Under that situation, it still has to be into positive caloric balance (some studies show you basically have to have >maintenance calories from carbs alone) you can get lipogenesis from carbs, the quantity from memory was over 700g/day for multiple days on end.
I recommend you read your two books from christmas there jeff. Lyle will even reference stuff in the keto book, but dont expect references from his UD book. If you go to his webpage he will even insult you for free 
I was interested in cycomikoâs criticism of Berardiâs discussion of certain studies and decided to check it out for myself. Unfortunately, my university doesnât subscribe to the journal that published the police officer study, but I was able to read the Agus high-GI versus low-GI study. For anyone else interested in the science, hereâs my take on that study.
-
It is true that the difference in total weight loss was not statistically significant. However, a t-test was used to compare the weight loss at the end of the 6-day period. Because the data from the other days are essentially thrown away by this comparison, this test lacks sensitivity. (As with all parametric statistical tests, it defaults to finding no difference unless the difference is large and/or the variablility very small. Hence, finding no difference never means we can be sure there is no difference; but when we do find a difference, we can be pretty sure there truly is one.) The systematic trend of the weight loss clearly favors the low-GI group. These data suggest that a significant difference may have been found if the authors had analyzed the trend over all the data (using a GLM, for instance), OR if they gained additional statistical power by using more subjects (they only had 10), or if the study lasted longer than 6 days (which is very hard to do though).
-
There were other significant and substantial differences between high-GI and low-GI: glycemic response and insulinemic response (to be expected); leptin; resting energy expenditure; and nitrogen balance. The difference in REE became greater at the end of the 6 days, suggesting a closer look at the long-terms effects of these diets. Also, nitrogen balance was POSITIVE with the low-GI diet and NEGATIVE with the high-GI one. Given the importance of these differences, particularly over time, I think Berardi represented the study fairly, though perhaps he oversimplified the weight loss aspect.
The thing that bothered me most about this study wasnât the systematic but small difference in weight loss, but the fact that the low-GI group ate more protein. In fact, the low-GI group ate almost TWICE as much protein. The authors concede that because of this confound, the differences cannot be attributed solely to GI. It is almost pointless to conduct a study in which confounding factors make it impossible to attribute causation. External validity (extending study results to the âreal worldâ) is ALWAYS hard; but if you donât have internal validity, whatâs the point of doing the study?
Someone needs to redo this study over a longer time period, with more subjects, and with identical macronutrient ratios. Still, this study showed that a low-GI, higher-protein diet has superior metabolic effects compared to a high-GI, lower-protein diet â and that these effects may increase over time.
Iâm late and new to this thread. However, The Atkinâs Diet sucks for the general public. Not because the diet itself sucks, but because most of the people following the diet are so fucking lazy that they havenât read the book. They heard about it from a friend or read a blurb in a newspaper. Therefore they kind of make up their own version of the diet and mess it all up. They arenât really on the diet. Just my two cents.
[quote]The systematic trend of the weight loss clearly favors the low-GI group. These data suggest that a significant difference may have been found if the authors had analyzed the trend over all the data (using a GLM, for instance), OR if they gained additional statistical power by using more subjects (they only had 10), or if the study lasted longer than 6 days (which is very hard to do though). [/quote]But to ASSume that something will have a result with longer time period and greater subjects doesnt always stand up in reality. In theory they got data on â20 peopleâ due to the crossover design.
If you really want to extrapolate, get figure 1 on page 903 and extend the findings further on from the last few days gradient (to allow for a change in water volume due to the reduced carbs) and you will notice that if they did, the groups weight loss would equal around day 10-12, but that is merely conjecture.
[quote]I think Berardi represented the study fairly, though perhaps he over simplified the weight loss aspect.[/quote]to go thru a study, not mention statistics nor the actual protein intake and claim that low gi will maintain RMR, and provides better weight loss is a touch more than an oversimplification.
[quote]The thing that bothered me most about this study wasnât the systematic but small difference in weight loss, but the fact that the low-GI group ate more protein. In fact, the low-GI group ate almost TWICE as much protein. The authors concede that because of this confound, the differences cannot be attributed solely to GI. [/quote]But the same authors also tried to state that both groups were protein sufficient. At 0.56g/kg (just over half the RNI) for the high GI group, on hypocaloric energy (~1500kcal for a 220lb average weight) is vastly too low (who here would survive with 56g of protein?), which could simply explain a reasonable amount of the data. Not to mention the differences in fat content, carb content, etc
[quote]whatâs the point of doing the study?[/quote]to keep money flowing thru from funding agencies.
[quote]Still, this study showed that a low-GI, higher-protein diet has superior metabolic effects compared to a high-GI, lower-protein diet â and that these effects may increase over time. [/quote]Extrapolation from limited data doesnt always add up in the real world. The differences in weight loss was 0.39kg which without even bothering with nitrogen balance is easily explainable by reduced glycogen/water volume.
To quote lyle
True. But in a between-groups design, 20 subjects is still far too few for a study like this.
Yes, perhaps. But no one (including the authors) represented weight loss as the benefit of the diet. The authors were mainly concerned with REE.
I disagree on this point. It is not the job of a consumer-oriented article to critique scientific studies and explain details of the methodology and design, let alone the statistics. No one cares about the statistics, just the ultimate interpretation of them. It is reasonable for an article to distill correctly the authorâs conclusions. It is up to researchers to debate whether those conclusions are correct. As youâve pointed out before, people without a knowledge of research methods are not in a position to judge those details anyway.
I didnât say that the study showed a superior effect for weight loss: it didnât. But the study did in fact show superior metabolic effects of one diet over the other.
[quote]So you know, this is one of the worst studies ever done. It had 4 count 'em 4 independant variables. It didnât prove shit [/quote] I donât know the study youâre referring to, but having 4 independent variables does nothing to reduce the validity of an experiment, absolutely nothing. In fact, multiple regression is powerfully illuminating when several important factors affect an outcome, especially when those factors interact. When there are interactions, the effects of a factor in isolation will be misinterpreted.
[quote]Yes, perhaps. But no one (including the authors) represented weight loss as the benefit of the diet.[/quote]Um, go back and read lean eating, because Berardi was representing weightloss as a benefit of this style of diet.[quote] The authors were mainly concerned with REE. [/quote]Hte only reason they were concernbed with teh REE was that was the only significant thing they could pick. If they didnt get a result, tehy wouldnt have talked about it.
[quote]I disagree on this point. It is not the job of a consumer-oriented article to critique scientific studies and explain details of the methodology and design, let alone the statistics.[/quote]So, whenever John writes an article, he can then delve into any data, whether significant differences were noted or not, and represent to people that it is relevant? [quote] As youâve pointed out before, people without a knowledge of research methods are not in a position to judge those details anyway. [/quote]John is a Phd candidate. People on this site, and his own, are trusting him to provide accurate representations of the available research to guide them to futher success.
If the research he bases the articles on are poor quality, or ignorning facets of statistics, shouldnt people be allowed to know this?
[quote]I didnât say that the study showed a superior effect for weight loss: it didnât.[/quote]But you were trying to draw out the fact that if it was longer, with more people, it will show a difference, but that is merely conjecture.
[quote]But the study did in fact show superior metabolic effects of one diet over the other. [/quote]yes, a change in REE, but berardis commetns were "The bottom line here is that when all else is equal, a diet containing mostly low-GI carbohydrates is superior to a high-GI diet for losing fat, preserving metabolic rate, and maintaining healthy insulin sensitivity and glucose tolerance."Everything else being equal. But their not.
[quote]I donât know the study youâre referring to, but having 4 independent variables does nothing to reduce the validity of an experiment, absolutely nothing. [/quote]Yes, if you actually read it, it said âa quote from lyle mcdonaldâ
and anywhosâ multiple variables can reduce the validity in terms of this research. No-one can state what caused the reduction in REE. if they produced more statistics possible, but I can almost bet that they did,and found nothing, which is why they wouldnt be in the published paper (AJCN is an expensive journal)
cyco,
for the most part i do agree with you, however i think your opinion is to simplistic.
your right about calories in versus calories out. i dont care if your eating high carb or low carb if your eating too many calories you will not lose bf.
that said insulin plays a major role in this as well. especially for the average american that is insulin resistant in the first place.
i have been training religiously and eating a damn good diet (imo) for years now. still my insulin sensitivity is not very great. diabetes type II runs in my family. i am no where near being diabetic but my insulin sensitivity isnt the greatest either.
when i eat low carb i drop bf quick when i cut carbs out completley and eat tons of bacon and steak i lose even more bf. believe me the few times i have completley cut out carbs my calories were still high. your calories add up real quick when your eating 2-3 ultimate cheeseburgers (minus the bun) per day.
insulin can be your best friend when your training for hypertrophy and your enemy when your dieting.
in the end i think it is again individual. for an athelte that is sprinting or playing basketball 3 hours per day eating a bunch of carbs will not hurt him at all. however for the average bodybuuilder person trying to make body comp improvements controlling insulin is key.
It is not my goal to pick a fight with you, cycomiko, and hijack this thread. However, this is not true. The entire motivation of the study, as discussed in the introduction and discussion, was to investigate the effects of food choices on preserving metabolic rate while dieting. You complained that Berardi didnât represent the study fairly in his article; but he basically presented the interpretation of the authors. Your statement above, though, doesnât represent the study fairly either. The authors found several significant hormonal differences, as well as a difference in REE which was the main point under investigation.
[quote]People on this site, and his own, are trusting him to provide accurate representations of the available research to guide them to futher success.
If the research he bases the articles on are poor quality, or ignorning facets of statistics, shouldnt people be allowed to know this?[/quote] I think that the debate is great and healthy. Thatâs one reason I like this site and why Iâm contributing to the debate.
[quote]But you were trying to draw out the fact that if it was longer, with more people, it will show a difference, but that is merely conjecture.[/quote] Yes, and I think my statement (âthe data suggest. . .it mayâ) shows that I was conjecturing. But if there were truly no difference in weight loss (rather than just a lack of statistical power), I would expect the difference to change direction randomly from day to day. Instead, weight loss was systematically greater for the low-GI group. So itâs a supported conjecture. However, you may be right that the difference may have disappeared over a longer time, rather than be maintained or increase. Your conjecture is supported by the apparent trends in the existing data.
In fairness to Berardi, his âbottom lineâ was supported by many other studies in addition to this one.
[quote]Yes, if you actually read it, it said âa quote from lyle mcdonaldâ
and anywhosâ multiple variables can reduce the validity in terms of this research. No-one can state what caused the reduction in REE. if they produced more statistics possible, but I can almost bet that they did,and found nothing, which is why they wouldnt be in the published paper (AJCN is an expensive journal)[/quote]
Iâm completely lost on the lyle mcdonald thing. I must have missed something. But multiple independent variables do not reduce validity, period. Actually, the amounts of fats, carbs, and protein in the Agus study are known. Perhaps you could get their data and analyze the effect on REE with each macronutrient as a separate variable. Multiple regression would show how much of the effect is due to each variable. That would be cool!
average american insulin resisnant.
not really
The figure is about 25% of the US population, and this figure has a bit to do with the obesity, and also with the lack of physical activity.
Once again, you are saying what works for you (not nessecarily everyone), and thats the way to go.
no way that stat is true!
i could have swore the last time i checked it was something like 60% of adult americans are overweight and about half of those are obese.
[quote]The entire motivation of the study, as discussed in the introduction and discussion, was to investigate the effects of food choices on preserving metabolic rate while dieting. [/quote]Not picking a fight either, but this is assuming that this was the plan when the study started. I have seen tons of papers that started under one design, and went to others to get published becuase what they were looking for wasnt significant.
[quote]Berardi didnât represent the study fairly in his article; but he basically presented the interpretation of the authors[/quote]basically? âIn a study by Agus et al (2000), it was demonstrated that during a short, 6 day, low-calorie diet, a low-GI carb intake preserved metabolism and enhanced fat loss vs. a high-GI dietâ preserved metabolism, yes(but due to protein differences?). enchanced fat loss, no. Even then he failed to state the huge area of protein intake.
[quote]the main point under investigation. [/quote]assuming this was so at the begining of the research.
[quote]Instead, weight loss was systematically greater for the low-GI group. So itâs a supported conjecture.[/quote]THe way that it shows, a rapid decrease for the low gi over 2-3 days adn then flatten out, and similar but slightly less rapid for the high GI, points to one main thing, water. But without studying water balance tis unknown.
[quote]In fairness to Berardi, his âbottom lineâ was supported by many other studies in addition to this one. [/quote]Just ot make it clear, Im not picking on it just for the sake of it. But, Berardis comemnts were supported by 3 other studies, 1 that shows an improved OGT, and teh other 2 were not controlled for caloric consumption, so not all is being equal.
[quote]Multiple regression would show how much of the effect is due to[/quote]Probably would be fun, but since they didnt do it, or at least didnt provide it makes me think that it would end up showing nothing.
Have a good new year
[quote]no way that stat is true!
i could have swore the last time i checked it was something like 60% of adult americans are overweight and about half of those are obese[/quote]
That was an population estimate of clinically insulin resistant people, not overweight nor obese.
Just because one is overweight, doesnt mean they are insulin resistant.
no but its a pretty good indicator.
The bottom line is that the Atkins Diet is popular right now because it works.
Jeff no one said it didnt work