't fast.
I'm curious about that as well. My girlfriend and I do 18:6 intermittent fasting (she's very strict about it while I'm... less so). We also get a decent amount of exercise, and while I don't count calories anymore, I do try to limit my calorie intake. That, plus shifting from eating out a lot to cooking nearly every day during COVID restrictions has caused me to lose 15-20 lbs.
I'd tried intermittent fasting in the past, without making any dietary changes, and it never did anything. The only thing that's consistently allowed me to lose weight has been changing my diet. IF probably helps me, due to the psychological effect you mentioned.
> _ When_ you eat your food has never directly related to weight loss
I am not sure this is true. Insulin levels rise naturally just before waking, and apparently the same calories consumed at breakfast as at lunch has a different impact on fat accumulation.
The general theme I keep coming across is that calories in, calories out is too simplistic and that weight loss is driven by hormones which fluctuate with the circadian cycle and _what_ you eat with respect to those calories.
This compares 3 'structured' meals a day, to ad libidum eating over an 8 hour window.
...what's the point? why would anyone conduct this study, how does precisely three meals with pre-calculated caloric value relate in any way to eating as much as you want in 8 hours.
"In the in-person cohort (n = 50), there was a significant decrease in weight in the TRE group using the in-person weight measurements but not in the CMT group " ?
as i said lower down:
"ehhh, really depends on what effect size you are looking for. presumably you would want to see a 10% effect size or more. (detecting that intermittent fasting caused 2 lbs more weight loss than CMT would be statistically valid, but not clinically that important)"
power depends not just on n but also on the expected difference between the means (and the standard deviation of the mean)
ehhh, really depends on what effect size you are looking for. presumably you would want to see a 10% effect size or more. (detecting that intermittent fasting caused 2 lbs more weight loss than CMT would be statistically valid, but not clinically that important)
It is a pet peeve when a failure to reject the null hypothesis is reported (in confident language) as a confirmation of the null hypothesis. However, I do appreciate how clear the results section is in breaking down exactly what they did measure: both groups lost weight, and the time-restricted group lost more but not significantly more in the overall study.