CrossFit | A Food-Based, Low-Energy, Low-Carbohydrate Diet for People With Type 2 Diabetes in Primary Care

A Food-Based, Low-Energy, Low-Carbohydrate Diet for People With Type 2 Diabetes in Primary Care

ByCrossFitJune 25, 2020

Question: Can a simple form of dietary restriction, without the use of artificial diets or supplements, rapidly improve glycemic control in diabetics?

Takeaway: This 2019 pilot study found a low-carb, low-calorie diet led to significant weight loss and improvements in glycemic control in diabetic subjects over a 12-week period. These results are consistent with growing evidence indicating diet, unlike standard-of-care methodologies, can reduce the severity of, and even reverse, diabetes.

Growing evidence suggests diabetes is not a lifelong, progressive condition and dietary changes alone can lead to diabetes control and remission (1). As previously discussed on, low-carbohydrate diets are a specifically promising intervention and have been shown in multiple trials to improve glycemic control, in some cases leading to outright remission (2). Previous research has found a variety of carbohydrate-restricting regimens to be effective, with those involving greater reductions in absolute carbohydrate intake — i.e., a ketogenic diet or a combination of calorie and carbohydrate restriction — leading to greater improvements in glycemic control (3). Most trials involving low-calorie, moderately carb-restricted diets have included artificial diets (e.g., shakes and supplements), which limits their generalizability. This 2019 pilot study specifically tested the impact of a food-based, low-calorie, low-carbohydrate diet on diabetes.

Researchers recruited 33 subjects from primary care practices in Oxfordshire, U.K. All subjects were diabetics with a BMI > 30. Randomization was 2:1, with 21 subjects randomized to the intervention group and 12 to control.

Control subjects maintained usual care (receiving standard dietary advice in line with U.K. guidelines) for 12 weeks. Intervention group subjects were instructed to limit calories to 800-1,000 per day, with less than 26% coming from carbohydrate and a minimum of 60 grams of daily protein (4). To simulate a broadly generalizable program, dietary education and instruction was minimal: four 15-20 minute sessions with a nurse practitioner during weeks two, four, and eight. After following this diet for eight weeks, subjects increased calorie intake to a level that led to weight stability (5).

The intervention group lost an average of 9.5 kilograms over 12 weeks (compared to 2 kilograms in controls). Intervention group subjects improved across a variety of measures of glycemic control including HbA1c (63.2 – 46.9 mmol/mol, 25%), fasting glucose (10.0-7.2 mmol/L, 18%) and fasting insulin (95.1 – 72.9 mmol/L, 23%). All changes were significantly different from both baseline and control; no significant improvements in glycemic control were observed in control group subjects.

Intervention group subjects were on an average of 1.4 anti-diabetes medications and 1.4 antihypertensive medications prior to enrollment. Seven subjects (33%) were able to discontinue at least one medication in each category over 12 weeks. No subjects in the control group discontinued any medication. Notably, subjects in the intervention group cited the desire to reduce or cease medication use as a significant motivating factor for continued compliance. No subjects reported side effects or negative events.

The success of this trial is consistent with a growing body of research indicating non-pharmaceutical, non-surgical interventions can reduce the severity of, or even reverse, diabetes. Given the inability of standard-of-care practices (medication, general dietary advice, surgery) to sustainably reverse diabetes without major side effects, additional research is warranted. Researchers should test the impact of various interventions that improve insulin sensitivity and glycemic control — including carbohydrate restriction and exercise — to broadly and sustainably reverse diabetes.


  1. Type 2 diabetes: Treating not managing; Primary-care led weight management for remission of type 2 diabetes (DiRECT): An open-label, cluster-randomized trial; Doctor referral of overweight people to low energy total diet replacement treatment (DROPLET): Pragmatic randomized controlled trial; James Lind Alliance Priority Setting partnerships: Diabetes (Type 2) Top 10
  2. Adapting diabetes medication for low carbohydrate management of type 2 diabetes: A practical guide; Low carbohydrate diet to achieve weight loss and improve HbA1c in type 2 diabetes and pre‐diabetes: Experience from one general practice
  3. Dietary approaches to the management of type 2 diabetes (DIAMOND): Protocol for a randomised feasibility trial
  4. Notably, subjects’ absolute carbohydrate intake — no more than 65 grams per day — is only slightly greater than that prescribed in many low-carb diets that do not involve deliberate calorie restriction (e.g., Atkins, ketogenic diets).
  5. From the authors: “Dietary advice focused on excluding sugary and starchy foods high in carbohydrates entirely from the diet (with the exception of dairy and limited fruit intake), strict portion control and minimal use of fats and oils. Participants were advised to eat fresh vegetables or salad and small amounts of lean meat and fish.”

Comments on A Food-Based, Low-Energy, Low-Carbohydrate Diet for People With Type 2 Diabetes in Primary Care


Comment thread URL copied!
Back to 200626
Erin Wysmuller
June 26th, 2020 at 12:28 pm
Commented on: A Food-Based, Low-Energy, Low-Carbohydrate Diet for People With Type 2 Diabetes in Primary Care

I would have liked to see them only use the decrease in carbs as a variable rather than both decreased carbs and decreased calories. Eating 1,000 calories/day is not sustainable long term, where reduced carb intake is.

Comment URL copied!
Tyler Hass
June 26th, 2020 at 5:59 am
Commented on: A Food-Based, Low-Energy, Low-Carbohydrate Diet for People With Type 2 Diabetes in Primary Care

There’s a lot of interesting things going in this trial (in addition to what was written in the summary). First, it’s “food-based”. That sounds almost silly, but there are programs out there where people go on pre-made smoothie diets. The unfortunate thing about such a diet is that it does not teach people anything. What happens when the training wheels come off? They can't just drink smoothies for the rest of their lives. When they re-enter the real world, they will still need to learn how to properly eat. I like the idea of skipping the shakes and going straight to real food.

Secondly, the control group was advised based on the DiabetesUK handbook: “What is a healthy balanced diet for diabetes?“ It would be more accurate to call it a comparison group, unless the researchers knew in advance the standard of care would have no impact. (I think they knew it was useless.)

It’s unfortunate they didn’t report the health biomarkers at the end of the weight loss phase and at the end of the stabilization phase. This would offer some insight into whether the improved biomarkers were driven by the low carb diet or the weight loss. Bonus points if they had a third arm do the stabilization phase first, followed by the weight loss phase.

Comment URL copied!
Susan Young
June 26th, 2020 at 11:05 am

Agree. Food-based is the only sensible way to proceed.

Comment URL copied!