It sounds like you have positive test results only, and not negative results. In that case, one approach would be to use a chi-square goodness-of-fit test to compare your counts to the proportions of blood types in the general population.

With the caveat that I wrote it, you might look at the race example here: https://rcompanion.org/handbook/H_03.html . You might scroll to near the bottom where the plots are also.

Yes, you can do a chi-square goodness-of-fit test as suggested by Salvatore S. Mangiafico. Another way for testing the relationship between Covid infection (Y/N) and the blood pressure is to run a logistic regression (Covid infection will be the binary response and the blood pressure is the predictor, assuming that you want to examine the effect of the blood pressure on the Covid infection). This will give you more informative results about the significance as well as to what extent the blood pressure affect the infection through the Odds, the interpretation of the odds is tricky so be careful about that.

Assuming categorical outcome (in this case covid-19 infection- Yes/No) is compared by different blood groups or categories. So, chi-square test is appropriate. Also find the links on decision tree for selecting statistical test.

https://images.app.goo.gl/wLjeEvWefjs6ptqY9 

https://images.app.goo.gl/aHWcvUwVSHHuBdvc9

Dear Dr Mohammed Sabah Hussein . See the RG link: Article Association between ABO blood groups and Rhesus antigen and ...

See the following RG link: Preprint Testing the association between blood type and COVID-19 infe...

See also the following link: https://www.karger.com/Article/FullText/515609

Also the following RG link: Article ABO Blood Group and Outcomes in Patients with COVID-19 Admit...

https://www.researchgate.net/deref/https%3A%2F%2Fwww.karger.com%2FArticle%2FFullText%2F515609

Kindly see also the following link: https://www.frontiersin.org/articles/10.3389/fcimb.2021.767771/full

Did you try simple Pearson correlation.?

You need to add individuals without COVID-19 to your sample, to have a binary outcome, then you can use chi-square or any other non parametric test

If you know the prevalence of the blood groups in the population you’re studying, you can use the goodness of fit chi-square to test if the distribution of the blood groups in your infected patients fits the known distribution of blood groups in the general population.