Hacker News Reader

Comment by kaiju0 | Hacker News Reader

Have a feeling this is a correlation and not direct causation. You have to eat less / better to lose the weight. That helps with your insulin resistance.

View on HN

> Have a feeling this is a correlation and not direct causation

Classic. An HN commenter's gut feel (no pun intended) apparently outweighs a serious piece of work. Here's the results from the actual study [0] referenced in the article:

In this study, 1028 participants were randomized (INT, 424 [41.2%] [166 women (39.2%)]; INT-DPM, 426 [41.4%] [147 women (34.5%)]; CON, 178 [17.3%] [70 women (%39.3)]) between January 1, 2011, and February 24, 2017. The mean (SD) age was 65.3 (10.0) years, mean (SD) body mass index 31.2 (5) (calculated as weight in kilograms divided by height in meters squared), and mean (SD) follow-up 24.7 (13.4) months. A total of 156 participants progressed to type 2 diabetes, which comprised 39 of 171 receiving CON (22.8%), 55 of 403 receiving INT (13.7%), and 62 of 414 receiving INT-DPM (15.0%). There was no significant difference between the intervention arms in the primary outcome (odds ratio [OR], 1.14; 95% CI, 0.77-1.7; P = .51), but each intervention arm had significantly lower odds of type 2 diabetes (INT: OR, 0.54; 95% CI, 0.34-0.85; P = .01; INT-DPM: OR, 0.61; 95% CI, 0.39-0.96; P = .033; combined: OR, 0.57; 95% CI, 0.38-0.87; P = .01). The effect size was similar in all glycemic, age, and social deprivation groups, and intervention costs per participant were low at $153 (£122).

[0] https://jamanetwork.com/journals/jamainternalmedicine/articl...

I don't fully blame OP, the HN title is simply wrong.

The paper does not say that weight loss lowers the risk of diabetes. Which would imply a causal relationship. The paper says that "lifestyle intervention" lowers the risk of diabetes. This intervention could also explain the weight loss, hence a corollary relationship.

That said, there is such a thing as bad science and there is literally decades of it when it comes to nutrition and chronic diseases. And more keeps coming out all the time. I don't blame the OP for being skeptical, but he/she was skeptical of the wrong thing in this case.

HN's medical advice is on par with it's advice on law and economics.

*its (cause if there's one thing we do well, it's pedantry!)

I don't think the OP is denying the results of the study, but trying to understand if it is the acts associated with losing weight or actually weighing less.

Yeah in this case OP is right though, you can gain muscle mass, be very active, and eat healthy. It is 100% a correlation.

Actually I would describe what you posted an example of a limitation of the correlation, since it uses BMI.

I wonder if you could form a study to disambiguate it by using liposuction patients in some way.

edit: https://www.webmd.com/beauty/news/20000825/study-looks-at-wh...

Yep. Diabetes (type 2 at least) is widely understood to be the result of chronic insulin resistance, which can be remedied with diet. No one has shown a causation between weight and and diabetes but the correlation is definitely there because they have the same root cause.

In fact, weight alone cannot be the cause because there are plenty of overweight older individuals who never develop the symptoms of diabetes.

It's also possible that people who are suddenly interested in losing weight and exercising more have an increased awareness in what they're eating.

This is crap. There are people, myself included, which are simply predisposed to insulin resistance. For reference, I'm 145lbs @ 5'10", exercise daily, and eat relatively well. My last A1C was 5.6. To some extent, yes, diet will help; however, insulin resistance is a complex metabolic process that is absolutely rooted in your genetics, so simply telling people to eat less and/or lose weight is facile at best.

There may be a genetic component to Type 2 diabetes but _most_ people who are diagnosed with it are indeed quite overweight and have decades of poor lifestyle choices behind them. If we can help those people understand that diabetes is not always inevitable, even if it's not helpful for a small percentage, then that sounds like a net win for me.

You do not have to eat better to lose weight. You only have to eat less.

edit To correct confusion, I am saying "eat less" to mean "be at a caloric deficit." Weight loss is caused by caloric deficit, nothing else.

This mindset is why people find it hard to lose weight. Can't we agree that it's both?

How can it be both though? If you eat a little healthy food or a little garbage you'll lose weight, and if you eat a lot of healthy food or a lot of garbage you'll gain weight.

As far as weight loss is concerned, "healthy" food helps insofar as it makes it easier to eat less and exercise more, which is a property that some (not all) healthy foods accidentally happen to have, and which they share with a variety of less healthy calorie sources.

From my point of view (currently mid 40s and undergoing my very first attempt at losing weight in my life), losing weight of course means consuming fewer calories than you burn. However, eating healthy foods is what makes that a less miserable experience. Eating foods that take a while to digest and sit in your stomach for a while making you feel full like whole grains, nuts, fresh fruit and vegetables, and enough protein to keep your muscle mass, will make it a lot more likely to lead to success than simply "eating less."

Think about why you want to lose weight. Is it to be healthier, or is to reach an arbitrary number? If it's the latter then maybe you can just pound down 1500 calories of cake a day. If not, then you actually do need to eat healthy. The health benefits don't come from weighing less, they come from having a healthier body composition and less visceral (internal) fat.

> Think about why you want to lose weight. Is it to be healthier, or is to reach an arbitrary number?

Many possible reasons. You would be surprised at the number of people doing it mainly to be able to fit into pretty clothes.

It's difficult to find motivation to lose weight to be healthier when you don't actually have any health problems yet.

This attitude strikes me as harmful.

1. Why exercise if you never plan on engaging in vigorous hobbies or professions?

2. Why seek physio therapy for a stiff joint if you spend your days on the couch or in a chair?

3. Why see a dentist if you only receive calories from drinking soylent?

4. Why bathe if you never leave your house?

There's some philosophy of morals argument to be made here similar to the concept of whether it is good or bad to eat arsenic.

Not sure what your point is.

1. Because it's healthy and feels good

2. Because it's healthy and may stop future problems

3. Because it may stop future problems

4. Because it feels good

So why lose weight for the sake of losing weight? It's neither healthy nor fun if you do it the wrong way.

> So why lose weight for the sake of losing weight?

Because chronic obesity is associated with numerous morbidities.

The same answer you gave for (3) applies to avoiding obesity.

But how true is that for weight loss specifically?

Sure the food that's phrased as 'unhealthy' might not be what's best for you, but we shouldn't spread falsehoods.

https://www.npr.org/2010/11/12/131286626/professor-s-weight-...

That professor is stupid. Obviously he lost weight, it's thermodynamics. Energy in, energy out, or whatever. But he is no doubt less healthy after the experiment and probably more susceptible to things like diabetes.

This thread all started as a result of talking about weight loss, not health. Also, you're ignoring all of the health improvements weight loss brings even without any other change in behavior (diet, exercise, sleep).

Hmm. Worked fine for me. Calorie counting but no other dietary modification let me drop 30kg. Better (and less processed) food choices came about naturally as an effect of caloric restriction and understanding of the makeup of food I was consuming, not as the driver.

The notion that one has to pound salads and give up enjoyed foods is also a mindset that causes people to struggle in maintaining a healthy weight.

30kgs! Good on you ( assuming that is a healthy amount to lose). From my own point of view, I have grown to like a quality salad, and they are very easy to make quickly. Some fresh variety of greens from the store, half an avocado sliced, cubes of roasted golden beet, add roasted/salted pistachio nuts or sunflower seeds, layer a few chunks of roasted chicken on top, add olive oil and your choice of tasty vinegar (e.g., balsamic), and a dash of salt and mix. 5 minutes to prep if you have all the ingredients. Divine!

It doesn't have to be either one! You could just exercise more.

You mostly have to control what you eat. One donut takes like 3 hours to burn off at a good clip on a treadmill.

I think that thinking you can make up for eating habits with exercise at all will just get you into trouble. You should exercise for health, but unless you’re a professional athlete, your voluntary physical activity is a rounding error in terms of weight management

> One donut takes like 3 hours to burn off at a good clip on a treadmill

I don't know where you get this idea but I see this meme really often in defeatist "it's impossible to lose weight" discussions. It's so easy to research, I don't know why it spreads so easily.

A Krispy Kreme caramel glazed doughnut is about 220 kcal[0] and a very light treadmill session (30 minute, 5% incline, 3 mph) will probably burn in the 220 range[1].

> unless you’re a professional athlete, your voluntary physical activity is a rounding error in terms of weight management

Absolutely not. It's true that for somebody that eats/burns ~2000 kcal in a day, about 1500 of that is for just existing (the resting metabolic rate[2]). But even small amounts of exercise are significant: that 30-minute speedwalk is 10% of your calorie intake! Turning it up to a jog can easily be twice that.

[0]: https://s3.amazonaws.com/kkd-e1-images.kktestkitchen.com/eco...

[1]: https://www.sparkpeople.com/resource/calories_burned.asp?exe...

[2]: https://www.verywellfit.com/metabolism-facts-101-3495605

and to add on an important extra data point, 1lb of fat is roughly 3500 calories. So that one 220 calorie treadmill walk is 0.06lb or about 1lb lost per 16 walks, changing nothing else.

Or put another way, if your office is a 30 minute walk away then you can lose just over 2.5 lb/mo by just changing your routine to walking instead of driving (5×2×4×220÷3500 = ~2.5) and not changing your diet at all.

> One donut takes like 3 hours to burn off at a good clip on a treadmill.

You're way off. A "Frosted Snowflake Donut" from Dunkin weighs in at 270 calories. [1]

A 155-pound person will burn 298 calories in 30 minutes doing 12-minute miles on a treadmill, [2] which is probably less than "a good clip".

> You should exercise for health, but unless you’re a professional athlete, your voluntary physical activity is a rounding error in terms of weight management

My personal experience just does not bear this out.

I used to ride bikes a lot. The I stopped because I didn't have as much time for it anymore. I eat noticeably less. I also gained 20 pounds.

[1]: https://www.dunkindonuts.com/content/dam/dd/pdf/nutrition.pd...

[2]: https://www.health.harvard.edu/diet-and-weight-loss/calories...

Sure, I was abstracting "Eat less" to mean "Be at a caloric deficit", which is the only requirement for weight loss. Exercising more obviously meets that criteria as well.

Exercise helps you eat less. It really does have to be all 3.

No, and in fact, most people report an increased level of hunger when starting an exercise regime because of their increased energy expenditure.

No, because that's not how calories work.

Can confirm, lost 25kg this year.

Calories in == calories out is probably the worst lie in mainstream medical/nutritional guidance today. In a purely mathematical model is it obviously true but the reality is that our body deals with macro-nutrients (carbs, protein, fat) very differently.

It does, but over long enough time spans chronic caloric deficiency will result in weight loss, regardless of the form of the calories consumed.

Some calories are better than others for short term outcomes, of course.

Calories in/calories out ignores that fact that the _type_ of calorie matters a great deal. It also ignores the fact that the body is capable of moderating its metabolic rate in response to energy requirements and supply.

If calories in and out were _really_ all that mattered, I would be able to calculate my energy needs precisely and then eat exactly the right number of cupcakes (and only cupcakes) every day for the rest of my life. And yet we intuitively know that would be extremely unhealthy. Why is that?

I'm not disagreeing with you on the good calories, bad calories concept. Simply that calories in/out is, when applied as a chronic deficiency, effective at reducing weight over long periods of time. No, you should not eat 750 calories of cupcakes and cupcakes alone.

You're mixing two different arguments together. Calories in/out is the only requirement for weight loss. Calories in/out is a percentage of the overall picture we define as "health".

To lose weight, you only have to be at a caloric deficit. You can eat _only_ three cupcakes a day and probably lose weight (unless they're kinda insane cupcakes). If you only eat 3 cupcakes a day and nothing else, you will be malnourished and nutrient deficient, which I would extend to mean "unhealthy", but you will lose weight.

Reply on news.ycombinator.com