This is a real-world efficacy test of the Pfizer vaccine in Qatar. This data is of particular interest because at the time the study was done, about half of the cases were the B.1.351 (SA) variant. This variant looks to be the one most likely to avoid antibodies produced after vaccination. Most of the rest of the cases were the B.1.1.7 (UK) variant.
As of the end of March, about 10% of the population had been fully vaccinated. Comparing fully vaccinated to unvaccinated people for infection, the vaccine was 90% effective against the B.1.1.7 variant, and 75% effective against the B.1.351 variant. That's not as good as Pfizer's trial in South Africa, but a real world trial is a harder test, as those who are at higher risk due to frailty or high exposure are vaccinated first. 75% in a real world test against this variant is really good, considering that the AZ vaccine doesn't work at all against it.
They also looked at the ability of the vaccine to reduce severe, critical, or fatal disease. It was 100% for both variants. That is, there were no fully vaccinated people during this study who had severe/critical or fatal covid. That said, the country's official count of covid deaths is suspiciously low relative to their cases, so I wouldn't be surprised if they were undercounting total deaths.
What does this mean for long covid?
We have data showing that they prevent infections, not just symptoms. If you don't get infected at all, you can't get long covid. But it is probably still possible to get long covid if you get infected in spite of being vaccinated. I haven't seen any data on that particular issue to make any further informed commentary.
Link to study: https://www.nejm.org/doi/pdf/10.1056/NEJMc2104974