Aug 29 In my Mint mathematics column yesterday, all I really did was take a look at some numbers. Not much mathematics, tell the truth. But with so much to learn from the numbers about the spread of Covid in India, I felt it was worth pulling out at least a few matters of interest.
So do take a look: Covid numbers: A snapshot & some questions, https://www.livemint.com/opinion/columns/covid-numbers-a-snapshot-some-questions-11598583238281.html Incidentally, you will see that I made two predictions in the column. One is for mid-Sept, we'll have to wait and see if that one comes true. The other was about total deaths due to covid. I wrote that India "will likely overtake Mexico to #3 in deaths today.” (i.e. yesterday 28th). It hardly thrills me that I was right. Death count yesterday: Mexico 62,594, India 62,713. Questions? Thoughts? Anything else? Always welcome. yours, dilip --- Covid numbers: a snapshot and some questions We’re past six months into this Corona pandemic in India. “Normal” life, whatever that may mean, has resumed to varying extents: shops are open, our nearby tennis courts have not; people throng seaside promenades for walks, the police clear them off by 7pm; you can in general eat your favourite restaurant’s fare, but not on the premises. All signs that while there’s still great need for caution, you can’t keep an economy on lockdown indefinitely. Still, perhaps there’s a need for some stocktaking. In particular, since there’s so much about a pandemic that’s to do with data, what do we learn from the data, six months into this pandemic? What do the numbers tell us about how we’ve coped with it? In this column, a few random observations and reactions — purely from poring over the covid numbers. * Odisha’s Health Minister Naba Kishore Das pronounced recently that the state’s “low fatality rate [due to covid] reflects our success in managing the pandemic.” Some part of this is true: the highest death-by-Covid count on any single day in Odisha is just 16, on August 10. For the three weeks since then, it has hovered around 10. But remember that Odisha’s Chief Minister Naveen Patnaik made more or less the same point on May 28 — his state’s mortality rate, he told us, “was one of the lowest in the world.” He was right: at that time, Orissa’s number of daily deaths varied between zero and two — not 10, not 16 — and that rock-bottom level continued till the end of June. But after that, it rose steadily to 16. Why? * Still with Odisha. Actually those death counts must be compared to the case counts to determine the fatality rate, and so whether it is actually low. Doing that, we find both Odisha Ministers were right: the state’s fatality rate is low indeed. But that masks something else about Odisha’s brush with covid. The count of new cases in that state was also relatively small for a long time: in the four months from the beginning of March till the end of June, it had risen only to somewhere between 175 and 250 every day. (Compare to Maharashtra, which at the end of June was adding about 5000 cases every day — more than 20 times Odisha’s count). But suddenly on July 3, Odisha’s new case count more than doubled, to 561. Since then, there’s been a steady and relentless rise, with an occasional leap upwards thrown in. Odisha’s case count is now increasing at nearly 3000 a day. (Maharashtra is now at about 10500 a day, less than 4 times Odisha’s count). Why this steep rise? Is that answered by the fact that Odisha allowed a Jagannath Yatra to take place in late June, just days before the July 3 leap? * Delhi was the first state in the country — and for weeks, the only one — whose daily new case count seemed to have peaked. That peak was 3947 cases, on June 23. Since then, Delhi numbers have declined steadily. Does Delhi hold lessons for the rest of the country? Possibly. The fly in that particular ointment, though, is that the decline seems to have stalled, or flattened out. Since about the beginning of August, daily new cases are holding steady at between 1000 and 1500, maybe even trending slightly upwards. What’s the explanation here? * Still, no longer is Delhi alone in having apparently peaked. In recent weeks, some other states are showing signs that their new case counts have peaked as well. Take Tamil Nadu. New cases were at nearly 7000 on July 27, declined somewhat after that and the daily count has been steady at about 6000 since. That’s a month at the 6000 level. It’s a good thing the number isn’t increasing, but why isn’t it decreasing? If this is a peak, it’s a long and flat one. * Then there’s Andhra Pradesh. After a high of nearly 11,000 cases on August 9, the graph suggests the tide is turning, but slowly and not yet definitively. August 17 saw about 6800 cases, but since then there’s been an uptick, with 10,300 on August 22 and 10,000 on August 25. Is the uptick just temporary? Was the peak itself a false one? * Bihar looks like it’s over the hump. For a long time, this state looked like it had escaped the real force of the pandemic. Daily new cases were in the low three-digits as recently as the first week of July. But there was a steep rise after that — a tripling in five days, a ten-fold increase in a month. On August 11, Bihar registered just over 4000 new cases. But since then, daily numbers have dropped and are under 2000 now. * And wait, there’s Assam. It had a peak of nearly 4600 new cases on August 12, though that is a clear aberration, an outlier. That’s because on the days just before and just after August 12, Assam saw numbers like 2700-2800. Not only that, the upward trend in the weeks before, and the downward trend afterward, suggest a peak of about 3000 that day; so that 4600 was indeed an aberration. New cases have now dropped below 2000 a day, so it’s possible Assam has peaked as well. * Goa and Punjab are the only other states that are hinting at having peaked. * Telengana reported zero new cases on July 26. That’s an obvious mistake, because there were about 1500 cases on both July 25 and July 27, and it’s been broadly rising since then (nearly 2500 on August 22, an all-time high). Every other tracked number for Telengana — the deaths, the recoveries, the number tested — stayed stubbornly at zero that day. Why that, and why has it not been corrected? I have found just one news report that mentions this, and it says simply that “Telangana did not disclose its numbers that day.” Has this had any effect on overall trends? Not really. About 1500 missing cases doesn’t seem like a lot, given that Telengana was testing about 15,000 of its citizens every day at the time, and is now testing nearly 40,000 every day; given too that we are well past the 3 million mark for total cases nationwide. But it remains a mistake, a statistical aberration. Will it ever be fixed? * At the end of July, Gujarat briefly held out hope that its rise in cases had peaked. Daily numbers started declining from about 1150. But that didn’t last long. Within a few days, they started climbing again and are now back to that apparent peak. These numbers and trends should be seen in the light of a report from August 23. It said that since March, the Gujarat Health Department has distributed a homeopathic drug called Arsenicum Album-30 to 34.8 million people in the state — more than half the state’s population — as a prophylaxis against covid-19. Now the graph of Gujarat’s daily cases since March is a steady increase that hasn’t really slowed at all, that false peak notwithstanding. That itself tells the story of just how effective this prophylaxis has been. But even the report underlines this: “There is no scientific evidence that the drug works against covid-19, a fact stressed not only by medical scientists but by some homeopathic practitioners themselves.” Why did the Gujarat Government decide to hand out ineffective pills to so many people in that state? * As a whole, India is now adding about 65,000 new covid cases a day. If that sounds like a lot, remember it’s a daily increase of just over 2 percent. There’s been a long decline in this growth rate. (Through July, for example, it hovered at around 3.5 percent.) The decline means there’s a peak coming, though I can’t predict when. In the meantime, let’s remember: even at 2 percent, this remains exponential growth, and thus worrying. Besides, remember too that over 3 million Indians have been infected. Nearly 60,000 have died. These numbers put us at or near the top of world covid rankings. We are #1 in daily new cases, will likely overtake Brazil to #2 in total cases by mid-September, will likely overtake Mexico to #3 in deaths today. Yet it isn’t because of the rankings that all this should be alarming — the numbers themselves do that job. Covid has killed far more Indians than, for example, terrorism has managed to do in all our independent years. Yet the absence of any significant outcry against this carnage suggests that we have — and maybe our government has — decided that, in Donald Trump’s famous recent words, “it is what it is.” Nearly 200,000 Americans killed by covid, yet he believes that his government has fought the disease and won. Will nearly 60,000 Indians killed by covid persuade us Indians any differently? Questions, questions. — (All data from the excellent site covid19india.org). --- My book with Joy Ma: "The Deoliwallahs" Twitter: @DeathEndsFun Death Ends Fun: http://dcubed.blogspot.com -- You received this message because you are subscribed to the Google Groups "Dilip's essays" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web, visit https://groups.google.com/d/msgid/dilips-essays/CAEiMe8rMsvhQPqpD4np%3D-hcBz8PKP2sUz6gGD1kG%3DcY38OaS-Q%40mail.gmail.com.
