I can see that, from the title of the second reference I gave, you might think that I am conflating the two. However, the section that deals with blood type is fairly clear, including -
"Blood type seems to be a predictor of how susceptible a person is to contracting SARS-CoV-2, though scientists haven't found a link between blood type per se and severity of disease.
Jiao Zhao, of The Southern University of Science and Technology, Shenzhen, and colleagues looked at blood types of 2,173 patients with COVID-19 in three hospitals in Wuhan, China, as well as blood types of more than 23,000 non-COVID-19 individuals in Wuhan and Shenzhen. They found that individuals with blood types in the A group (A-positive, A-negative and AB-positive, AB-negative) were at a higher risk of contracting the disease compared with non-A-group types. People with O blood types (O-negative and O-positive) had a lower risk of getting the infection compared with non-O blood types, the scientists wrote in the preprint database medRxiv on March 27; the study has yet to be reviewed by peers in the field.
In a more recent study of blood type and COVID-19, published online April 11 to medRxiv, scientists looked at 1,559 people tested for SARS-CoV-2 at New York Presbyterian hospital; of those, 682 tested positive. Individuals with A blood types (A-positive and A-negative) were 33% more likely to test positive than other blood types and both O-negative and O-positive blood types were less likely to test positive than other blood groups. (There's a 95% chance that the increase in risk ranges from 7% to 67% more likely.) Though only 68 individuals with an AB blood type were included, the results showed this group was also less likely than others to test positive for COVID-19."
This is a review article and so the full details of the experimental design are not given. The hospital cases, it could be argued, will represent more serious cases and so both susceptibility to infection and susceptibility to severe disease may be factors. However, on the face of it, the 23,000 sample tests
appear to be based on antibody testing in the community which would link it to susceptibility to infection. It is likely, if my suggestion is correct about cell surfaces having different levels of resistance to viral attachment, that both types of susceptibility could be linked. There are, of course, other factors relating to both types of susceptibility.
The Scottish figures that I started with, to me at any rate, suggest a much lower level of infectivity than expected, which needs an explanation.