DiabetesNewsletterPublic Health

Pre-diabetes diet advice

Executive Summary

* This post was inspired by two events. First, an email containing diet advice given to someone recently diagnosed with pre-diabetes. Second, the latest document from Diabetes UK reporting that there are now £5m people living with diabetes in the UK.

* Type 2 diabetes can be put into remission. We have reviewed the evidence for that in previous Monday notes. The main schools of evidence are DiRECT (low-calorie diets) and VIRTA (low-carbohydrate diets).

* The diet advice given in the email was truly shocking. Starchy food should form up to one third of the meal. Aim for 5-8 portions of starchy foods a day. Avoid fat.

* I share 10 basic principles of diabetes (covering type 1 as well as type 2).

* If one understands diabetes at all, there can be only one conclusion. Diabetes is the condition of not being able to handle glucose. All carbs provide glucose. Therefore, minimise carb/glucose intake.

* Low fat diet advice is, by default, high carb diet advice. Our long-term fear of fat is driving catastrophic diet advice, which is driving obesity and type 2 diabetes.


A couple of events coincided last week. First, someone I know in the real food world emailed me to say that his wife had been diagnosed with pre-diabetes. The reason for the email was to share the diet advice that she had been given (Ref 1). The next morning, Diabetes UK reported that “The number of people living with diabetes in the UK tops 5 million for the first time” (Ref 2). (Please note Diabetes UK – on the diabetes.org web site – is the charity for diabetes in the UK. It gets confused with the diabetes.co.uk website, which is a commercial venture helping people to manage their diabetes with diet – low-carb diets especially).

When you see the UK National Health Service diet advice sheet for pre-diabetes, you will completely understand the Diabetes UK headline. The diet advice for pre-diabetes is a perfect prescription to give someone type 2 diabetes.

The Diabetes UK report

The Diabetes UK report provided new figures that 4.3 million people had been officially diagnosed with diabetes in the UK. The charity estimated that 850,000 were type 2 diabetic, but not yet diagnosed. I suspect that the number is even higher than this.

The vast majority of people with diabetes, 90%, have type 2 diabetes (T2D). I see T2D as the body’s natural response to too much carbohydrate, too often, over too long a time. The body cannot try to maintain the required glucose level of just 4 grams in the entire bloodstream when people are consuming 300-400 grams of carbohydrate on a daily basis. Every gram of glucose, beyond the four grams required, needs to be removed from the blood stream by insulin. The body can only do this so many times. I think of T2D as the body saying, "enough is enough."

The Diabetes UK report made chilling reading. "Every week, diabetes leads to 184 amputations, more than 770 strokes, 590 heart attacks and 2,300 cases of heart failure."

Type 1 diabetes is life changing, as insulin needs to be taken for life from the time of diagnosis. Both types of diabetes can be life shortening conditions, if not managed properly. Both can be substantially alleviated with diet. Type 2 can be put into remission with diet. We have reviewed evidence for this in previous Monday notes, with the UK DiRECT and the US Virta trials being the prominent fields of research (Ref 3).

The DiRECT trial has shown that T2D can be put into remission with a very low-calorie diet (liquid meal replacement). Virta has shown that T2D can be put into remission with a very low-carbohydrate diet. I favour the low-carb diet approach by a margin, as it is more achievable, more sustainable and more nutritious. I don’t think that people can stick to c. 800 calories of low-fat milk shakes for as long as would be necessary to maintain remission. Low-carb diets, on the other hand, can be enjoyed indefinitely, while providing the nutrients we need in real food, not synthetic shakes.

It is important to note that the DiRECT trial was so low in calories that it was also low in carbohydrate (below 130g/day) (Ref 4). Hence both diet approaches, which have achieved T2D remission, are low carbohydrate. Only one of the diets reports itself as such.

The diet advice

Let’s call the wife Jane. Jane had been to see her doctor in London. Following some blood tests, Jane was told that she was pre-diabetic. She is below 50kg in weight, so this was quite a surprise. Jane was sent a PDF called “Diabetes – Dietary Advice for Pre-Diabetes.” At the top of the page were the words: “This dietary advice sheet gives some general information to help you make the recommended changes to your diet. If you need more detailed advice or if you are following a special diet that makes it difficult to make these changes, please ask your doctor to refer you to a registered dietitian.”

I put “Diabetes – Dietary Advice for Pre-Diabetes” into a search engine and this link was returned (Ref 5). Even though this document is branded “Norfolk and Norwich University Hospitals”, it is identical to the one that was emailed to Jane.

The document opens with "If you have pre-diabetes your body is not controlling your blood glucose (sugar) properly resulting in your glucose being raised."

It continued "The hormone insulin controls glucose levels. If you have pre-diabetes it means that you are either not producing enough insulin or your body is not using it properly – this is called insulin resistance. You can help to overcome insulin resistance by: Losing weight; Improving your diet; Being more active."

I agree with all of that so far. It is saying that (type 2) diabetes is a condition of not being able to control blood glucose properly. Hence the solution must be to reduce the amount of glucose going into the blood – yes?

No! The diet advice was as follows. I’ve put the verbatim advice in italics and I’ve put ZH next to my comments:

– “Eat regular meals and include a portion of starchy food at each meal, such as bread, pasta, potatoes, rice or breakfast cereals. Starchy food should make up ¼ to of a main meal.” (ZH – filling up to one third of the plate with – essentially – glucose is insane. Meals should be few, rather than regular).

– “Avoid added sugar and sugary foods (see table below).” (ZH – there was no table below.)

– “Eat plenty of fruit and vegetables – all fruit contains natural sugar so spread your fruit intake through the day.” (ZH – i.e., avoid sugar, but fruit contains sugar, so consume that regularly throughout the day).

– “Aim to include regular, peas, beans and lentils – these release glucose slowly.” (ZH – you can’t handle glucose but consume foods that release glucose regularly and slowly. By releasing glucose slowly, your body will need to deal with that glucose over a longer period of time).

– “If you are overweight, cut down on fats. Use small amounts of monounsaturated fats and oils, such as olive oil and rapeseed oil. Avoid saturated fats including butter, lard, suet and ghee. Avoid fried foods, cut fat off meat and limit pastry, cakes and biscuits. Choose low-fat milk and cheese.” (ZH – fat is the one macronutrient that doesn’t impact blood glucose levels or insulin levels. Protein impacts insulin, but glucose negligibly. Carbohydrate impacts blood glucose levels and insulin substantially. This advice is thus saying don’t consume the one macronutrient that won’t impact your condition and replace it instead with the one that will.)

This final piece of advice is so full of errors, it’s difficult to know where to start. At the principle level, the ‘don’t eat fat’ advice comes from the erroneous dietary guidelines set back as far as 1977 in the US and 1983 in the UK. This was why I did my PhD – to look for evidence for restricting fat to no more than 30% of calories and saturated fat to no more than 10% of calories. There was no evidence for either of those recommendations at the time they were set (Ref 6). There would be no more evidence to justify them, if recommendations were being set today (Ref 7).

Given that there are only three macronutrients and given that protein is a fairly constant 15% of any natural diet, an instruction to consume no more than 30% of the diet in the form of fat is, by default, an instruction to consume 55% of the diet in the form of carbohydrate (Ref 8). Since the introduction of this government diet advice, there has been far higher incidence of obesity and type 2 diabetes. The connection between the high carb advice, obesity and T2D does not seem to have been made by those issuing the advice.

At the detailed level, the errors in that final point appear ignorant of the fact that every food that contains fat contains all three fats (Ref 9). Olive oil contains more monounsaturated fat than saturated (it also contains polyunsaturated fat) but it still contains more saturated fat than a typical pork chop (Ref 10). Rapeseed oil is predominantly polyunsaturated, not monounsaturated, fat (although it contains all three fats of course – saturated, mono and poly). Butter is not pure fat – it contains water and protein. Butter has more saturated than unsaturated fat, but it still has all three fats. Lard is mostly monounsaturated fat – not saturated. That’s a common error. People trying to put T2D into remission should not be eating pastry, cakes and biscuits at all. Low-fat dairy products should be called low-fat nutrient products. We need fat-soluble vitamins (A, D, E and K) and dairy products are a rich source of these (as are meat, fish and eggs). We need the whole fat version of these products to obtain the fat-soluble nutrients.

The second page was so bad it was almost funny. It started with, "There are 2 main forms of carbohydrates in our diet: sugars and starches." It split sugars into two: i) those added to foods and drinks and ii) those found naturally in milk and fruits. It said that the added sugars should be kept to a minimum and the natural sugars should be included in moderate amounts.

There are three monosaccharides (single sugars). These are glucose, fructose and galactose. All carbohydrates break down into these basic sugars. The sucrose in a can of cola is 50/50 glucose/fructose. A banana is also fairly equal in terms of glucose and fructose amounts. The body knows that it needs to deal with glucose and fructose. Trust me, it has absolutely no way of knowing where the glucose and fructose came from. A can of cola and a banana give the body the same single sugars to deal with.

Starch next – starchy foods are seen as different to sugars by this diet sheet. They’re not. They break down into sugar like every other carbohydrate. The examples given under starches ("bread, potatoes, pasta/noodles, flour, oats, rice and other grains – quinoa, couscous etc") are all mostly glucose in terms of single sugars. They will chuck more glucose into the blood stream than any other food group that could be consumed.

The diet sheet lied again "Starchy carbohydrates are an important part of your diet." Wrong – carbohydrates are not essential nutrients (Ref 11). That means we do not need to consume them, Full stop.

The closing piece of advice in relation to starchy foods was "Aim for 5-8 portions a day." The following examples were given for a portion: a medium slice of bread, a small scone, one medium jacket potato, etc. Five to eight bread slices/baked potato equivalents a day should create frequent glucose havoc in the body. The pre-diabetic will be fully diabetic in no time.

I was truly shocked by this diet sheet. I thought such high-carb advice had been quietly dropped, as even the low-fat high-carb diet fans had come to realise that diabetes is a condition of glucose, i.e., carbohydrate, intolerance.

The basics of diabetes

Since my brother’s diagnosis with type 1 diabetes when he was a teenager, I’ve been fascinated by diet and the power of sugar and insulin. For this note, I have summarised my findings about diabetes from family experience, research and conference presentations:

1) Four grams of glucose circulates in the blood of an average person (Ref 12). That’s one teaspoon.

2) Two primary hormones (released by the pancreas) are responsible for maintaining blood glucose levels. Insulin removes glucose from the bloodstream. Glucagon is able to put glucose back into the blood stream.

3) Diabetes is essentially the inability to manage glucose. Glucagon still works in diabetics. Insulin is the impaired hormone.

4) Type 1 diabetics don’t produce insulin and need to inject insulin to survive. Type 2 diabetics produce insulin, but this mechanism doesn’t operate well and they have difficulty removing glucose from the blood stream (and moving fuel into cells).

5) If type 1 diabetes is poorly controlled, over time type 2 diabetes can be developed on top. This is known as double diabetes (Ref 13). The type 1 diabetic can become insulin resistant and the injected insulin has a reduced effect over time. This requires more and more insulin to be injected and a vicious spiral of decline ensues.

6) Blood glucose levels in excess of four grams are toxic to the body. Prolonged elevated blood glucose causes damage to cells and organs. The extremes tend to be affected first. Uncontrolled diabetes can cause eye damage and foot amputation, as serious examples.

7) The type 1 diabetes challenge is as follows: in the short term, the most dangerous scenario is critically low blood glucose (a hypo). This could induce coma and even death. In the long term, the most dangerous scenario is persistently high glucose (hyperglycemia). This can damage cells and organs as above and shorten life.

8) Type 1 diabetics (including my brother, following years of terrible dietary advice given by dieticians) invariably think that low blood glucose is caused by insufficient carbohydrate. This is wrong and the belief causes dangerous behaviour. There is only one thing that can induce a hypo in a type 1 diabetic and that’s insulin (Ref 14). A type 1 diabetic might not consume sufficient carbohydrate for the insulin taken. However, the ‘but for’ is the insulin injected. I once received a backlash from mums of type 1 diabetics on twitter for saying that type 1 diabetics don’t need carbohydrates. “They literally save my child’s life”, they raged. They are needed when too much insulin is administered. The ‘but for’ is the insulin injected.

9) All carbohydrates put glucose into the bloodstream. The summary in this post may be useful (Ref 15). The three disaccharide’s (two-sugars) all contain glucose. Sucrose (table sugar) is one molecule of glucose and one of fructose. Lactose (milk sugar) is one molecule of glucose and one of galactose. Maltose (the least well-known sugar) is two molecules of glucose. I repeat, all carbohydrates put glucose into the bloodstream.

10) Knowing that diabetes is the inability to manage glucose, and knowing that all carbohydrates put glucose into the bloodstream, why would a diabetic consume carbohydrate?

The diet for diabetes and pre-diabetes

The diet advice should simply have been “Diabetes is the inability to manage glucose. Carbohydrates provide glucose. Cut your carbohydrate intake dramatically. Base your diet on meat, fish, eggs and dairy products and green vegetables (which happen to be the most nutritious foods) (Ref 16). You should also be able to consume low-sugar fruits (e.g., berries), high cocoa-content dark chocolate and possibly legumes on occasions.”

Additional advice should have been – wear a continuous glucose monitor for two weeks. Test a variety of foods – especially starchy foods and fruit – and you will quickly see for yourself that carbohydrates must be minimised. Finally, eat no more than three times a day.

This diet sheet that was given to Jane should be a piece of evidence in a criminal trial for manslaughter. I jest not.


Ref 1: https://www.nnuh.nhs.uk/publication/download/diabetes-dietary-advice-for-prediabetes-v5/
Ref 2: https://www.diabetes.org.uk/about_us/news/number-people-living-diabetes-uk-tops-5-million-first-time
Ref 3: https://www.zoeharcombe.com/?s=type+2+diabetes
Ref 4: https://www.zoeharcombe.com/2020/09/soups-shakes-for-type-2-diabetes/
Ref 5: https://www.nnuh.nhs.uk/publication/download/diabetes-dietary-advice-for-prediabetes-v5/
Ref 6: Harcombe et al. Evidence from prospective cohort studies did not support the introduction of dietary fat guidelines in 1977 and 1983: a systematic review. Br J Sports Med. 2016.
Harcombe et al. Evidence from randomised controlled trials did not support the introduction of dietary fat guidelines in 1977 and 1983: a systematic review and meta-analysis. Open Heart. 2015.
Ref 7: Harcombe et al. Evidence from prospective cohort studies does not support current dietary fat guidelines: A systematic review and meta-analysis. Br J Sports Med. 2016.
Harcombe et al. Evidence from randomised controlled trials does not support current dietary fat guidelines: a systematic review and meta-analysis. Open Heart. 2016.
Harcombe Z. Dietary fat guidelines have no evidence base: where next for public health nutritional advice? Br J Sports Med. 2016.
Ref 8: https://www.zoeharcombe.com/2019/03/the-all-party-parliamentary-group-for-type-2-diabetes/
Ref 9: https://www.zoeharcombe.com/2018/01/saturated-fat/
Ref 10: Olive oil: http://nutritiondata.self.com/facts/fats-and-oils/509/2
Pork chop: http://nutritiondata.self.com/facts/pork-products/10298/0
Ref 11: The Panel on Macronutrients. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids (Macronutrients): The National Academies Press, 2005. (See p275)
Ref 12: David H. Wasserman. Four grams of glucose. Am J Physiol Endocrinol Metab. 2009. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2636990/
Ref 13: Cleland et al. Insulin resistance in type 1 diabetes: what is ‘double diabetes’ and what are the risks? Diabetologia. 2013. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3671104/
Ref 14: We should briefly mention alcohol. Alcohol impairs the working of glucagon – the hormone that works to naturally elevate blood glucose levels. Some might argue that alcohol can also cause a hypo in type 1 diabetics. While alcohol can impair glucagon raising blood glucose levels, the ‘but for’ will still be the insulin that lowered them.
This post has more https://www.zoeharcombe.com/2019/07/how-alcohol-affects-weight/
Ref 15: https://www.zoeharcombe.com/2022/05/why-eat-carbohydrate/
Ref 16: https://www.zoeharcombe.com/2021/08/what-should-we-eat/

19 thoughts on “Pre-diabetes diet advice

  • The thing with medical negligence is that you can’t be (legally) negligent even when wrong/misguided if you follow standard of care, so there is a lot of value to collective stupidity in liability terms. You would have to sue the guidance setters, and they seem to be in a similar huddle of moronhood

    • So with respect to HRT, the doctors prescribing the drugs were off the hook, so long as they did so according to established practices of the time. But God help you if you color outside the lines of standard “agreement/practice,” sometimes even if you’re right. As for the people responsible for setting the standard of care, they are hopelessly biased against any change in the paradigm.

    • Hi AJ
      The issue is too – what’s 1/4 by plate arrangement? The body registers calories. Half the plate as non starchy veg is mostly carb (with a trace of protein) but it’s not many calories. The starchy quarter will form most of the calories given that the ‘protein’ will need to be lean. So stupid to have protein as a word anyway, given protein is in everything.
      Check out this issue with the UK plate on volume/calories – it changes things – a lot!
      Best wishes – Zoe

      • That’s an excellent point/distinction. Especially with starchy carbs, it wouldn’t take too much to ramp of the Calories, especially were butter, mayo, or olive oil, cheese, et cetera to be added. Even something like plain rice, could conceivably cover 1/4 of your plate and contribute hundreds of Calories. I mean, it’s a 2D design, after all. How high do you pile it? How much space is between each thing, if any? Also, how big is the plate? 10”? Even so, does it have a big, deep lip on it, where no food usually goes? Or is lipless, where you can easily pile the food out to the farthest edges? Such things could change the overall Calorie count tremendously. I mean, you could put a big, 3” tall piece of quiche that covers a full 1/4 of the plate, then scoop a big mound of scalloped potatoes potatoes to cover another 1/4, and it really matters little how high you pile the other of half with roasted asparagus or a lightly-dressed arugula+spring mix salad in terms of Calorie count. And even if they add guidelines to the plates size, height, weight, etc., I doubt few would pay any attention beyond the 2D graphic.

  • Hi Zoë,

    I wish I could say that I was shocked by this advice, but it is pretty much what I’ve heard that they’ve been dispensing for decades now. It really should be considered the equivalent of medical malpractice IMO.

    It’s interesting to note that in James DiNicolantonio’s The Obesity Fix, which I just finished, he states that the amount of protein as a % of kilocalories that we consumed used to be closer to 20% before these dietary guidelines came out, and now it’s closer to 15%. This would seem to be a natural consequence of the low-fat high-carb dogma being dispensed. Considering that carbs are easy to overeat, especially when you add fat to them, while protein is highly satiating, this clearly aided weight gain to some degree, which we know is an (in)dependent risk factor for T2DM. By demonizing red meat and telling everyone to eat lots of bread and pasta, it would seem that the dominoes were carefully lined up to create a death spiral of diabetes. And now those chickens have come home to roost.

    It’s hard to reconcile the fact that the organizations that are ostensibly leading the charge in the fight against diabetes and other chronic health diseases seem to have a terrible habit of dispensing advice that is completely at odds not only with established medical science, but also with common sense. “Eat a bunch starchy foods on every plate…many times a day….” Are you bleeping kidding me?!

    And then there’s the contradictory sugar advice with respect to fruit! Fruit has fiber, which helps some, but it’s not something you want to be gobbling if you are diabetic or pre-diabetic. And if you are going to eat fruit, it’s not as if they are all created equal. Berries are one thing, grapes are another, for instance. For that matter, not all starchy foods are created equal, either. Parboiled rice, for instance, usually elicits much less of a BS or insulin response than ordinary rice. And one variety of potato can have a dramatically different effect than another. Et cetera, et cetera.

    Sorry for blathering. Stuff like this makes my blood boil.

    All best,


    • Hi AJ
      You can probably tell this one got to me too! I really did think they had quietly dropped this dreadful non-sense. I heard on good authority that Diabetes organisations had known for many years that they needed to change advice but they were worried about the consequences if they effectively owned up to the fact that they had harmed people.
      So keep quiet and keep harming them! :-(
      Best wishes – Zoe

      • Hi Zoë,

        I think few people want to end up with egg on their face. “We really thought we were doing a good a job, but it turns out that we really screwed the pooch. Oops, so sorry about that. Hope none of you died as a a result.” Yeah, I somehow doubt that would go over well.


      • One more thing:

        I assume you know who Dr. Neal Barnard is. If you’ve watched any vegan propaganda film published in the last decade, he invariably makes at least one appearance. He is very much of the opinion that fat—and very specifically animal fat—is what causes diabetes. How would you reply to him?


        • Yes I do!
          I’m trying to work out how I can add the one image I would use to reply to this! It’s one of Nina’s…

        • I wonder how he deals with the fact that when type 2 diabetic patients reduce their carbs and increase their fats, sugar control, blood pressure, hdl and triglycerides all improve, even before any weight loss has happened? Many low carb critics try to argue that any ‘benefits’ from low carb diets arise from the weight loss that could happen in any diet, rather than anything intrinsic in the diet itself, but that would appear not to be the case.
          Also, how does he explain Professor Lustig’s findings, in this?https://pubmed.ncbi.nlm.nih.gov/26499447/
          in which he restricted the fructose intake alone in obese teenagers with metabolic syndrome, improving all the parameters including insulin resistance and glucose control. He deliberately replaced the fructose with starch so that the participants lost no weight. He also made no changes to the fat intake. If Dr Neal Barnard’s assertion that insulin resistance, and thus T2DM was caused by saturated fat (particularly animal fat) intake, how does he explain the improvement in all the parameters of metabolic syndrome with no reduction in saturated fat intake?
          Lustig also points out that one of the hallmarks of insulin resistance, a fatty liver, can arise from metabolism of excess fructose, and exactly the same changes happen with metabolism of excess alcohol. It would make a very plausible explanation for the strong association of type 2 diabetes with alcohol abuse, and why there is a very weak association between red and processed meat and type 2 diabetes – lots of people drink beer and wine at barbecues!

  • Excellent, as always Zoë.
    I used to tell my patients with type 2 diabetes that the years of carbohydrate had made their bodies deaf to insulin, so the pancreas was shouting so loud soon it would lose its voice, and the best way to save the pancreas was to cut out the cause of the deafness, the carbs, and then the pancreas could talk normally again.
    I get the feeling from posting on DrsNet that the number of low carb converts is slowly rising.
    Interestingly, the worst criticism of Dr David Unwin’s data on diabetes remission using low carb diets that I’ve heard from his doubters was that the patients hadn’t been randomised, or blinded. I think that had David set out at the time to do a randomised trial of low carbs, it wouldn’t have got past the ethics committee because it differed do markedly from standard dietary advice. It would be impossible to ‘blind’ a dietary study, because you can’t easily disguise fats and carbohydrates. One thing you can be sure of though, more patients will end up blinded on standard dietary advice than on low carb advice.

    • Hi Meddyg
      Thank you! This one shocked me more than most – you can tell! I got angry too. I’ve waved at the NHS trust and NHS generally on twitter and radio silence replies of course. I’ll not let it drop.
      The criticism of Virta was the same – you didn’t randomise – you asked people which diet they wanted to do – the intervention or the control! And in the real world – would people not be asked if they wanted to try a diet solution for their diabetes?
      It’s clutching at straws in my view. We need a law suit!
      Best wishes – Zoe

  • I’m not a huge fan of CGMs, but imagine how quickly this kind of rubbish advice could be dismissed if everyone was issued a CGM immediately upon diagnosis with T2D or prediabetes. It would be pretty easy to see what happens if you have one of those conditions and you go out of your way to have starches make up 1/3 of your meal.

    • You only need them for 2 weeks – less if you try lots and learn fast. It’s so obvious!
      This was what finally got to my brother – after years of not understanding that porridge and oat cakes were not good for him!

  • The mind absolutely boggles.

    “If you have prediabetes your body is not controlling your blood glucose (sugar) properly resulting in your blood glucose being raised.”

    My comments in parentheses:
    Therefore: “Include a portion of starchy food (which raises blood glucose) at each meal. Starchy food (which raises blood glucose) should make up 14 to 1/3 of a main meal.”

    The cognitive dissonance is stunning. It’s as if these “experts” have NO IDEA that you can live a perfectly happy (and healthy!) life with *no* starchy foods at *any* meals.

    I honestly wonder how long it will be before there are some class action lawsuits for “guidance” like this. How do these people sleep at night?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.