Which Diet Is Best For Diabetics?

What Did This Study Show? 

Author: Dr. Stephen Chaney 

High Blood SugarWhen you were first diagnosed with diabetes, your doctor probably told you that your life will forever be changed. Among other things he or she probably told you that you would need to make some radical changes to your diet.

But what changes? Both the American Diabetes Association (ADA) and Diabetes UK (the British version of ADA) recommend:

  • An individualized approach. This recognizes that we are all different. What works for some diabetics may not work for others.
  • A diet that incorporates more non-starchy vegetables and minimizes added sugars and refined grains.

But these recommendations are vague. Most people want a specific diet to follow. It’s here that Diabetes UK and the ADA part ways.

  • Diabetes UK gives its highest recommendation to the Mediterranean diet.
  • The ADA gives equal recommendations to the Mediterranean diet and both low-carbohydrate and very-low carbohydrate diets.

But which diet is best? It’s hard to know because most studies compare one of these diets to the standard American diet (SAD), and anything is better than the standard American diet.

Fortunately, one recent study (CD Gardner et al, American Journal of Clinical Nutrition, 116: 640-652, 2022) directly compares the two extremes of ADA-recommended diets, the Mediterranean diet and the Keto diet.

How Was This Study Done?

clinical studyThis study recruited 33 participants with diabetes or prediabetes from the San Francisco Bay area. The participants in the study:

  • Were between 41 and 77 years old (average age = 60.5).
  • Were 61% male, 45% non-Hispanic white, and mostly (85%) college educated.
  • Had either prediabetes (61%) or diabetes (39%).
  • Had BMIs ranging from 22.7 (normal) to 39.7 (obese) (average BMI = 30 (borderline obese).
  • Had elevated levels of HbA1c (hemoglobin A1c, a measure of long-term blood sugar control).

People were excluded from the study if they were:

  • Underweight (<110 pounds) or morbidly obese (BMI ≥40).
  • Had extremely high cholesterol (LDL cholesterol >190 mg/dL) or blood pressure (>169 mmHg).
  • On insulin or certain medications to lower blood sugar levels.

This was a randomized, crossover, interventional study. Simply put, that means:

  • The study started with participants eating a typical American diet. The intervention was either a Keto diet or a Mediterranean diet.
  • Each patient was randomly assigned to one of the diets for 12 weeks. Then they “crossed over” to the other diet for 12 weeks. In this type of study each patient serves as their own control.
  • Finally, there was a 12-week follow-up period in which they could choose which of the two diets to follow.

It was a very well-controlled study:

  • Participants were given detailed guidelines to follow and received weekly individual education sessions by a registered dietitian and certified diabetes educator.
  • During the first 4 weeks of each diet, participants were provided at no cost all meals and snacks from a local food delivery service.
  • During the next 8 weeks of each diet, the participants purchased their own foods using the same guidelines they had been given during the first 4 weeks.
    • They were also provided with a recipe booklet and suggestions for diet-compliant menu items at local restaurants for each diet.
  • This was not designed as a weight loss diet. The participants were provided with 2,800 calories of food per day and instructed to eat until they were full.
  • Compliance with the diet was assessed in three ways:
    • During week 4 and week 12 of each diet phase, 3 unannounced 24-hour dietary recalls (2 on weekdays and 1 on a weekend day) were administered over the phone by a trained nutritionist.
    • Participants were also given an app to log in their food intake daily.
    • Participants on the Keto diet were given blood ketone monitors and strips.
  • Finally, at the beginning and end of the study and during weeks 4 and 12 of each diet phase participants went to a medical facility for blood work and weight measurements.

The primary focus of this study was measuring the effect of each diet on HbA1c. HbA1c measures blood sugar control over the previous 12 weeks (which is why each diet phase was 12 weeks long). But the study also measured the effect of each diet on LDL cholesterol, HDL cholesterol, and triglycerides.

What Were The Diets Like?

Vegetarian DietThese were not ordinary versions of the Mediterranean and Keto diets:

  • Sugar and refined flour are often part of the diet in Mediterranean regions. So, this study used the “Mediterranean Plus (Med-Plus)” diet which restricts both sugar and refined grains.
  • Keto convenience foods are often a witch’s brew of artificial ingredients. So, this study used the “Well-Formulated Keto Diet (WFKD)” which is composed of whole, unprocessed foods. In fact, both diets were whole food diets.

In summary, the two diets were:

  • Alike in that both emphasized non-starchy vegetables and minimized sugar and refined grains.
  • Alike in that they were both whole food diets.
  • Different in that the Keto diet eliminated legumes, fruits, and whole grains while the Mediterranean diet included them.

The macronutrient composition of the two diets was about what you would expect.

USDA

Guidelines

Baseline Keto

(Weeks

1-4)

Keto

(Weeks

5-12)

Med

(Weeks

1-4)

Med

(Weeks

5-12)

Protein 10-35% 18% 25% 22% 19% 21%
Carbs 45-65% 41% 12% 18% 37% 37%
Fat <30% 41% 63% 60% 44% 42%
  • The baseline diet was typical of the American diet. It was higher than recommended for fat. While carbohydrate intake appeared to be moderate, it was high in sugar and refined grains.
  • The Keto and Mediterranean diet interventions were separated into 2 phases. In phase 1 (weeks 1-4) every meal and snack were provided to the participants. In phase 2 (weeks 5-12) they purchased their own food.
  • As expected, carbohydrate intake was much lower, fat intake much higher, and protein intake slightly higher than baseline for the Keto diet. And this pattern was maintained during the 8 weeks the participants purchased their own food.
  • Macronutrient composition on the Mediterranean diet was not much different than baseline and did not change much during weeks 5-12.

The fat composition of the two diets was also different.

Baseline Keto

(Weeks

1-4)

Keto

(Weeks

5-12)

Med

(Weeks

1-4)

Med

(Weeks

5-12)

Monounsaturated 42% 48% 43% 52% 45%
Polyunsaturated 23% 15% 19% 23% 25%
Saturated 35% 37% 38% 25% 30%
  • The Keto diet was significantly lower in percent polyunsaturated fat and slightly higher in percent monounsaturated and saturated fat than baseline (the typical American diet) in weeks 1-4. However, remember that the Keto diet was 50% higher in total fat than the other diets. This makes it significantly higher in saturated fat than either the baseline or Mediterranean diets.
  • As expected, the Mediterranean diet was significantly higher in percent monounsaturated fat and lower in percent saturated fat than baseline in weeks 1-4.
  • Not surprisingly, both diets trended towards the baseline diet in the 8 weeks participants were buying their own food.

Other interesting differences between the two diets:

  • The Keto diet contained around 12% plant protein and 88% animal protein, while the Mediterranean diet contained about 50% of each.
  • Fiber intake decreased by 33% compared to baseline on the Keto diet, while fiber intake increased by 50% on the Mediterranean diet.
  • In terms of nutritional adequacy, the Keto diet was significantly lower in fiber, vitamin C, folate, and magnesium than the Mediterranean diet.

What Did The Study Show?

Question Mark1. Participants consumed around 300 fewer calories/day and lost about 15 pounds on both diets.

    • The authors speculated this was because both diets were more filling than the baseline diet, presumably because both diets were whole food diets while the baseline diet contained lots of processed foods high in sugar and refined grains.

2) Both diets reduced HbA1c (a cumulative measure of how much the diets improved blood sugar control compared to the baseline diet) by about the same extent.

3) LDL cholesterol (bad cholesterol) increased by about 10% on the Keto diet, while it decreased by about 9% on the Mediterranean diet. This difference was highly significant.

4) HDL cholesterol increased by about the same extent on both diets.

5) Triglycerides decreased by around 20% on the Keto diet and by 10% on the Mediterranean diet. This difference was also highly significant.

6) Finally, adherence to the Keto diet was less than for the Mediterranean diet. Plus, more people chose the Mediterranean diet during the follow-up phase when they were allowed to choose their own diet.

The authors concluded, “HbA1c values…improved from baseline on both diets, likely due to several shared dietary aspects. The WFKT [Keto diet] led to a greater decrease in triglycerides, but also had untoward risks from elevated LDL cholesterol and lower nutrient intakes from avoiding legumes, fruits, and whole, intact grains, as well as being less sustainable [easy to follow long-term].

Which Diet Is Best For Diabetics?

Mediterranean Diet Foods

Animal Protein Foods

Vs

 

 

 

 

Once again, I have covered lots of information in this blog. But if you are diabetic, you are probably wondering, “What does this mean for me?” Let me start by reviewing the purpose of this study.

  • This study was designed to compare the two extremes of recommended diets (Mediterranean and Keto) with respect to their effectiveness at keeping blood sugar under control.
  • These were both more restrictive versions of the two diets than most people follow. In this study, both diets:
    • Were whole food diets. No sodas, processed, or convenience foods were allowed.
    • Minimized the consumption of sugars and refined grains.

Now let me divide the discussion into two sections:

  1. Which diet is best for diabetics in the short term (in this case, 12 weeks)?
    • Participants consumed 300 fewer calories and lost about 15 pounds on both diets in spite of being given more than they could eat and not being encouraged to lose weight.
      • The authors attributed this to whole food diets being more filling.
      • However, it is also consistent with my contention that any restrictive diet will lead to short-term weight loss and improvement in blood sugar control. I summarize the 5 reasons for this phenomenon in last week’s “Health Tips From the Professor” article
    • Blood sugar control over 12 weeks, as measured by HbA1c, improved by the same amount on both diets.
      • That is consistent with the American Diabetes Association’s position that a variety of diets, ranging from Mediterranean to Keto, are suitable for diabetics.
      • This also means that you can forget the advice that diabetics need to follow a low carb diet and give up fruits, whole grains, and legumes to keep their blood sugar under control.
      • However, this is not a “get out of jail free card”. Diabetics do need to avoid sodas, processed, and convenience foods and minimize sugar and refined grains.
    • There was considerable individual variability. Some people did better on the Mediterranean diet. Others did better on the Keto diet.
    • This is consistent with the American Diabetes Association’s recommendation that diabetic diets should be individualized.

In short, this study suggests that in the short term (12 weeks) the Med-Plus and WFKD Keto diets are equally effective at promoting weight loss and improved blood sugar control for both diabetics and prediabetics.

However, there is considerable individual variability, meaning that diabetics can chose the diet that works best for them.

2) Which diet is best for diabetics in the long term?

If both diets are equally effective short term, the important question becomes whether they are equally successful and equally healthy long term.

As noted in the author’s conclusion, this study raised several “red flags” which suggest the Keto diet might be less successful and less healthy long term. But this is a short-term study.

You may be wondering, “What do long-term studies show?” Unfortunately, there are very few long-term studies to guide us. But here is what we do know.

    • There are multiple studies showing that the Mediterranean diet reduces the risk of diabetes, heart disease, and some cancers long term. There is no evidence that meat-based low carb diets are healthy long term. This includes the Atkins diet, which has been around more than 50 years.
    • A 6-year study reported that the group with the lowest carbohydrate intake had an increased risk of premature death – 32% for overall mortality, 50% for cardiovascular mortality, 51% for cerebrovascular mortality, and 36% for cancer mortality.
    • A 20-year study reported that women consuming a meat-based low carb diet for 20 years gained just as much weight and had just as high risk of diabetes and heart disease as women consuming a high carbohydrate, low fat diet.

In short, the few long-term studies we do have suggest that the Mediterranean diet is a better choice for long-term health and reduced risk of diabetes than low-carb diets.

The Bottom Line 

If you are diabetic or prediabetic, the American Diabetes association recommends a diet that is individualized and ranges from Mediterranean to low carb and very low carb (Keto).

However, low carb and Keto enthusiasts insist that diabetics need to follow a low carb or very low carb diet. And that seems to make sense. After all, aren’t carbs the problem for diabetics?

To resolve this question, a recent study was designed to compare the two extremes of the ADA-recommended diets (Mediterranean and Keto) with respect to their effectiveness at keeping blood sugar under control.

These were not ordinary versions of the Mediterranean and Keto diets:

  • Sugar and refined flour are often part of the diet in Mediterranean regions. So, this study used the “Mediterranean Plus (Med-Plus)” diet which restricts both sugar and refined grains.
  • Keto convenience foods are often a witch’s brew of artificial ingredients. So, this study used the “Well-Formulated Keto Diet (WFKD)” which is composed of whole, unprocessed foods. In fact, both diets were whole food diets.

In short, this study found that in the short term (12 weeks) the Med-Plus and WFKD Keto diets are equally effective at promoting weight loss and improved blood sugar control for both diabetics and prediabetics.

The authors concluded, “HbA1c values…improved from baseline on both diets, likely due to several shared dietary aspects. The WFKT [Keto diet] led to a greater decrease in triglycerides, but also had untoward risks from elevated LDL cholesterol and lower nutrient intakes from avoiding legumes, fruits, and whole, intact grains, as well as being less sustainable [easy to follow long-term].

If both diets are equally effective short term, the important question becomes whether they are equally successful and equally healthy long term.

As noted in the author’s conclusion, this study raised several “red flags” suggesting that the WFKD Keto diet may be less successful and less healthy long term than the Med-Plus diet. However, this was a short-term study.

So, the question becomes, “What do long-term studies show?” There are few long-term studies of low-carb diets, but the few long-term studies we do have suggest that the Mediterranean diet is a better choice for both long-term health and reduced risk of diabetes than most low-carb diets.

For more details on this study and what it means for you, read the article above.

These statements have not been evaluated by the Food and Drug Administration. This information is not intended to diagnose, treat, cure, or prevent any disease.

Health Tips From The Professor