Last checked at 
* Estimated 89% effectiveness against symptomatic disease
* Analysis is first to study two AZ doses in real world
setting
* Pfizer shot effectiveness drops off 10 weeks after first
dose
(Adds detail)
By Alistair Smout
LONDON, May 20 (Reuters) - Two doses of the
Oxford/AstraZeneca COVID-19 vaccine are around 85% to 90%
effective against symptomatic disease, Public Health England
(PHE) said on Thursday, citing an analysis of real-world data
from the rollout of the shot.
Britain has suffered one of the worst death tolls globally
from the pandemic, but has also had one of the fastest vaccine
rollouts, generating a lot of data about the use of the shots in
real world settings.
In a weekly surveillance report, Public Health England said
the estimated effectiveness of the AstraZeneca vaccine, invented
at the University of Oxford, was 89% compared to unvaccinated
people.
That compares to 90% estimated effectiveness against
symptomatic disease for the Pfizer/BioNTech
vaccine.
"This new data highlights the incredible impact that both
doses of the vaccine can have, with a second dose of the
Oxford/AstraZeneca vaccine providing up to 90% protection,"
vaccines minister Nadhim Zahawi said.
PHE said the analysis was the first of its kind on the
effectiveness of two doses of AstraZeneca in a real-world
setting but was not yet published in a journal, adding more data
would boost its confidence in the finding.
Britain has been rolling out the shots manufactured by
Pfizer and AstraZeneca since December and January respectively,
and in April also started rolling out Moderna's
vaccine.
PHE said there was a "small reduction in vaccine
effectiveness" from 10 weeks after the first dose of the Pfizer
shot before the second shot is given.
Britain extended the gap between doses to 12 weeks, though
Pfizer warned there was a lack of evidence of its efficacy
outside the three-week gap used in trials.
Last week, Britain cut the gap between doses down to 8 weeks
for the over 50s, aiming to give maximum protection to more
vulnerable people in light of concern about the B.1.617.2
variant first found in India.
(Reporting by Alistair Smout; Editing by Kate Holton and Giles
Elgood)
