When we talk about complications, we need to talk about “Risk” #TalkAboutComplications

It’s hard not to have seen the recent Twitter and blog campaign that came about as the result of Grumpy Pumper’s diagnosis of a foot ulcer linked to neuropathy. If you haven’t seen it, it is called #TalkAboutComplications and is intended to change the way people talk about and discuss complications, removing the blame game and judgemental language, and make it much more matter of fact.

Having observed the ongoing discussions on Twitter, it seems that the hashtag has encouraged people to discuss complications, on Twitter, in a way that perhaps they hadn’t previously, but it isn’t the first place that many go for those types of discussion.

Lessons for Twitter could be learned from many of the posts in Facebook groups and diabetes forums related to support of people with type 1, which  have a lot of real world discussion amongst people with diabetes (of all kinds) where the discussion is how you deal with living with complications and what the treatment options are. Most of this is also matter of fact and supportive of each other, and as it’s people with diabetes, generally helpful and non-judgemental.

One observation is that while there’s a lot of discussion about what people have and that the way we should talk needs to be supportive and non-judgemental, there’s been limited discussion of risk and managing that risk. Why is that important?

It’s important because complications are not an inevitability of living with diabetes, however, there is a strong likelihood that they will occur, and whilst, due to the fact that as a person with diabetes, you have a screwed up body, about some of which we still know very little, why wouldn’t you look to understand what that really means. And as someone that has a diagnosis of maculopathy (or in other words, something similar to retinopathy, but on the macula, which is the central part of the retina used for close and detail vision) I also have a vested interest.

Risk and Likelihood

As a person diagnosed with type 1 at a young age 30 years ago and spending my first five days of diagnosis in a hospital ward with older people with diabetes, a number of whom were there for amputations, I was obviously exposed to the risks at a very early age. I’m not going to say that was fun, and neither am I going to suggest that it was not formative. It was laid out to me at that point that it was more than likely that something nasty would happen. Back to the same set of issues that we have with Type 2, where the disease itself or the complications of the disease were always considered to be progressive. And yet, there are many people who don’t have these issues.

Why? We need to look at DCCT (Diabetes Control and Complications Trial) and EDIC (Epidemiology of Diabetes Interventions and Complications) and more recent papers looking at glucose variability to understand this better.

The key takeaways from DCCT and EDIC are:

This is also represented in the below graph which has been widely used in discussing the risk of complications:

What do these statements tell us? They tell us, very factually, based on a 25 year study of people with diabetes, that bringing Hba1C as close to non-diabetic as possible brings the risk of having complications close to those of people without diabetes. Large amounts of data have been collected and analysed to arrive at this discussion point, but it is stark in its message. A lower a1C is better in terms of long term health. That’s not to say luck won’t play a part, but there’s a very, very strong relationship.

This is also reflected in the UK’s NICE guidelines for Type 1 Diabetes, where the target Hba1C is stated as 6.5% (48 mmol/mol). There is clear, factual, evidence for this.

Perhaps more controversially, there is also growing discussion about the impact of glycaemic variability on complication risk and how variation also increases risk. There are numerous papers and discussions available on the internet which cover this, but as a starter, this article from Diabetes Care in 2015 by Hirsch looks at the arguments for and against glycamic variability having an effect on complications, and draws the following conclusions:

In the end, there is a biological rationale and enough supporting evidence to endorse the concept that GV is an important risk factor directly involved in the pathogenesis of the vascular complications of diabetes.

Dexcom certainly believes so and has started to include the concept of Glucose Variability in its reports (although we can question whether this is linked to commercial or clinical reasoning). I’m also of a similar opinion, as, although I maintain an Hba1C that’s around 5.7%, I have had much greater variation around that level over the past three months, compared to the preceding eye examination, which showed “only” background retinopathy.

But what does “Risk” mean in this context? The Beaver Dam eye study, which looked at the link between retinopathy and the onset of diabetes and hypertension over a 15 year period gives us a baseline number for population with retinopathy and no diabetes. It states that 7.8% of the 3,402 members of the population studied (aged 43-86 years) had retinopathy signs. Or in other words, you’d expect that 78 in every thousand people would have retinopathy. It also identified that retinopathy was a pre-cursive indicator for future diabetes.

The DCCT figures suggest that if you maintain an Hba1C level that is 42mmol/mol, you would have a similar incidence of retinopathy, whereas if you took 1,000 people with an Hba1C at 53mmol/mol (7%), you’d have 156 people with retinopathy.

Or in other words, you’ve got a 7.8 in a hundred chance of getting retinopathy, whether or not you have diabetes, or more importantly, you have a 92.2 in a hundred chance of not getting retinopathy, regardless of whether you have diabetes. Simply by having diabetes, stuff is just more likely to happen.

What does all this data mean?

It comes down to the point that, if we are to talk about complications, we can’t do so without stating that managing glucose levels has a direct impact on the risk of complications. This isn’t “Project Fear”. It’s hard fact. That’s not being judgemental or trying to force anything on anyone.

Those of us living with Diabetes have a much greater chance of things happening to us than those without Diabetes, and if we want to bring the chance of those happening more in line with our non-Diabetic peers, all the evidence points to managing our glucose levels in such a way that we aim for lower levels and less variation.

If you’re a T1D online, you’ll likely have heard of Dr Richard Bernstein, who’s method for living with Type 1 seems extreme. It aims for normal glucose levels, and he suggests that this is possible with predictable, low carb eating. He holds himself up as an example of what can be achieved with normal levels, by the fact that he has no “active” hyperglycaemia related complications, and was able to reverse a number that he had. I’m not suggesting that everyone wants to follow this approach, but it does suggest an alternate hypothesis.

How can we reduce risk?

As I’ve already mentioned, there’s rather a lot of evidence that demonstrates clearly what drives conditions that are considered to be complications.

It comes down to treatments that reduce long term glucose levels, and most probably variability. There’s a reason that time in range is now considered to be a gold standard in research circles. The question has become “How can we improve time in range”, and sadly, as I’ve mentioned here before, SMBG simply isn’t doing it for the majority of people. That’s why research continues into “Smart Insulin” and Artificial Pancreas systems. It’s why people like me have built our own.

As it stands at the moment, we reduce the risk of complications by reducing our overall glucose levels and our variability of glucose levels. There are a number of ways of doing this. They come down to education, diet and technology, and more realistically a combination of all three. And the evidence base suggests that the sooner Hba1C levels are reduced and maintained at a reduced level, the better the outcome with developing longer term complications.

So should we continue with the view that complications are an inevitability of life with Type 1? Or should we, when talking about complications, consider that there are things we can do to reduce those risks? And when we talk about complications with those who are newly diagnosed, should we frame the conversation in that context?

Is it wrong to have a conversation with a newly diagnosed person with type 1 about the risk of complications, what those are and what the evidence base tells us about how best to reduce those risks?

If we were to offer technological marvels, such as CGM, to all those newly diagnosed, so that they immediately understood the impact of living life on how diabetes works, would that change the way they chose to live life? Would that affect their longer term chances (and chance is an important word here)?

Is it right that the cost of the technology to inform those choices is still astronomically high, or that at diagnosis, the frequency with which SMBG glucose testing is suggested (four times daily) undermines the ability to understand clearly what is going on? Or that the NICE guidelines suggest that structured education should take place a long period after an insulin dosing regimen has been prescribed?

Should we be stratifying the “at diagnosis” or “soon after diagnosis” education that people are given based on an assessment of the Sophie Harris types?

I suspect that answering many of these questions is particularly hard. If I look back at when I was diagnosed (before the DCCT study was finished), nobody could have answered these questions. Now though, I think there is a duty to explain what diabetes does and what the evidence shows. We need to discuss the risks and how we can best mitigate them, without doing so in a blaming or judgemental way. We should also be providing technology solutions and explaining how to use them in ways that will assist with this.

There’s an old adage in relation to managing most situations:

“Communicate early; communicate often”

In relation to communicating about complications and how to mitigate the risk of them affecting those of us with diabetes, this is more relevant than ever.


  1. As always, great article!
    To me this is best Diabetes site on the web.
    Thanks for putting in the work!

  2. Excellent article on an area that can be difficult to discuss without appearing to drift into the realm of the judgemental-something I have struggled with for a while-good work, thank you

  3. Very well written, well done. I particularly support the idea that early CGM. I wish cost would come down soon to enable this

  4. I’ve been T1 for almost 20 years. I try to balance the risk of long-term complications and short-term hypos by targeting a higher level than many (115 mg/dl (6.4mmol) vs., say 95), but trying to minimize variation via a low-carb diet. A large factor in my decision to do so is the continued improvement of treatment/measurement which has allowed continued improvement in control without meaningful lifestyle changes.

    This has resulted in higher A1c’s for me (generally in the mid 6’s) with median sensor readings in the high 120’s but with low standard deviations in the 20 range and few hypos (used to be a few times per week, now more like a few times per month). I wish there was more research on the relative importance of variation versus mean/median sensor readings, since I often compare myself to people with lower medians and A1c’s but significantly higher variability and more hypos.

Leave a Reply

Your email address will not be published.