UKPDS risk engine related to CHD/CVD

NorthernPam

Member
Messages
12
Hi everyone,
At my recent clinic appointment The DN was giving me a lecture about risk of heart attack and stroke and brought up a % on her screen to show me. Today I have been looking for information about how that risk is assessed and the nearest I can find is based on the UKPDS. Does anyone know if this is correct or not please?

If it is then I think I might have found something interesting to print off and take with me for next time. It is an external validation of the UKDPS risk engine in T2 patients written/published 2010/2011 so fairly recent. It says: -

[The UKPDS risk engine showed moderate to poor discrimination for both CHD and CVD (.....), and an overestimation of the risk (224% and 112%). The calibration of the UKPDS risk engine was slightly better for patients with type 2 diabetes who had been diagnosed with diabetes more than 10 years ago compared with patients diagnosed more recently, particularly for 4 and 5 year predicted CVD and CHD risks. Discrimination for these periods was still moderate to poor.][/quote] and [To enhance the prediction of CVD in patients with type 2 diabetes, this model should be updated.][/quote].

Full report can be found at [http://link.springer.com/content/pdf/10.1007%2Fs00125-010-1960-0.pdf][/url]
Thanks, Pam
 

phoenix

Expert
Messages
5,671
Type of diabetes
Type 1
Treatment type
Pump
The NICE guidelines suggest it is used if you are over 40
http://cks.nice.org.uk/cvd-risk-assessm ... rification

I briefly scanned the paper you mentioned and these are the reasons it suggests for the poor performance
the UKPDS risk engine was developed from a cohort that started including patients in 1977 [8]. Treatment of type 2 diabetes and prevention of CVD has improved since 1977 and the risk of developing CVD has declined with better treatment of type 2 diabetes [22]. Also, as diabetes is now detected at an earlier stage, therapeutic intervention can be initiated earlier, reducing CVD risk even further. Altogether, this is likely to explain the large differences in predicted and observed absolute risks that have led to poor calibration.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3017299/

So things may have improved. (that's good isn't it!) However, women today have at least an equal risk with men today; it was a comparatively lower risk at the time of the UKPDS.
One point is that the original study on which the engine is based was from the UK. This validation study used Dutch and German subjects, there may have been cultural and genetic differences then and now. A second is that the follow up period was only 8 years so could not validate the 10 year risks.
 

NorthernPam

Member
Messages
12
Hi Phoenix,
Thanks for the reply. Yes, I agree that in terms of identifying and treating things such as CVD/CHD the world has moved on and it’s much better now than it was when the original UKPDS took place. The point is that the risk assessment is based on that study. When later diabetes studies are compared there seems to be an opinion that UKPDS was flawed from the outset as it was based on participants keeping their bg below the arbitrary figure of 182 mg/dl (below 10 ish) set in the US which is the same target in the UK today. The 2010/2011 external validation that I referred to states that the risk has been overestimated. My understanding of this paper is that the criticism of ‘moderate to poor’ refers to the tool (software) that is used to make the assessment irrespective of outcomes in different geography. I hope I have explained that well enough.

I was 64 in May so at some stage in the next 30 years it is likely that something is going to see me off (!) and due to age alone I think my risk of all sorts of things increases. I was not unduly concerned when told that my risk of a heart attack in the next 10 years was around 19% and know that in NHS terms that will decrease anyway as I lose a further stone though I might be being too complacent here. Others are being quoted much higher figures. I am somewhat sceptical of any facts and figures they quote at me. What bothers me is that in addition to being given duff dietary advice some people are being scared witless by a risk assessment that could be 200+% out and that this appears to be based on a software tool that independent validation has deemed to need an overhaul as it produces moderate to poor results.

I am new to all this and although I have learned a lot in the last 4 months I know I have a great deal more to learn so if I am off-track with this please be patient with me.

To change the subject - I’ve seen your name on numerous forum posts and think you have been around a long time so please can I ask you a really dumb question? When you test 2 hours after meals should it be 2 hours after starting the meal or 2 hours after finishing it as the time could vary by 30 minutes or more? I did ask this question twice at my clinic review a couple of weeks ago and was met with blanks looks followed by ‘errr’ then ‘probably’ 2 hours after finishing. This is what I have been doing but am not sure it is right. Next week I have a home week and plan to do some specific meal and exercise testing and it seems daft if I don’t get the timings right so hope you can advise please.
Thanks, Pam