for daily coronavirus deaths
...4,529 deaths in the past 24 hours...
The numbers are almost certainly undercounts.
The previous record for most daily deaths from the coronavirus was set on Jan. 12 in the United States, when 4,475 people died, according to data from Johns Hopkins University.
...
Experts say new infections in India, which had been rising steeply, may finally be slowing. But deaths have continued to rise and hospitals are still crowded with patients.
AP is correct about that last. I looked (for the first time I am shamed to say) at India's stats on Hopkins two or so days ago and was shocked to see India's cases bar graph descending.
There is something weird about this thing in India. Is it undercounting? Is the two-headed Indian variant a real thing? The last time I read on that scientists were pretty comfortable that it was the British variant responsible for the explosion. But this thing is behaving in India as if it is different. In no time flat it raced like wild fire through the country and now is burning itself out--with no help from the Indian government either. They ran out of everything, they vaccinated practically nobody.
You see that sequence? New infections dropping, deaths continuing to rise. That's the normal interval. Cases first, then hospitalizations, then death (or recovery). What is perplexing about the U.S. is that that normal interval has not held recently.
