Posted by on Thursday 15 January 2015

6.5% - Are they mad?

I haven't really got time to write this today, but I have waited too long with too many blog posts bouncing around in my head so I'm just going to rattle this one in quickly to clear some head space.

The situation has been exacerbated in recent years because of my position as one of the patient members of the NICE Guideline Development Group (GDG) for the update to the Type 1 Diabetes in Adults Guideline. It has been an enormous privilege to be part of creating the guideline. Filtering through all the research and discussions looking for definitive, evidence-based, best-practice approaches to treating and managing type 1 diabetes has been incredibly interesting (and quite hard work!). Part of the NICE process is that members of the GDG are sworn to secrecy until the guideline is published. This is partly to promote a healthy, open dialogue within the group and also because - research being what it is - new stuff emerges all the time and blabbing about what the group thinks about a topic early on, might cause upset, especially if it changes later as new results/research are found or economic models are run.

As a consequence there are all sorts of really interesting things I have come across, or thought, or wanted to comment upon over the last 2 years which I have not been able to. Around the middle of December though, the 'consultation' version of the Guideline was published - you can read it here. The consultation document is the first glimpse the world gets of what we have been working on and it allows stakeholders to comment on the recommendations before they are finally published. So at last it feels like I can pop my head over the parapet.

One of the aspects of the consultation document that immediately prompted some comment on diabetes forums and Twitter was the recommendation regarding aiming for an HbA1c of 48mmol/mol (6.5%).

On the face of it, this sounds like madness. In a world where so many people struggle to meet the existing target of 59mmol/mol (7.5%), which fewer than 30%* of people with type 1 can manage, why on earth would you set the bar at 48mmol/mol (6.5%). Currently (and rather appropriately) achieved by just 6.5% of people with Type 1*. Why on earth would you set a target that almost 95% of people cannot reach?

Why indeed.

The section starts on page 160 and the recommendations are made on page 194.

8.1.2 Review question: In adults with type 1 diabetes, what is the optimum target HbA1c level that should be achieved to reduce the risk of complications?
39. Support adults with type 1 diabetes to achieve and maintain a target HbA1c level of 48mmol/mol (6.5%) or lower, to minimise the risk of long - term vascular complications.[new 2015]

Surely the patient reps on the group were up in arms at such lunacy?

Well actually no. If anything I was pushing for that very recommendation. And the real reason is the question. Questions are important. And without them, answers rarely make any sense.

I wasn't interested in knowing whether research said achieving 6.5% was easy. I was more interested in whether it was worth it. And even more importantly - how much it was worth it.

From the papers reviewed there was that familiar 'curvy' shape when complications were plotted against HbA1c values. At higher levels the line is really steep - zooming upwards with frightening menace. As HbA1c values fall down toward 8.5% things are curving off, though it is clear that even small improvements could potentially make a big difference. The argument that I was always given for the target of 7.5% was that below that level the added difficulty outweighed the more modest improvements of the levelling curve. But actually the improvements below 7.5% are far from level and when it comes to all the nasty things diabetes can throw at you - worth having if you can do so safely and with a decent quality of life.

The next recommendation is critical too.

40. Agree an individualised HbA1c target with each adult with type 1 diabetes, taking into account factors such as the person’s daily activities, aspirations, likelihood of complications, co-morbidities, occupation and history of hypoglycaemia. [new 2015]

So far from being a 'failure stick' to beat us with - I hope that this new recommendation (if it makes it to the published version) - becomes a source of support for anyone who has the time, skill, good fortune and technology to aim for a sub-7 A1c. And that the next one promotes a tailored programme of support, strategies and treatment options for everyone to achieve their own perfect balance between diabetes management, results and still having a life.

Mostly I hope no one gets told off again for having an HbA1c that is 'too low' by a clinic that assumes you *must* be having severe hypos all the time.

What did you make of it? Do you think it will help or hinder your efforts?

Follow up: How do you achieve these numbers?

* National Diabetes Audit 2009 - 2012


  1. 6.5 has been the highest Hba1c that I have had in the past 10+ years (the last was 5.9 or 41). I am constantly being berated that it's "too low", but really I achieve it without disabling hypos - OK I have lows, but I do feel them, doesn't everyone have lows? I can understand how some people find an Hba1c that low hard to achieve, but the other side of the coin is that I find a higher Hba1c harder to achieve!

    1. how do you achieve these numbers?


    2. Do u fidget with your basal. I do for my son. Only when there's a dire need but I do. He has had a couple of morning lows recently. But alhamdolillah I'm controlling the highs well (except when there's a birthday cake involved). His A1c Is 6.6.

  2. Thanks Patti! Yes it kinda makes sense to me that the optimised targets for A1c and BG testing should relate to each other. Staying staying 4-9 including post meals sounds more like a 5.x% HbA1c than a 7.5% HbA1c to me.

  3. Thank you ever so much Mike - yes, we should aim for much better control than at present.

    Keeping the BGs around 4 to 7 between meals and aiming to minimise post-prandial spikes is just common sense when you look at all the long-term diabetic complications.

    Despite being a physician myself, I was berated by a hospital diabetic specialist nurse for achieving this using a Dexcom CGM in my first year post diagnosis.

    Initial close control is important to maintain beta-cell mass. All my HbA1cs since presenting with DKA three years ago have been in the low forties - and my GP practice nurses think the CGM is wonderful technology :)

    I've found the main thing is getting the basal insulin right (which the CGM software makes easy), everything else then fits into place.

    And doing regular exercise - just 20 minutes here and there, but regularly. The CGM has helped avoid spikes by demonstrating the effects of timing on the rapid acting insulin boluses, and the need to avoid too much high GI food, such as white rice etc.

    Thank you for all your time with the committee work and the great journalism on the blog.

    Very best wishes,


  4. These numbers are impossible to reach...
    Or at least give me some suggestions on how to rech these levelas. Thanks.

  5. Hi Tim/Tamer

    My reply got a bit long, so I posted it as a new post.


  6. Hi Anon

    The short answer is most definitely YES. In my 3 years of pump therapy so far I have found my basal needs ongoing adjustment to keep highs and lows at bay. My meal ratios need tweaking far less often (perhaps only 3-4 times in the years I have been on the pump) but I am adjusting my basal patterns probably once every two weeks. Sometimes I just 'wing it' and make a small adjustment because I am having more high or low readings than expected with no obvious cause, but if things get really unpredictable/out of hand I spend a couple of days basal testing (see link in the sidebar). It is amazing how much this transforms the 'randomness' of my diabetes. When my basal is right, everything else just falls into place, and if it's wrong, no amount of messing with meal doses will reliably get me back on an even keel.

  7. Oh... forgot to say - MASSIVE congratulations on your son's 6.6%. That is a huge achievement of you both.

  8. It is so interesting how different we all are.
    Since I have been on a pump, plus now able to see results with the Libre, my background blood glucose levels are often a straight line. I'm sure I would be better if I didn't eat at all! But my response to food varies dramatically, I can have a spike one day but not the next for exactly the same thing - currently I am struggling with high sugars, lows after corrections then rebound highs. My standard deviation has finally avaraged just less than 3 for last year, not exactly impressive.
    Sorry, this wasn't meant to be a personal diary! Just to say that I can't imagine ever acheiving a consistent magic 6.5% HbA1c either!

  9. Hi heasandford,

    I know just what you mean about the day-to-day variations in insulin sensitivity - and it isn't always possible to work out why.

    For me, exercise usually plays a big part in explaining it - if muscle glycogen stores are depleted from exercise earlier that day, then insulin sensitivity is increased. Conversely, if I haven't exercised much then it takes more insulin to bring the BGs back down after a meal.

    A CGM with an alarm on it is really useful for breaking the cycle of lows followed by overcompensation. You catch the low earlier (set the alarm for 4.5) and then don't need to take so much dextrose.

    Have you been given the opportunity to chat to your local diabetic dietician at the hospital? Absolutely priceless, particularly if you take along examples of where things have gone wrong unexpectedly.

    Very best wishes,


  10. I had my first HBba1c result this week ( diagnosed, at age 54, with type 1 in August) and was thrilled to hear it was 43 ( 6.1%) and then dismayed when the nurse said she was worried it was too low and indicates too many hypos: in three months I've only gone as low as 4.6 a couple of times, I can feel if I go under 5.5 and always eat to bring myself up to 6.5/7.0. I eat low carb and exercise to keep in a 5.0 - 9.0 range so my low hba1 c is more to do with no high numbers than too many lows. My hospital won't send me on the DAFNE course until I have had diabetes for a year. Thanks to the internet and Dr Bernstein I have learned how to count carbs & have continued my pre-diabetes healthy eating ( ignoring the diabetes nurses telling me I must eat more starch) and regular exercise. It would really help if there wasn't a 'one size fits all' apporoach.