Some confusion about macronutrient divisions in Direct meals

[Burylancs]

@Burylancs - can you point me to the "famous Lancet article of 2018"? Do you think it was another formal paper from the DiRECT study or something from the Lancet editorial team? If I knew that it would help in hunting it down

You have set me off on another puzzle! Hope I can get further than I am with trying to make my minimalist monitor work with the mac mini running linux.

It's in the Lancet in 2018. Basically it was the closest thing to an account of the study. A link to the full article was given early last year. I'll have to check my downloads for it again. Otherwise its a trip to the Library for the paper version.

 

[Docb]

Having had a second look at the 2017 paper, specifically to look for what was used as a replacement diet, I can only find a reference to a product called Counterweight-Plus where the replacement meals provide 825-853 cal/day made up of 59%carbs, 13% fat, 26% protein, 2% fibre. That adds up to 100%.

Going to have to read it a few times even to get begin to get to grips with the statistics, but I am beginning to get the sense that the trial was as much about the problem of devising an experimental protocol which at least had a chance of providing statistically significant result as it was about finding a treatment for T2 diabetes. The results suggest that they did find such a protocol.

Going to have to find the 2018 paper!

 

[Docb]

Update. As far as i can see, having searched the Lancet contents lists, there is no 2018 paper. Taylor did respond to some comments on the original paper in a letter to the journal (vol 392 10155 p1307 Oct 13 2018) but that is all I have been able to find. These comments make no reference to composition of replacement meals.

The main paper, which I believe the preprint identified by @everydayupsanddowns is a copy, was published in Vol 391, 10120, pp541-551 December 2017.

@Burylancs. I am at a loss to find a source for the numbers you quote with respect to the formulation of meal replacements used in the direct trial. I do not think they are from anything published in the Lancet. Can you help any further?

 

[Burylancs]

Update. As far as i can see, having searched the Lancet contents lists, there is no 2018 paper. Taylor did respond to some comments on the original paper in a letter to the journal (vol 392 10155 p1307 Oct 13 2018) but that is all I have been able to find. These comments make no reference to composition of replacement meals.

The main paper, which I believe the preprint identified by @everydayupsanddowns is a copy, was published in Vol 391, 10120, pp541-551 December 2017.

@Burylancs. I am at a loss to find a source for the numbers you quote with respect to the formulation of meal replacements used in the direct trial. I do not think they are from anything published in the Lancet. Can you help any further?

Still tracking it down. It's the article where Taylor goes all Mystic Meg and predicts the future ... Saying ....'Our findings confirm that type2 diabetes of up to six years duration is not necessarily a permanent, lifelong condition.'

 

[everydayupsanddowns]

Saying ....'Our findings confirm that type2 diabetes of up to six years duration is not necessarily a permanent, lifelong condition.'

That’s the opening of the ‘Discussion’ section of the original peer-reviewed 2017 paper (page 14 of the PDF below) which details the diets totalling 100% as mentioned previously.

The original 2017 (pre publication) paper seems to be available here:
https://eprints.gla.ac.uk/153078/13/153078.pdf

 

[Burylancs]

That’s the opening of the ‘Discussion’ section of the original peer-reviewed 2017 paper (page 14 of the PDF below) which details the diets totalling 100% as mentioned previously.

The lead author of that paper is Lean with Taylor's name tagged on at the end of a long list. I'm talking about an article by Taylor.

 

[Docb]

Still tracking it down. It's the article where Taylor goes all Mystic Meg and predicts the future ... Saying ....'Our findings confirm that type2 diabetes of up to six years duration is not necessarily a permanent, lifelong condition.'

Keep going @Burylancs. The only thing I can find is the eprint from Glasgow referenced by @everydayupsanddowns which turns up as the only paper published in the Lancet and the comments you make about the DiRECT trial are not justified by any objective reading of it.

I can't find anything else in my searching. I am sure you are not making things up, so you must have got the numbers from somewhere and as you might have guessed I like to go back to the source documents.

 

[Docb]

Still to find anything referring to formulations but I do see some reporting of the ReTUNE study in 2021-2022 which seems to be a follow up to the DiRECT study. Cannot see a peer reviewed paper from ReTUNE, but from what I have seen, much of the reporting of it has lost sight of the basis for the DiRECT study.

Maybe something was said by somebody in relation to that study which is confusing the issue.

 

[everydayupsanddowns]

The lead author of that paper is Lean with Taylor's name tagged on at the end of a long list. I'm talking about an article by Taylor.

Prof Mike Lean and Prof Roy Taylor are co-authors (and principal investigators) of DIRECT as far as i understand it. I believe the 2017 paper i linked to is an open access preprint (author’s approved version) of the paper that was published in the Lancet, and presented at various conferences (including IDF, the slides of which you have seen.

See: https://www.directclinicaltrial.org.uk/DiRectStudyTeam.html

 

[Docb]

Update... I think I might have found it in a report of the ReTUNE study.

Very Low-Calorie Diet and 6 Months of Weight Stability in Type 2 Diabetes: Pathophysiological Changes in Responders and Nonresponders

OBJECTIVE. Type 2 diabetes mellitus (T2DM) is generally regarded as an irreversible chronic condition. Because a very low-calorie diet (VLCD) can bring abo

diabetesjournals.org

The Experimental Protocol says....

The VLCD consisted of a liquid diet formula (43% carbohydrate, 34% protein, and 19.5% fat; 2.6 MJ/day [624 kcal/day]; OPTIFAST; Nestlé Nutrition, Croydon, U.K.)

These percentages add up to 96.5%. They refer to a commercial product and my guess is that the "missing" 3.5% is not "fresh air" but the sort of inactive component of any food stuff we normally refer to as fibre. If you go back to the DiRECT study you will find that the formulations quoted there included the fibre content (2%).

I have yet to find the discrepancies quoted by @Burylancs. They are certainly not in any Lancet paper as far as I can see but maybe the numbers in this paper have been misquoted somewhere and then picked up.

 

[Docb]

One thing I might add.... Roy Taylor seems quite happy to add his name to the end of the list of authors on the papers, giving more emphasis to those more in need of recognition and in any case probably did most of the work. Most professors I came across would insist at being at the top of the authors list on any paper coming out of their department whether or not they had made any contribution to the work. He begins to earn my respect for that alone.

 

[Burylancs]

Update... I think I might have found it in a report of the ReTUNE study.

Very Low-Calorie Diet and 6 Months of Weight Stability in Type 2 Diabetes: Pathophysiological Changes in Responders and Nonresponders

OBJECTIVE. Type 2 diabetes mellitus (T2DM) is generally regarded as an irreversible chronic condition. Because a very low-calorie diet (VLCD) can bring abo

diabetesjournals.org

The Experimental Protocol says....

The VLCD consisted of a liquid diet formula (43% carbohydrate, 34% protein, and 19.5% fat; 2.6 MJ/day [624 kcal/day]; OPTIFAST; Nestlé Nutrition, Croydon, U.K.)

These percentages add up to 96.5%. They refer to a commercial product and my guess is that the "missing" 3.5% is not "fresh air" but the sort of inactive component of any food stuff we normally refer to as fibre. If you go back to the DiRECT study you will find that the formulations quoted there included the fibre content (2%).

I have yet to find the discrepancies quoted by @Burylancs. They are certainly not in any Lancet paper as far as I can see but maybe the numbers in this paper have been misquoted somewhere and then picked up.

The formula including the fiber content is for American audiences who list fiber separate to carbs. Thus the IDF slide says 61 % of carbs but the 2017 Lancet says 59 carbs and 2% fiber.
The example you quote just looks like slipshod carelessness and of course they're not expecting anyone to actually do the maths.
Still ploughing through my downloads, looking for that article where Taylor says putting T2 into 'remission' won't stop you getting complications.
How come Retune says 624 cals a day but DiRECT is supposed to be 880 ?

 

[Burylancs]

Prof Mike Lean and Prof Roy Taylor are co-authors (and principal investigators) of DIRECT as far as i understand it. I believe the 2017 paper i linked to is an open access preprint (author’s approved version) of the paper that was published in the Lancet, and presented at various conferences (including IDF, the slides of which you have seen.

See: https://www.directclinicaltrial.org.uk/DiRectStudyTeam.html

I thought Lean and Taylor were splitting up and Taylor had gone to the Dark Side, tagging along on papers by the low carb brigade such as Dr. Unwin.

 

[Docb]

@Burylancs - are we at the point where we can agree that your suggestion that in the Direct trial,

"50% of calories from carbs, 10-20% from Protein and 20-30% from fats. Which means Taylor's guinea pigs were getting up to 20% of their calories from Fresh Air."

is not a good summary of the low calorie diet used? The record shows that a commercially available meal replacement was used with a basic formulation, 59% carbohydrate, 13% fat, 26% protein, 2% fibre as given in the preprint of the Lancet Article presenting the Direct trial results.

It is a pity that neither of us has been able to identify the primary source for the numbers you quote. My suspicion is that they derive from a misrepresentation, which may or may not have been deliberate, of the formulations used in the DiRECT and ReTUNE trials and the differences in the way fibre content is assessed in different countries.

Could we also agree that, no matter what you think of basis for the trial and its ultimate value, it was a well conducted and reported piece of scientific work?

As for the relationship between Profs Lean and Taylor, I'd rather base my judgements on the formal record rather than academic gossip.

 

[Burylancs]

@Burylancs - are we at the point where we can agree that your suggestion that in the Direct trial,

"50% of calories from carbs, 10-20% from Protein and 20-30% from fats. Which means Taylor's guinea pigs were getting up to 20% of their calories from Fresh Air."

is not a good summary of the low calorie diet used? The record shows that a commercially available meal replacement was used with a basic formulation, 59% carbohydrate, 13% fat, 26% protein, 2% fibre as given in the preprint of the Lancet Article presenting the Direct trial results.

It is a pity that neither of us has been able to identify the primary source for the numbers you quote. My suspicion is that they derive from a misrepresentation, which may or may not have been deliberate, of the formulations used in the DiRECT and ReTUNE trials and the differences in the way fibre content is assessed in different countries.

Could we also agree that, no matter what you think of basis for the trial and its ultimate value, it was a well conducted and reported piece of scientific work?

As for the relationship between Profs Lean and Taylor, I'd rather base my judgements on the formal record rather than academic gossip.

No we're not at the point you mention, I'm still looking for the article in question.
As for being well conducted we actually know very little about how it was conducted. It gives the impression that DiRECT was conducted over a specific three month period in 2017 but then says that people were recruited over a period from 2014 to 2017. Does that mean different people were doing it at different times ? It was a rolling experiment ? What about the Control Group on standard measures. Did it actually exist in one three month period alon gside the Intervention Group or were the Control Group records just retrospective results taken from random three month periods between 2014 and 2017. What happened with the diet, were the Cambridge Diet shakes delivered once a week and the participants left to get on with it. Did the researchers have control of three hundred shopping trolleys, fridge doors and freezer doors ? In the article I'm still searching for again Taylor mentions Carers reporting on what they were feeding their charges. This shows third party interference between researcher and subject. It suggests that DiRECT was self-reported anecdotal evidence, generally thought to be the weakest form of medical evidence but I don't think you will find the word 'anecdotal' anywhere in the published reports on DiRECT.
In Taylor's latest outing in an article by Dr. Unwin in January Taylor says we randomly assigned 300 T2s to one arm of the study or other in DiRECT. Well we know that to be a bare-faced 'terminological inexactitude' . Because surgeries were randomly assigned to one arm or the other not individuals. Lean says they did that because they couldn't randomly assign patients because that would mean some people at the same surgeries following different protocols.
But no doubt Taylor's erroneous claim that they randomly assigned individuals will pass into the Mythology of DiRECT and make it sound like an RCT.

 

[Docb]

OK so, I will try another way of looking for some common ground!

If you look at Table 1 in the preprint, the first two sections give the weight and HbA1c results from the intervention and control groups. Those are physical measurements. Would you agree that those numbers are not invalidated by the issues you raise?

By the way, I also am a bit surprised by the lack of information regarding the practicalities of the test programme and wonder how it must have been organised. Perhaps where we differ is that I don't see it as something that calls into question the base data produced.

Keep looking for the document!

 

[everydayupsanddowns]

By the way, I also am a bit surprised by the lack of information regarding the practicalities of the test programme and wonder how it must have been organised. Perhaps where we differ is that I don't see it as something that calls into question the base data produced.

In a diabetes trial where I led the PPI team, recruitment took place over an extended period, but the actual interventions were organised so that intervention and comparator ‘courses’ for want of a better description took place at the same time.

I don’t think this is particularly unusual in clinical trials

It can take a while to get volunteers to sign up and agree, and then you might need to run the interventions in a way that ‘balances’. So someone may agree in principle to take part in the spring, but the thing may not happen until recruitment numbers had sufficiently been reached in the early autumn.

 

[Docb]

Agree @everydayupsanddowns, and I imagine that much of the detail could be found in the various dissertations and theses of many of those listed as authors to the paper.

To me, the clear, statistically significant differences, in weight loss and HbA1c reduction between the control and intervention groups is all the more remarkable in view of the practical difficulties inherent in conducting a trial of this sort.