That Guardian report is about a meta-analysis.
Meta-analysis is what I do for a living (among other things).
There's no adjustment for age, no quality assessment (to identify which reports are more reliable than others). There are 2 studies that are huge outliers (for mortality, probably some of the other outcomes, too) so skew the results. It's a huge effort I respect but it's not a very informative picture, yet.
It looks like the obesity risk in the meta-analysis was based on who went to hospital or even who was in ICU; not based on all obese people in the community who get covid; or maybe it's a mix of both those who died at home & died at hospital. We don't know. The authors don't tell us.
From what I can work out (say, from ICNARC reports): age is like 3-6x more important than obesity as a risk factor. Obesity is exciting because it's something you can try to control, but it takes a long time to change, anyway.
I don't like obesity being called a 'major' risk factor when we don't know how much that risk is modified by age (don't know what the ages were of the patients in each primary study).