Coronavirus: Diabetic people offered support

Status
Not open for further replies.
trophywench said:
Hi @Jane50 - you and me both! It still depends on any number of other factor you haven't mentioned though - eg the natural colour of your skin/ethnicity, your day to day and therefore long term control of your BG, being too fat or too thin to name just 3.
Click to expand...
Blood glucose is ok sometimes, but do tend to get higher levels over 4 or 5 tests a day, think it is because it is difficult to get the insulin in without it going into a lump, I am white british, and am obese (over 30 bmi), just cant seem to get this down. blood pressure is normal sometimes, but other times it is far too high so have to get plenty of exercise
 
Well, I have now read the second of the two papers. I think I know what has been done. They have done an enormous data collection for the period from 1st Jan to 1st May. They have found out how many people there are with a diagnosis of diabetes, what their age is, what their BMI is, their latest HbA1c and a couple of other factors. They have done this using input from GP records using NHS number to identify individuals. They have then looked at death stats for the same period looking for death certificates where COVID is referred to as a contributor to the death. The NHS number provides the link to the diabetes data.

They have then entered the data into a statistical package which looks at all the data and looks for the effect of each of the factors as separate entities. It does this by calculating a hazard ratio for each factor ( HbA1c, BMI ethnicity, etc) which says something about the additional risk of being in a particular category. This hazard ratio increases with age, HbA1c, and generally with BMI although are higher for anorexic levels of BMI.

So far so good. The next thing I have to work out is what the hazard ratio means and what the risks are being compared to. I would expect the comparisons to be to people without a diabetes diagnosis but on a first read, I cannot see how that can have been done. Also, the raw data shows an expected increase in death rate with age but not for HbA1c or BMI, that only comes out in the statistical analysis. I need to get comfortable with that and the conclusions you can draw.

Might run out of steam on this one.
 
Docb said:
Well, I have now read the second of the two papers. I think I know what has been done. They have done an enormous data collection for the period from 1st Jan to 1st May. They have found out how many people there are with a diagnosis of diabetes, what their age is, what their BMI is, their latest HbA1c and a couple of other factors. They have done this using input from GP records using NHS number to identify individuals. They have then looked at death stats for the same period looking for death certificates where COVID is referred to as a contributor to the death. The NHS number provides the link to the diabetes data.

They have then entered the data into a statistical package which looks at all the data and looks for the effect of each of the factors as separate entities. It does this by calculating a hazard ratio for each factor ( HbA1c, BMI ethnicity, etc) which says something about the additional risk of being in a particular category. This hazard ratio increases with age, HbA1c, and generally with BMI although are higher for anorexic levels of BMI.

So far so good. The next thing I have to work out is what the hazard ratio means and what the risks are being compared to. I would expect the comparisons to be to people without a diabetes diagnosis but on a first read, I cannot see how that can have been done. Also, the raw data shows an expected increase in death rate with age but not for HbA1c or BMI, that only comes out in the statistical analysis. I need to get comfortable with that and the conclusions you can draw.

Might run out of steam on this one.
Click to expand...

The HR's are relative to a chosen level within each risk factor category, all referenced to the diabetic cohort. So hazard relative to being diabetic and female for sex, to diabetic and age 50-59 for age, to diabetic and BMI 25-30 for BMI etc, as independent factors. It's the standard multivariate analysis.

What I might be interested in is: What is my risk of dying if I catch COVID-19, as a 59 yr old T2D with BMI=20, HbA1c=34 etc etc. The paper won't tell you that and no 2-dimensional paper could, presumably. But maybe an interactive app could.
 
everydayupsanddowns said:
My understanding is that the majority of the risk for T1s was in the older group (70+). The 3.5x is a composite, not universal. Risk to T1s in their 40s is very low, and no deaths in T1 population younger than 20.
Click to expand...
Well that gives me a few years in which they can hopefully find a vaccine!!
 
Eddy Edson said:
The HR's are relative to a chosen level within each risk factor category, all referenced to the diabetic cohort. So hazard relative to being diabetic and female for sex, to diabetic and age 50-59 for age, to diabetic and BMI 25-30 for BMI etc, as independent factors. It's the standard multivariate analysis.

What I might be interested in is: What is my risk of dying if I catch COVID-19, as a 59 yr old T2D with BMI=20, HbA1c=34 etc etc. The paper won't tell you that and no 2-dimensional paper could, presumably. But maybe an interactive app could.
Click to expand...

Thanks Eddy, just about beginning to work that out and you have clarified it for me. I was sort of getting to the position that the analysis was saying something about relative risk within the categories but would be of little use in calculating specific risk under a given set of circumstances. As such the noise that is being made in the press about diabetes and covid needs to be listened to with your ear defenders on.
 
Bruce Stephens said:
https://twitter.com/x/status/1262997620865327104
Click to expand...

Thanks Bruce. Average age of 72. Not sure what the median or mode would be, but I think there were many T1 deaths in their 80s. Which is of course, incredibly sad (my Mum and Dad both turned 80 this year and are fit as fleas), but I think 80 is a pretty good innings with T1D, especially given the insulins etc available in 1940 when insulin was still a relatively newfangled therapy.
 
everydayupsanddowns said:
… especially given the insulins etc available in 1940 when insulin was still a relatively newfangled therapy.
Click to expand...

We haven't even achieved our first century yet! Bloomin good job so many folk worldwide were carnivores, eh?
 
Aye, TW, we Creonistas would be right up **** creek without pigs too.

(That’s a deliberate graphic use of the coarse word)
 
To save having to use that particular word when she absolutely felt she needed to, a girl at work always used a mixed metaphor instead, which made us all chuckle appreciatively anyway. Hence, I offer it now - Up a gum tree without a paddle!
 
Jane50 said:
Blood glucose is ok sometimes, but do tend to get higher levels over 4 or 5 tests a day, think it is because it is difficult to get the insulin in without it going into a lump, I am white british, and am obese (over 30 bmi), just cant seem to get this down. blood pressure is normal sometimes, but other times it is far too high so have to get plenty of exercise
Click to expand...

I think you need to speak to your GP @Jane50 - there certainly seem to be several boxes ticked in your case that could infer additional risk if you got the virus badly. I think you need to be as careful as you possibly can be with social distancing, and ask your GP their advice. It may be that they advise self-isolation and could give you a letter for additional support (online shopping slots etc)
 
Status
Not open for further replies.
Back
Top