#434: Is a Vegan Diet Really Best for Diabetes?

1. Detailed Study Notes
2. Transcript

To view our Premium content, please log-in to your account or subscribe to Premium.

Danny’s Key Idea from this episode is: Evaluating claims of “This is the best diet for X”

If someone makes that claim, on hearing that, your interest should be immediately piqued and your brain should be racing into critical appraisal mode. This is often followed up with evidence showing such a diet can have beneficial impacts. But that is not dealing with the original point of contention. If someone is going to make a claim that the BEST way to treat diabetes is with diet X, then the burden of proof becomes a sufficient amount of evidence showing actual superior magnitude and/or prevalence of beneficial outcomes in humans, of using that dietary intervention, relative to others.

In the context of treating type 2 diabetes, as we discussed in this episode, we need to first be clear on the outcome we’re concerned with. And then define that precisely. Putting type 2 diabetes into remission has a specific meaning. We can’t use “curing diabetes” as a synonym.

In 2021 a Consensus Report was published by a group of experts (Riddle et al. 2021), agreeing on a global definition of diabetes remission. They have agreed that a person with type 2 diabetes is in remission if:

1. they have had an HbA1c level below 6.5% (48 mmol/mol) for a least three months* AND
2. they have not taken any medications to manage their blood glucose levels during this time.

So if that’s going to be our outcome, and someone wants to advocate for a specific type of dietary intervention as being the “best” way to achieve that, then they need evidence that:

1) That dietary intervention leads to remission in humans

2) It is superior to other dietary interventions aiming to achieve the same thing.

With diabetes treatment, from a dietary standpoint, we know that the greatest impact on outcomes has been down to what induces the greatest amount of weight loss.

Other diets may have some benefits. And this is the focus of current research looking at high-protein diets, low-carb diets, and indeed plant-based diets. However, as of right now, saying one of those is universally better is not in alignment with current evidence. And none of them have achieved what has been seen in the very low-calorie meal replacement diets of the past few years.

Until such a time that one of them does, if ever, saying something like a WFPB diet is the best diet to treat diabetes is just not accurate. It can certainly help an individual. But let’s just make sure what is being communicated isn’t over-selling what is actually occurring.

You are currently not signed-in as a Premium subscriber. Key Ideas are for Premium subscribers only.

What is Diabetes Remission?

Until relatively recently, the word remission was defined differently by various organizations or research groups. In 2021 a Consensus Report was published by a group of experts (Riddle et al. 2021), agreeing on a global definition of diabetes remission.

This was a consensus statement from the American Diabetes Association (ADA), the Endocrine Society, the European Association for the Study of Diabetes and Diabetes UK jointly published in Diabetes Care, the Journal of Clinical Endocrinology & Metabolism, Diabetologia, and Diabetic Medicine, respectively.

They have agreed that a person with type 2 diabetes is in remission if:

1) they have had an HbA1c level below 6.5% (48 mmol/mol) for a least three months*

AND

2) they have not taken any medications to manage their blood glucose levels during this time.

*In situations where an HbA1c test is determined to be an unreliable marker of chronic glycaemic control, fasting plasma glucose <7.0 mmol/L or eA1c <6.5% calculated from CGM values can be used as alternate criteria.

Diabetes UK position statement on remission:

• “Diabetes UK’s position is that the term “remission” currently best reflects the situation where HbA1c levels return to below the diagnostic criteria for diabetes and the person is relieved of diabetes symptoms without glucose lowering medications, without downplaying the risk of relapse and essential need for ongoing review. There is currently no evidence that this state could remain permanently. For this reason, Diabetes UK does not support the use of the term “cure” and any other that suggests a permanent situation. We know that people with type 2 diabetes, and some within the healthcare system, sometimes talk about reversal of diabetes. Whilst we will sometimes use this term to engage people with the concept of remission and to ensure they are accessing our information and advice on remission we consider the term “remission” to be more accurate.”

In previous episodes of the podcast, diabetes remission has been discusses in detail. If you haven’t listened to any of these three episodes, then adding them to your to-listen list will help dive into more details:

1. #294: Nicola Guess, PhD, RD – Prediabetes & Type 2 Diabetes Nutrition
2. #331: Prof. Roy Taylor – Diabetes Remission, Very-low Calorie Diets & the Twin Cycle Hypothesis
3. #405: Adrian Brown, PhD – Dietary Strategies in Type 2 Diabetes

Hypothesis: Low-fat, plant-based diets are “best” for diabetes

While there are many claims about different dietary strategies, in this episode we focused specifically on the claim that a low-fat, whole-food plant based (WFPB) diet is the “best” diet to treat and/or “reverse” type 2 diabetes.

There are various examples of this claim, typically from the vegan community, but most commonly there are two sources that people online (colloquilly) cite:

1. The ‘Mastering Diabetes’ program
2. Work from Dr. Neal Barnard (either his gen. pop. books, or his published studies)

‘Mastering Diabetes’ Program

The “Mastering Diabetes” program is a lifestyle and dietary program, which at its core is a plant-based, high-carb, low-fat diet. Content on their website states that “..the best lifestyle to reverse insulin resistance and get your blood sugar levels back to normal is a low-fat, plant-based, whole-food diet.”

In addition to only consuming plant food sources, they advocate for keeping fat intake “to a minimum”, even from sources like avocado, olive oil, chia seeds, etc. The online coaching program provides 16 weeks of instruction about how to transition to a low-fat, plant-based, whole-food diet, using videos, recipes, and downloadable worksheets. Clients are provided with continued support for one year via online discussion forums and live video conference coaching.

Their website states that “Our Method is Scientifically Proven to Lower A1c, Body Weight, and Reduce Medication Use”.

This is based on a pilot study that was published in the American Journal of Lifestyle Medicine. Some details:

• This pilot study was an online survey of current and former clients of Mastering Diabetes (MD)
• Only 8.9% of participants responded to the survey!
  • In addition, a notable limitation is that people who do respond to the survey are most likely to be those who experienced favorable outcomes, thus resulting in an “emotional incentive to participate”, as the authors note.
• Demographics: 80.6% were female, mean age was 56 years
• Self-reported results:
  • 78.4% reported weight loss
  • 68.8% reported decreased HbA1c
  • 52.4% reported decreased medication use
• The paper itself states: “Study limitations prevent drawing generalizable conclusions”
• 2 of the authors are the owners of the program

So does a low-fat WFPB diet provide the best dietary treatment option for type 2 diabetes? Let’s look at different lines of evidence…

Epidemiology

1. Adventist Health Study-2
   • Long-term cohort study
   • Prevalence of diabetes was 49% less among vegans and 46% less among lacto-ovo vegetarians compared with non-vegetarians
2. Epic-InterAct (subcohort of EPIC)
   • High intakes of fruit and vegetables and lower intakes of processed meats, SSB, and refined grains, were associated with lower T2D risk (surprise, surprise!)
3. Jannasch et al., 2017 – Systematic review of dietary patterns
   • Alternate Healthy Eating Index, DASH, and Med diets, all associated with lower risk, with group food components more informative than single foods.
4. Chen et al., 2018 – Rotterdam Study
   • Nice epidemiological work using plant-based dietary index, with baseline and follow-up measures of glucose/insulin to determine HOMA-IR.
   • Higher PDI scores associated with .05 lower HOMA-IR and 13% lower T2D risk.

The healthy eating index (HEI), developed and updated in 2015, is a measure by which assess how well a set of foods aligns with recommendations of the Dietary Guidelines for Americans. The alternative healthy eating index (AHEI), as the name suggests, was created as an alternative to the HEI, with the focus on chronic disease risk reduction.

• See podcast episode with Prof. David Jacobs (#403)

The whole food plant-based (WFPB) diet is not really represented in epidemiology. Those advocating for such a diet tend to cite vegetarian research, but this is not the same dietary pattern. Vegetarian diets show statistically significant reductions in diabetes risk. And the foods/food groups that make up much of the WFPB diet have beneficial associations with diabetes risk. Again however, there is no solid epidemiological work that looks at the WFPB diet specifically.

Suggested Mechanisms

A review paper by Jardine et al. (2021) published in Advances in Nutrition put forth some mechanisms by which WFPB diets can treat type 2 diabetes:

• “Plant-based eating patterns are beneficial for patients with diabetes by improving insulin sensitivity and improving body weight.”
• “… low-fat, plant-based (especially vegan) diets improve glycemic control because of their ability to reduce lipid accumulation in muscle and liver”

But if we’re trying to work out is there something uniquely beneficial about this intervention, then we need to ask whether these mechanisms are unique to a WFPB diet?

Similarly, consider other components of the diet that have been hypothesised to reduce insulin resistance and/or put diabetes into remission:

• High fibre intake?
  • Potentially not for liver fat, but for peripheral insulin sensitivity
  • e.g., see Prof. John Kirwan’s group and wholegrain intakes:
    • Malin et al., 2018 – insulin resistance and postprandial blood glucose
    • Malin et al., 2019 – beta-cell function
• Low saturated fat intake?
  • Highly plausible
  • Hannele Yki-Jarvinen’s group has shown the role of saturated fat in increasing liver fat and insulin resistance
    • Clinical trial: Luukkonen et al., 2018
    • Review: Yki-Jarvinen, 2015
• Reducing red meat intake?
  • Some components linked with T2DM – e.g. review by Kim et al., 2015
• Focus on whole foods reduces added sugars/refined grains?
  • Also plausible given relationship between insulin resistance, liver fat, and the role of added sugars in driving triglyceride formation:
    • Evidence looking at diet and non-alcoholic fatty liver disease suggests that, under hypercaloric conditions, added sugars increase de novo lipogensis (DNL).

This leaves us with two questions: Which of these have evidence? And of those that have evidence, are those components exclusive to a plant-based, low-fat diet?

• e.g. if we are suggesting the benefit is from having a high-fibre, low SFA diet, primarily from minimally-processed or whole foods, that includes lots of vegetables… such a diet can be acheived with an omnivorous dietary pattern. Even a low-carb diet can be set-up to meet these criteria.

While some in the vegan community claim dairy increases risk of type 2 diabetes, strong evidence exists for a protective effect of yogurt and low-fat dairy foods on risk for type-2 diabetes. For more on this, see this Sigma Statement on Dairy & Health.

Crucially, it is well established that the primary driver of diabetes remission (and risk reduction) is weight loss.

• Guess, 2018: “Overall, these trials and supporting data show that weight loss is the primary driver of T2D risk reduction in people who are overweight”
• Brown et al., 2021: “The most significant predictor of remission is weight loss”

Given that adoption of WFPB diet commonly leads to weight loss, the most plausible conclusion is that it is just one (of several) dietary approaches that could lead to weight loss and thus improved outcomes. Pointing to trials showing a WFPB diet leads to weight loss can’t be used as evidence they are “best” for diabetes remission.

Human Interventions: RCTs

BROAD Study

Wright et al., 2017 – The BROAD study: A randomised controlled trial using a whole food plant-based diet in the community for obesity, ischaemic heart disease or diabetes

• A community-based program with 65 participants (aged 35-70) having a BMI in either the overweight (BMI>25) or obesity (BMI>30) category.
  • People in GP practices in a region with New Zealand’s highest rates of socioeconomic deprivation, obesity and type 2 diabetes.
• Participants had a diagnosis of one of:
  • Type-2 diabetes
  • Ischaemic heart disease
  • Cardiovascular risk factors of hypertension or high cholesterol
• Intervention group:
  • Low-fat, plant-based diet (emphasising wholegrains, legumes, vegetables and fruit).
  • Dietary fat intake of 7-15% – Participants were asked to avoid all oils and high-fat plant foods (e.g., nuts and avocados).
  • No restriction on total calories (counselled to eat until satiation).
• Relevant results:
  • Two participants in the intervention group with type-2 diabetes no longer met the diagnostic criteria at both 6 and 12-months.
  • HbA1c decreased 5 mmol/L in the intervention group compared to controls at 6-months.
• Massive bias issues:
  • The researchers discussed prescribing with the participants’ physicians – Attending physician bias
    • 29% reduction in medications but large blood pressure increases suggests participants were taken off meds prematurely.
    • Contrast with Lyon Diet Heart Study, which blinded doctors to the fact participants are in a study.
  • Authors say they “ethically discontinued” the control group at 6 months (and offer the intervention to them) due to magnitude of benefit
    • This is despite the magnitude of the difference at 6 months being very small in relation to the main outcomes
    • So therefore the 12-month data is essentially uncontrolled
  • Investigator bias
    • Situations where researchers “convey to research participants in subtle or unconscious ways the results or patterns the researcher would prefer to see.”
    • In this study, the reseachers discussed the potential benefits of the intervention with participants before randomisation
      • They were also shown media content advocating for the dietary pattern.
      • “Special events included screening the documentary ‘Forks Over Knives’ and an accompanying film endorsing the WFPB diet”
  • Hawthorne effect: individuals modify an aspect of their behavior in response to their awareness of being observed.

Barnard et al., 2006 – RCT in Diabetes Care

Barnard et al., 2006 – A low-fat vegan diet improves glycemic control and cardiovascular risk factors in a randomized clinical trial in individuals with type 2 diabetes

• 22-wk randomized controlled trial
• Compared a low-fat, vegan diet to a conventional portion-controlled diet (as per American Diabetes Association (ADA) guidelines).
• Vegan diet = ∼10% of energy from fat, 15% protein, and 75% carbohydrate
• The ADA portion-controlled diet = 15–20% protein, <7% saturated fat, 60–70% carbohydrate and monounsaturated fats
• Those with a BMI >25 kg/m2 were prescribed calorie deficits of 500–1000 kcal/d to bring BMI into normal range
• Relevant results:
  • 43% (21 of 49) of the vegan group and 26% (13 of 50) of the ADA group participants reduced their diabetes medications
  • Among participants who made no medication changes, the vegan group had a significantly greater reduction in HbA1c (1.23% vs. 0.38%)
• Body weight decreased 6.5 kg in the vegan group and 3.1 kg in the ADA group
• Crucially: To test whether the effect of diet on A1C was mediated by body weight changes, a regression model was constructed, including baseline A1C, weight change, and diet group as predictors of A1C change, among those whose hypoglycemic medications remained constant. In this model, the effect of diet group was no longer significant.

Kahleova et al. 2016 – RCT on beta-cell function/insulin

Kahleova et al., 2016 – A Plant-Based Dietary Intervention Improves Beta-Cell Function and Insulin Resistance in Overweight Adults

• 16-Week RCT
• 75 participants living with overweight/obesity
• WFPB diet (~75% carbohydrates, 15% protein, and 10% fat) vs. habitual diet (control)
• Most relevant finding is reduction in basal insulin secretion and increase in insulin secretion as a function of plasma glucose concentration
  • Note however, that this is only reported in a figure in the paper and not in the main table of results.
• There were no significant differences in oral glucose insulin sensitivity, mean plasma glucose or insulin values, total insulin secretion, and other parameters.

Kahleova et al., 2020 – Liver Fat

Kahleova et al., 2020 – Effect of a Low-Fat Vegan Diet on Body Weight, Insulin Sensitivity, Postprandial Metabolism, and Intramyocellular and Hepatocellular Lipid Levels in Overweight Adults

• Authors reported: “Low-fat vegan diet reduced hepatocellular lipid concentrations by 34.4% and intramyocellular lipid concentrations by 10.4%. These changes in hepatocellular and intramyocellular lipid concentrations correlated with changes in insulin resistance”.
• However, in a subgroup of 44 participants, liver fat decreased from 3.2% to 2.4% in the WFPB group and increased from 3.3% to 3.6% in the control group…
  • The threshold for NAFLD is 5%, thus all participants were already in normal ranges.
  • The change was correlated with the loss of bodyweight in the WPFB group (r = .42).
• The strongest correlations were for changes in bodyweight and fat mass, which were correlated with the change in insulin resistance. So the reported direction of effect omits the central role of weight loss.
• Significant reduction in fasting insulin levels which is an encouraging finding, but again, this was primarily mediated by weight loss.

Anderson & Ward, 1979 – Very high-fibre intake

• Metabolic ward study on lean men with type 2 diabetes who were taking insulin
• High-carbohydrate, high-fiber (HCF) diet:
  • < 10% calories from fat
  • 70% from carbohydrates
  • 65 g of fiber/day
• Body weights were kept stable by simply having participants eat more if they lost weight on the HCF diet
• Relevant results:
  • ~50% were able to discontinue insulin, and the remainder were able to significantly reduce their insulin
  • The average insulin dose decreased from 26 units on the baseline (control) diet to 11 units on the HCF diet.
  • Fasting and postprandial plasma glucose values were lower in most patients on the HCF diets than on the control diets, despite lower insulin doses.
• Important: An interesting and key feature of this study is that participants were lean. Therefore this means that these people had T2D that was not complicated by the high amount of visceral/central adiposity, that is much more common in T2D today.
• Additionally, it is not possible to extrapolate these results to the current day context, especially given that this study has never been replicated.
• Based on the work on wholegrains from Prof. John Kirwan’s group referenced earlier, it is still certainly plausible that very high fibre diets like this could have a unique benefit via increased peripheral insulin sensitivity and improved glucose disposal.
• For more discussion of the Anderson & Ward paper, listen to episode 385 of the podcast on Insulin Resistance and Diet.

Yokoyama et al., 2014 meta-analysis on vegetarian diets

Yokoyama et al., 2014 – Vegetarian diets and glycemic control in diabetes: a systematic review and meta-analysis

• Meta-analysis of 6 RCTs involving 255 study participants with type 2 diabetes
• Found that vegetarian diets were associated with a greater reduction in HbA1C

Key Point: No intervention has directly tested and shown remission or reversal of T2D in a modern diabetic population using a WFPB diet. Interventions to date show proxy outcomes (e.g. reduction in liver fat or improved HbA1c), which although relevant, are not remission, and are often in participants that do not have T2D.

Key Question: Is a WFPB diet better than other dietary approaches?

From Brown et al., 2021: “Current available evidence suggests that a range of dietary approaches, including low energy diets (mostly using TDR) and low carbohydrate diets, can be used to support the achievement of euglycaemia and potentially remission. The most significant predictor of remission is weight loss and, although euglycaemia may occur on a low carbohydrate diet without weight loss, which does not meet some definitions of remission, it may rather constitute a ‘state of mitigation’ of T2DM. This technical point may not be considered as important for people living with T2DM, aside from that it may only last as long as the carbohydrate restriction is maintained.”

The range of dietary approaches that could potentially lead to remission (most likely via weight loss) include:

• Very low-calorie diets (see discussion below)
• Low-carbohydrate diet
  • Commonly cited is a non-randomised trial by Virta Health (Hallberg et al., 2018) showing 25% remission after 1 year, 18% after year 2
• Low-fat diet
  • Look AHEAD trial reported a remission rate of 11.5% after a year and 7% at 4 year follow up
• Mediterranean diet
  • Esposito et al., 2014 reported a remission rate of 15% after a year and 5% at 6 year follow up

Therefore, there seems to be no superiority for a low-fat WFPB diet.

As noted above, weight loss currently seems to be the primary driver of achieving remission via dietary interventions. In addition, when it comes to ‘reversal’ of diabetes, this requires the actual restoration of beta-cell function. The only interventions that have so far been shown to restore this beta-cell function are those that lead to massive weight loss (i.e. 15% body weight loss). These interventions are very low-energy diets, typically in the form of meal replacements.

Very Low Calorie Diets

The highly influential work in this area comes from trials done by Prof. Roy Taylor and collegues:

• DiRECT Trial – Lean et al., 2018
• 2-year results of the DiRECT Trial – Lean et al., 2019
• Lim et al., 2011 – normalisation of beta cell function
• Steven et al., 2016 – Very Low-Calorie Diet

For more background context, these review papers are worth reading:

• Taylor, 2008 – Pathogenesis of type 2 diabetes: tracing the reverse route from cure to cause
• Taylor, 2019 – Calorie restriction for long-term remission of type 2 diabetes
• Taylor et al., 2019 – Understanding the mechanisms of reversal of type 2 diabetes

A group in Qatar replicated the findings of DiRECT, in the DIADEM-1 trial.

Conclusion

Currently, there is no direct evidence showing that a low-fat, whole food plant-based diet:

• Leads to reversal of diabetes (i.e. permanent restoration of beta-cell function)
• Leads to remission of diabetes (beyond the ability to induce sufficient weight loss to drive remission)
• Is superior to all other dietary interventions for treating type 2 diabetes

Therefore any claims that a low-fat WFPB diet is the “best lifestyle to reverse insulin resistance” or treat diabetes are not evidence-based.

You are currently not signed-in as a Premium subscriber. Detailed study notes are for Premium subscribers only.

You are currently not signed-in as a Premium subscriber. Episode transcripts are for Premium subscribers only.