There are two papers submitted today by the group of Jonathan Valabhji. These can be downloaded from
https://www.england.nhs.uk/publication/type-1-and-type-2-diabetes-and-covid-19-related-mortality-in-england/ Others have pointed to these papers already in this thread. Here are my initial thoughts after speed reading the papers.
One paper (24 pages) is a full population study. It shows odds or hazard ratios between males and females, different age groups compared to 60 to 69 year olds, with T1 and T2 diabetes compared to no diabetes and also for other conditions (see Figure 2). It also shows the mortality rate as a function of age (see Figure 1).
Their findings were reported today, e.g in the Guardian
https://www.theguardian.com/society/2020/may/20/type-1-diabetics-type-2-coronavirus-nhs-study which focused on the difference between T1 and T2. What are the findings of the paper
1) the odds of dying for people with T1 was 3.5 times that of people without diabetes.
2) the odds of dying for people with T2 was 2.0 times that of people without diabetes.
3) for young people the overall risk is still very low. There were zero deaths with T1 reported under the age of 50. Since average age of T1s is 47 years, more than half of T1s have a very low risk of dying
This is the first study I have looked at which separates T1 and T2 (there could be other papers before) and it confirms that T1 is different from T2. The overall numbers are consistent with what was reported in other studies, like the one I mentioned at the start of this thread.
The second paper (22 pages) looks in more detail at people with diabetes. It gives mortality ratios within people with diabetes. For example for T2 (see p20, Figure 2) it shows the dependence on the latest Hb1Ac measurements
< 48 1.11
49- 53 1 (reference value)
54 -58 1. 05
59-74 1.23
75-85 1.37
> 86 1.62
Interesting to note is that very low Hb1Ac is slightly worse than 49 to 53.
The corresponding numbers for T1s have large uncertainties due to the smaller sample size, this is why there are horizontal bars through the data points (see p19, Figure 2), so it is difficult to make clear statements.
I also would like to point out this statement:
"Our observations show that risk of COVID-19 death is related to hyperglycaemia in people with either Type 1 or Type 2 diabetes risk and is proportional to the level of hyperglycaemia."