-
The ONS study may help with this, but hospital admissions are the only thing that can really provide that idea of its prevalence in the population (with a two week lag).
I'm not sure I understand why.
ONS data will lag by a week or so, but it will give a reasonably accurate measure of the prevalence of the virus (including those who are asymptomatic and unaware they even have it, and those who are symptomatic but have mild enough symptoms that they just stay at home).
The ONS study is up to 240,000 households now, so that means roughly 60,000 households being tested a week. Testing 1% of the population a month should be pretty reliable.
Hospital admissions has tended to lag infection data by 2 weeks, and the assumption is that "double vaccination plus two weeks" will mean that most people who get it again will be far more likely to get no symptoms or just very mild symptoms that are more like hayfever than the "classic" Covid symptoms (cough, temperature, anosmia - as per the gov.uk site currently). So measuring hospital admissions alone means you're going to miss asymptomatic cases and mild cases, thus giving an inaccurate representation of prevalence amongst the general population.
Anecdata but I know 15 people who've tested positive in the last three weeks and none of them have been a hospital admission.
Greenbank-
I suppose it's because it's not clear to me how herd immunity will present itself in terms of testing data.
I'm expecting to test positive for some time having had covid a few weeks back. As you say, plenty of asymptomatic folk may also test positive but have some immunity to serious illness due to previous infection / vaccine. Perhaps these people are also less infectious, preventing the onward transmission?
I dunno. All I do know is that if hospitalisations fall during a period with no restrictions in place, something pretty seismic has changed. This has not happened before. All the other data might help understanding of how this could happen, but hospitalisation remains the key metric in my view.
Scrabble
Ah, I meant "the big question is does the fall in positive tests reflect a broader reduction in real world infections".
The ONS study may help with this, but hospital admissions are the only thing that can really provide that idea of its prevalence in the population (with a two week lag).
Next ONS release is this Friday for weekending the 24th. Will be interesting to see how it tallies with the daily testing figures.