The New York (not-worth-the) Times

February 26, 2021. I am no fan of the New York Times. The paper is NOT the paper of record. It should not be the go-to source for credible journalism. It's strong biases are notorious. It is as "full of itself" as any newspaper can be. Nonetheless, it can't be ignored, if only because so many of the elite believe it to be the gospel truth, yet also because it can provide a good exercise in critical thinking, not to mention provide some laughter -- if only it wasn't doing such serious damage.

A colleague sent the four following excerpts from the Times' February 25 vaccine news. Following those four claims will be some of my thoughts on the claims. Here goes:

1. Nursing homes deaths are way down.

They have plummeted, falling by more than 60 percent between late December and early February. The main cause is straightforward: Nursing home residents have been among the first people to be vaccinated.

This chart — from my colleagues Matthew Conlen, Sarah Mervosh and Danielle Ivory — tells the story:

It’s another sign of how powerful the vaccines are. The decline in deaths happened surprisingly fast, said Dr. Sunil Parikh, a Yale University epidemiologist. It occurred even though most nursing home residents and employees have not yet received both of their vaccine shots — and it has likely continued over the past two weeks, which are not shown in this chart.

“I’m almost at a loss for words at how amazing it is and how exciting,” said Dr. David Gifford, the chief medical officer for the American Health Care Association, which represents long-term-care facilities.

The nursing home data add to the evidence that the vaccines don’t just work in research trials — they work in the real world, too. (A new study of Israel, published yesterday in The New England Journal of Medicine, offered the same message.)

2. Another vaccine looks excellent.

The Food and Drug Administration released a report about a vaccine that it has not yet approved — from Johnson & Johnson — and the data were extremely positive. Like the two vaccines that are already being administered in the U.S. — from Moderna and Pfizer — Johnson & Johnson’s eliminated both death and hospitalization in its research trial: About 20,000 people received the vaccine in the trial, and not a single one was hospitalized with Covid-19 symptoms a month later.

“I’ll never stop being amazed at zero hospitalizations among vaccinated in study after study,” Dr. Aaron Richterman of the University of Pennsylvania wrote. “It’s astonishing.” Dr. Isaac Bogoch, an infectious disease expert, called the results “terrific.” Dr. Kavita Patel wrote: “I would definitely recommend it for myself and my patients.”

The Johnson & Johnson vaccine also substantially reduced the number of moderate and asymptomatic Covid cases. It didn’t eliminate them, but the vaccines don’t need to eliminate all Covid cases in order to end the crisis. A sharp reduction — and sharper reduction in severe cases — can eventually turn this terrible coronavirus into yet another manageable virus.

(The nursing home data help make this point, as well: The number of confirmed cases has fallen by more than 80 percent, which is even bigger than the decline in deaths.)

A key advantage of the Johnson & Johnson vaccine is that it requires only one shot, making it easier to administer than the Moderna and Pfizer vaccines, which require two. An F.D.A. committee will meet on Friday, and the agency could approve the vaccine shortly afterward.

3. But caseloads are no longer falling.

The number of new cases has stopped declining in the U.S.:

By The New York Times | Sources: Health agencies and hospitals

I don’t want to overreact to one week of data. But you can see a change in those lines. The most likely explanation is the more contagious variants of the virus, like the B.1.1.7 variant, which was first detected in Britain.

Tellingly, cases first stopped falling in much of Europe, where that variant is more widespread. On Friday, a top health official in Germany warned that country could be heading toward another “turning point,” after weeks of falling infections.

It’s a reminder that the pandemic is far from over. The variants have the potential to cause new outbreaks, especially if unvaccinated people become lax about mask wearing and social distancing.

4. And vaccinations have stalled.

This is not a good trend:

The storms of the past week are the major cause of the vaccination slowdown, having temporarily closed sites and delayed vaccine shipments. Whatever the reason, though, it will have consequences: Fewer vaccinations mean more deaths.

The biggest task facing the Biden administration over the next two months is accelerating the pace from the current 1.4 million vaccines per day to about three million per day.

*****

Okay, now you've read it, maybe drawn some of your own conclusions and done your own analysis. Here are my reactions.

Of the four claims, only the second rings true for me, where the data on the Johnson & Johnson vaccine are promising. As for the other three, not so much.

One: Nursing home deaths

The reporter makes the classic NYT mistake of jumping to conclusions, reinforced by comments from a biased collection of 'experts.' For instance, the head of the nursing home association has a clear bias in interpreting any data on deaths produced by his industry. Worse, the author uses the false logic of assigning a cause to an event that occurred simultaneously with a dramatic shift in the cure, when there is at least one important alternative explanation worth considering. Perhaps nursing home deaths might be down so much because they were up so much in 2020. True, fewer people in nursing homes are dying now and that could easily be due to the fact that the weakest, sickest and most vulnerable contributed to the huge number of deaths and now there are not so many -- if any -- of this vulnerable population left in nursing homes. This is the ‘early harvest’ theory, combined with some evidence that we had a lot of ‘postponed’ deaths from a low influenza death rate in the previous flu season.

And don’t get me started on the 'Mario Cuomo effect' for counting nursing home deaths. While 'at home' is the largest singe location category for Americans to die, nursing homes, along with hospitals and hospices, are the place where about sixty percent of Americans die. From a 2008 NCBI report, "Most Americans prefer to die at home, but less than a quarter actually do. The majority die in hospitals or nursing homes." That deaths at home count has been climbing over the last ten years, to just over thirty percent, but some of that increase is due to dying patients choosing to leave a hospital for hospice care at home.

Three: A rising caseload

There has long been a serious conflation of the false positive PCR tests that get classified as cases -- due primarily to excessive amplifications -- when the test could actually be detecting past cases and current immunity. We might just have the amplification of minute RNA remains of a past asymptomatic case. A positive test constitutes a 'case' when a case should include significant symptoms. None of the tests effectively predict that a person will show symptoms, get seriously ill, require hospitalization, need intensive care, or die.

Without the counts of symptomatic cases, hospitalizations, intensive care treatments or deaths we have a muddied if not uninterpretable or useless measure by using 'caseload'. Somewhere along the line the metrics changed and ‘flattening the curve' approaches went out the window. At the onset of the pandemic a year ago, we primarily counted hospitalizations, intensive care bed use and Covid affiliated deaths. Testing and masks were not in the public domain. The economic, financial, social, emotional and mental impact of serious lockdowns was ignored. The PCR test was not designed to test for Covid, but it was judged to be the best we had. But testing access, reliability, lab facilities, costs and financial incentives, population representativeness and reporting were problematic, making the data problematic for meaningful analysis. For all we knew by summer 2020, the virus might have been much more widespread and, in addition, millions of people may have contacted the virus, shown no symptoms and recovered. Mortality rates for the virus had to rely on unreliable tests, rare post mortem, and medical directors with instructions to -- when in doubt -- classify a death as Covid.

Despite my lack of medical credentials, my belief is that overall US mortality needs to be given closer examination in judging the impact of Covid-related deaths on overall mortality. The concept of 'excess deaths' needs to start with a clear understanding as to the expected deaths in 2020, and not simply the change from 2019 mortality rates. Besides the flu and pneumonia death counts, we need to see 2020 projected versus actual total mortality rates to get a better picture of the incremental impact of Covid. Before 2020, the United Nations projected a 2020 USA overall mortality rate of 900 deaths per million. Early figures on actual overall mortality in 2020 show 939 per million, about a 4.3% overage, about 125,000 excess deaths and probably within the standard error for the projection.

Four: Stalled vaccinations

The pandemic has led 'journalists' to use colorful, inflammatory and fear-filled terms like 'surge,' 'spiked,' and 'stalled,' more for click bait purposes than to inform the reader. What exactly constitutes a spike or a surge? How much and how fast must the data change to warrant that frightening description? As to 'stalled,' how long must the growth remain flat to earn this designation? Or is it just temporary? Or 'catching its breath' after rapid growth?

I lost faith in journalism usage of terms when they decided that data was a singular noun. Scary concepts have little operational on interpretive value. But they promote clicks, sell papers, and scare people.