Looking for some perspective on statins...

[SimonP]

IMO QRisk2/3 has its own problems in terms of the size and composition (T1 & T2 lumped iirc) of diabetics, so the same comment stands as above, are you represented by the cohort who were included. With QRisk2/3 I also don't really understand why such a large factor is applied if the "diabetes" box is ticked, iirc there's no way to look at the input data to determine just exactly what causes quite such a large effect as it's based on private/protected data reported directly by surgeries and it looked like special dispensation and a sprinkling of Holy water would be required to gain access.

iirc the proportion of T2 diabetics was vastly larger than that of T1 and the centrally reported HbA1c values for both types (which are not overly recent, at least a couple of years old was the most recent data I could dig out when I was looking) indicate that on average control isn't very good at all. These are UK population level figures, not specific to the dataset, but one assumes that they should be vaguely close to one another.

It's been a while since I looked at both this and the statins question, so I'm afraid I can't remember the exact details, but I'd certainly encourage people to look at the data that backs up the calculations/recommendations in both cases (there is summary data about the inputs to QRisk2/3). I came away quite happy to ignore both regarding their "you're diabetic therefore you should do this because the box was ticked" recommendations, but YMMV, my analysis may be flawed and I may have different numbers to you.

 

[Daemonik]

NICE guidelines seem to be to offer statins to people with type 1 over the age of 40.

The qrisk3 calculator uses Cholesterol/HDL ratio, it doesn't even ask for total cholesterol level, which agrees with what some experts say that the ratio is more important than the total amount. There is also the Triglyceride to HDL ratio. These ratios seem a better predictor of risk of heart disease than the total levels. The fact that the qrisk3 calculator doesn't even ask for the total level is significant.

I would recommend calculating these two ratios and also calculating your qrisk score.
Total:HDL ratio: 3.5 optimal, <5 desirable, >5 elevated risk
Trygliceride:HDL ratio: 2 optimal, >3 elevated risk, >4 high risk, >6 very high risk

Interesting. My triglyceride to HDL ratio is <1 (1.23:1.29). I guess that's pretty good seeing as lower numbers seem better?

 

[Daemonik]

IMO QRisk2/3 has its own problems in terms of the size and composition (T1 & T2 lumped iirc) of diabetics, so the same comment stands as above, are you represented by the cohort who were included. With QRisk2/3 I also don't really understand why such a large factor is applied if the "diabetes" box is ticked, iirc there's no way to look at the input data to determine just exactly what causes quite such a large effect as it's based on private/protected data reported directly by surgeries and it looked like special dispensation and a sprinkling of Holy water would be required to gain access.

iirc the proportion of T2 diabetics was vastly larger than that of T1 and the centrally reported HbA1c values for both types (which are not overly recent, at least a couple of years old was the most recent data I could dig out when I was looking) indicate that on average control isn't very good at all. These are UK population level figures, not specific to the dataset, but one assumes that they should be vaguely close to one another.

It's been a while since I looked at both this and the statins question, so I'm afraid I can't remember the exact details, but I'd certainly encourage people to look at the data that backs up the calculations/recommendations in both cases (there is summary data about the inputs to QRisk2/3). I came away quite happy to ignore both regarding their "you're diabetic therefore you should do this because the box was ticked" recommendations, but YMMV, my analysis may be flawed and I may have different numbers to you.

Thank you for that. Bedtime reading!

 

[Daemonik]

Welcome @Daemonik 🙂 I started a thread asking about statins for Type 1s and I found the replies very helpful in making my decision about whether to take them. Although we hear about the notorious side effects, we don’t hear enough about the many, many people who have no side effects. I decided to start statins this year - and have had no side effects. They reduced my cholesterol and my LDL, and I’m also pleased they’re helping to stabilise any arterial plaque I have.

Here’s my thread:

Thread 'Type 1s - are you taking statins?'

Nov 6, 2023

I asked a question about the pros and cons of statins on the Statins thread, but I’m wondering if I’m unusual in not taking statins as a Type 1, so this poll is for Type 1s and asks about statins. Any comments or information welcome.

TIA

P.S - I’m only asking Type 1s as I’m Type 1 myself and so want to compare what other Type 1s do. Thanks 🙂

• Inka
• Replies: 55
• Forum: General Message Board

• 

.

Thank you. I've bookmarked to have a read through

 

[MarcR]

Interesting. My triglyceride to HDL ratio is <1 (1.23:1.29). I guess that's pretty good seeing as lower numbers seem better?

How about the total cholesterol to HDL ratio? (that's the one used in qrisk3)

 

[MarcR]

IMO QRisk2/3 has its own problems in terms of the size and composition (T1 & T2 lumped iirc) of diabetics, so the same comment stands as above, are you represented by the cohort who were included. With QRisk2/3 I also don't really understand why such a large factor is applied if the "diabetes" box is ticked, iirc there's no way to look at the input data to determine just exactly what causes quite such a large effect as it's based on private/protected data reported directly by surgeries and it looked like special dispensation and a sprinkling of Holy water would be required to gain access.

iirc the proportion of T2 diabetics was vastly larger than that of T1 and the centrally reported HbA1c values for both types (which are not overly recent, at least a couple of years old was the most recent data I could dig out when I was looking) indicate that on average control isn't very good at all. These are UK population level figures, not specific to the dataset, but one assumes that they should be vaguely close to one another.

It's been a while since I looked at both this and the statins question, so I'm afraid I can't remember the exact details, but I'd certainly encourage people to look at the data that backs up the calculations/recommendations in both cases (there is summary data about the inputs to QRisk2/3). I came away quite happy to ignore both regarding their "you're diabetic therefore you should do this because the box was ticked" recommendations, but YMMV, my analysis may be flawed and I may have different numbers to you.

The formula might be flawed, but its probably better to use some sort of equation than none. It seems to me better to follow some sort of algorithm when making a decision. Is there a better algorithm out there?

 

[Daemonik]

How about the total cholesterol to HDL ratio? (that's the one used in qrisk3)

3.5, so well under the recommended <5

 

[Daemonik]

So on that QRISK, the biggest single contributing factor to changing my score was.... postcode
If I changed my postcode from where I am to another nearby city, it went up 2%. If I change to a different one, it was down 1.2%
Changing my ratio, weight, BP - none of those had anything like that level of impact. Wonder if the GP would prescribe me a new house?

 

[MarcR]

So on that QRISK, the biggest single contributing factor to changing my score was.... postcode
If I changed my postcode from where I am to another nearby city, it went up 2%. If I change to a different one, it was down 1.2%
Changing my ratio, weight, BP - none of those had anything like that level of impact. Wonder if the GP would prescribe me a new house?

I think if your score is over 10% then NICE recommends taking a statin, over 5% consider taking (although they would if your over 40 with type 1).

 

[Daemonik]

I think if your score is over 10% then NICE recommends taking a statin, over 5% consider taking (although they would if your over 40 with type 1).

As it stands, under 10%, though moving 10 miles would take it over - I guess one of the factors they use in the calculation is prevalence of CVD within an area. If you move to an area with a high incidence, it increases your risk score, though on a practical level, that's nonsense (unless the location has objectively higher risks, like moving from the open countryside to a house next to a busy motorway or power station or something).
But the over 40, over 5%, Type 1 elements all apply, so I guess that's why they're pushing it.

 

[MarcR]

As it stands, under 10%, though moving 10 miles would take it over - I guess one of the factors they use in the calculation is prevalence of CVD within an area. If you move to an area with a high incidence, it increases your risk score, though on a practical level, that's nonsense (unless the location has objectively higher risks, like moving from the open countryside to a house next to a busy motorway or power station or something).
But the over 40, over 5%, Type 1 elements all apply, so I guess that's why they're pushing it.

I asked Google Gemini about the postcode thing, it says it's to do with social deprivation:

Why deprivation matters for CVD risk:

Lifestyle factors: More deprived areas often have poorer access to healthy food options (food deserts), fewer safe spaces for physical activity, and higher rates of smoking and unhealthy diets. These factors directly contribute to higher CVD risk.
Environmental factors: People in deprived areas may be exposed to higher levels of air pollution, poorer housing conditions, and less green space, all of which can negatively affect cardiovascular health.
Access to healthcare: While the NHS aims for universal access, there can still be inequalities in the quality and accessibility of primary care, preventative services, and specialist care in more deprived areas.
Stress and psychosocial factors: Living in deprived conditions can lead to chronic stress, financial insecurity, and a lack of social support, which are also linked to increased CVD risk.
Early life development: Disadvantage experienced in early childhood can have long-lasting effects on health, increasing the risk of chronic conditions like CVD later in life.

 

[Daemonik]

I asked Google Gemini about the postcode thing, it says it's to do with social deprivation:

Why deprivation matters for CVD risk:

Lifestyle factors: More deprived areas often have poorer access to healthy food options (food deserts), fewer safe spaces for physical activity, and higher rates of smoking and unhealthy diets. These factors directly contribute to higher CVD risk.
Environmental factors: People in deprived areas may be exposed to higher levels of air pollution, poorer housing conditions, and less green space, all of which can negatively affect cardiovascular health.
Access to healthcare: While the NHS aims for universal access, there can still be inequalities in the quality and accessibility of primary care, preventative services, and specialist care in more deprived areas.
Stress and psychosocial factors: Living in deprived conditions can lead to chronic stress, financial insecurity, and a lack of social support, which are also linked to increased CVD risk.
Early life development: Disadvantage experienced in early childhood can have long-lasting effects on health, increasing the risk of chronic conditions like CVD later in life.

Seems for this to be purely on current prevalence within the area. All the above affect that, but is complicated to work those out. Basing purely on area existing CVD rate is simpler, and also explains why moving to Harrogate (affluent, green etc - the factors Gemini gives don't really apply) increases the risk score significantly. Lots of old people = lots of CVD = my risk score goes up if I move there from a more suburban area which arguably has more detrimental environmental factors.

 

[MarcR]

Seems for this to be purely on current prevalence within the area. All the above affect that, but is complicated to work those out. Basing purely on area existing CVD rate is simpler, and also explains why moving to Harrogate (affluent, green etc - the factors Gemini gives don't really apply) increases the risk score significantly. Lots of old people = lots of CVD = my risk score goes up if I move there from a more suburban area which arguably has more detrimental environmental factors.

One day AI programs will probably do a better job.

 

[rebrascora]

It also doesn't take into account how long you have spent in any particular environment or location, let alone all the other generalisations, which is why, despite my Qrisk3 being above 10, I am still resisting statins.

 

[MarcR]

It also doesn't take into account how long you have spent in any particular environment or location, let alone all the other generalisations, which is why, despite my Qrisk3 being above 10, I am still resisting statins.

It's a personal decision, but however flawed qrisk is, it provides some sort of objective measure to base a decision on. I think if I was well over 10% I would consider it.

 

[rebrascora]

My decision is not based on QRisks flaws to be honest, it is based on my own assessment of my own risk in the same way as my diabetes management is based on my own assessment. My own knowledge of my lifestyle and history and fitness and diabetes management etc. To me, only I can make those assessments because I am the only one who lives with me everyday and knows what factors I am or am not exposed to and have been over the course of my life.
What annoys me about QRisk is that each factor is generalised so you might be at the lower end of every risk factor which could all add up to quite a significant percentage reduction but that not be recognised.

 

[Burylancs]

IMO QRisk2/3 has its own problems in terms of the size and composition (T1 & T2 lumped iirc) of diabetics, so the same comment stands as above, are you represented by the cohort who were included. With QRisk2/3 I also don't really understand why such a large factor is applied if the "diabetes" box is ticked,

Because 85% of diabetics are said to die of cvd. Type 2 Diabetes in particular is treated to all intents and purposes as a sub-species of cardio/ heart disease but T1s aren't immune from it.

 

[HayleyR]

Hi @Daemonik , I'm in exactly the same position as you. Have had diabetes for 32 years and for the last few my GP suggested it may be a good idea to go on statins. Have managed to avoid it so far as I don't really want to have to take another tablet and also worried about side-effects. However when I saw the Diabetes Consultant recently, he was quite adamant that I really should start taking them now to avoid serious health issues so that put it all into perspective! Just waiting for the prescription to come through......

 

[Inka]

The risks are scary and although you can possibly reduce them by controlling your blood sugar well, Type 1 itself is a risk factor, eg:

Type 1 diabetes mellitus (T1DM) is associated with an almost threefold higher mortality than the general population [1]. Premature atherosclerosis is the main driver of this excess mortality for both men and women, with cardiovascular events occurring more than a decade earlier [2].

Cardiovascular Risk in Type 1 Diabetes Mellitus - PMC

Type 1 diabetes mellitus (T1DM) is associated with premature cardiovascular disease (CVD), but the underlying mechanisms remain poorly understood. The American Diabetes Association and the European Association for the Study of Diabetes recently ...

pmc.ncbi.nlm.nih.gov

Even if your cholesterol isn’t too bad and just slightly above the target, statins also stabilise the plaque in your arteries, reducing the risk of it breaking off. That’s a big reason why I decided to take statins.