It's interesting to compare time-series graphs of doses/(100 persons) for various nations. The USA had an inflection point in mid-April, at about 60 doses per

It's interesting to compare time-series graphs of doses/(100 persons) for various nations. The USA had an inflection point in mid-April, at about 60 doses per 100 persons. I don't know the ratio of single-dose vaccines to double-dose vaccines, but assuming the latter to be in the majority, this suggests that the eventual coverage will be in the range of 60%. By contrast, Canada seems to have hit an inflection point in mid-June, at about 90 doses per 100 persons. In Canada, about 13 percent of persons are under 12 years of age, and the vaccines are not permitted for them. Therefore, in rough terms, the pattern so far is indicating that essentially every eligible Canadian is likely to receive the vaccine. Linear extrapolation based on vaccination rates over the past 10 days suggests that this result may be achieved in under 2 months. Naturally, linear extrapolation is a problem in a system like this, but the Canadian data are not sufficiently past the inflection point to justify nonlinear regression. The Canadian story is being played out in several other countries as well: inflection points occur somewhere above 80 doses per 100 persons. As in so many things, the USA is exceptional. Graphs, as of today, are at https://imgur.com/a/hANzuWt https://imgur.com/a/hANzuWt, based on data from covid.ourworldindata.org. (I don't want to post my website where this graph is auto-updated daily, along with case data for various nations, plus provincial data in Canada and state data in the USA, simply because it is on a server in a research group that cannot withstand hackernews-scale traffic.)