"Herd immunity" and "protect the vulnerable", a thread. These terms have become very divisive, and having an extreme libertarian group promote them doesn't help. But for me, they reflect the central dogma of epidemiology, which is to *target* interventions. /1

The language of the "herd" is terrible; we should say "population". But either way, this should be about the strong protecting the weak. Those of us who can get vaccinated do, then our immunity protects those who can't. /2

Will this pandemic end will immunity via natural infection? Let's hope not, but we can't bet everything on elimination, treatment and vaccination. So we need a strategy for mitigation, which is where "protect the vulnerable" comes in. /3

COVID can be terrible in all ages, but the first wave has exacted the most horrific toll on the elderly. Now is the worst time to be pessimistic about our prospects for doing better in the second wave. What could we do that doesn't involve extreme isolation of the vulnerable? /4

It doesn't look like test, trace isolate alone can contain at the population level, but it could be central in generating "chains of trust" - see https://www.gov.uk/government/publications/protecting-high-risk-individuals-as-an-approach-to-controlling-covid-19-outbreaks-4-august-2020 - e.g. offer those who want to visit relatives in care homes the regular tests they are asking for. /5

Also to use @WesPegden 's phrase, why not throw money at the care home sector? I suspect most of us would be happy for care home workers to get a pay boost to support safe working as well as clapping and the cost of this would be comparable to other schemes. /6

There are doubtless other things we can do as part of mitigation that are low-cost, or even overall beneficial - e.g. better ventilated buildings and cycling infrastructure - please comment with others. /7

So I don't have all the answers, but my big appeal here is not to be too negative about prospects for targeting - risks are so variable with age particularly that even modest success will have a significant impact on mortality. /end