The Corner

U.S.

COVID-19 Projection Models Are Proving to Be Unreliable

A patient is wheeled from the Wyckoff Heights Medical Center to an ambulance during the coronavirus outbreak in Brooklyn, N.Y., April 7, 2020. (Stefan Jeremiah/Reuters)

To describe as stunning the collapse of a key model the government has used to alarm the nation about the catastrophic threat of the coronavirus would not do this development justice.

In a space of just six days starting April 2, two revisions (on April 5 and 8) have utterly discredited the model produced by the University of Washington’s Institute for Health Metrics and Evaluation. I wrote about the IHME’s modeling at National Review on Monday, the day after the first revision — which was dramatic, but pales in comparison to Wednesday’s reassessment. This was not immediately apparent because the latest revision (April 8) did not include a side-by-side comparison, as did the April 5 revision. Perusal of the new data, however, is staggering, as is what it says about government predictions we were hearing just days ago about the likelihood of 100,000 deaths, with as many as 240,000 a real possibility.

As I noted in my last post on this subject, by April 5, the projection of likely deaths had plunged 12 percent in just three days, 93,531 to 81,766. Understand, this projection is drawn from a range; on April 2, IHME was telling us cumulative COVID-19 deaths could reach as high as approximately 178,000. The upper range was also reduced on April 5 to about 136,000.

On April 8, the projected cumulative deaths were slashed to 60,145 (with the upper range again cut, to about 126,000). That is, in less than a week, the model proved to be off by more than 33 percent.

My use of the term “off” is intentional. There is no shortage of government spin, regurgitated by media commentators, assuring us that the drastic reductions in the projections over just a few days powerfully illustrate how well social distancing and the substantial shuttering of the economy is working. Nonsense. As Alex Berenson points out on Twitter, with an accompanying screenshot data updated by IHME on April 1, the original April 2 model explicitly “assum[ed] full social distancing through May 2020.”

The model on which the government is relying is simply unreliable. It is not that social distancing has changed the equation; it is that the equation’s fundamental assumptions are so dead wrong, they cannot remain reasonably stable for just 72 hours.

And mind you, when we observe that the government is relying on the models, we mean reliance for the purpose of making policy, including the policy of completely closing down American businesses and attempting to confine people to their homes because, it is said, no lesser measures will do. That seems worth stressing in light of this morning’s announcement that unemployment claims spiked another 6.6 million (now well over 16 million in just the past couple of weeks), to say nothing of the fact that, while the nation reels, the Senate has now chosen to go on recess, having failed, thanks to Democratic obstinacy, to enact legislation to give more relief to our fast-shrinking small-business sector.

As I detailed in the last post, the revised April 5 model was grossly wrong even in predicting conditions that would obtain on April 5 itself. It had predicted that on that day, New York, the epicenter of the crisis, would need about 24,000 hospital beds, including 6,000 ICU beds. In fact, the model was off by a third — New York had 16,479 hospitalized COVID patients, 4,376 that were in ICU.

On April 8, IHME reduced the total number of hospital beds it had predicted would be needed nationally by a remarkable 166,890 — down to 95,202 from the 262,092 it had predicted less than a week earlier (i.e., it was nearly two-thirds off). The ICU projection over that same week was cut in half: to 19,816 on April 8, down from 39,727 on April 2. The projected need for ventilators also fell by nearly half, to 16,845 from 31,782.

Because of the way the media report on skepticism about models and a desire to get reliable facts (which used to be the media’s job), I pause to stress that I am not belittling the threat of the virus, particularly to people who are especially vulnerable — the elderly and those with underlying health problems, especially respiratory problems. The question is one of balance. American lives are being shattered by the restrictions that have been put in place. The decision to do that was based on models. Those models have no credibility. They now tell us that about 61,000 may die of coronavirus this year — although, if the last few days are any indication, that number could be revised downward soon, perhaps substantially.

To compare, the CDC estimates that 61,000 people died from the flu in the extraordinarily bad 2017–2018 period. It has become fashionable to ridicule flu comparisons, but they are surely relevant, even if it is true that coronavirus is more readily transmissible and has a higher fatality rate. For this year, the CDC projects that flu deaths will range between 24,000 and 63,000, and that hospitalizations could surge as high as 730,000 (out of the 18 to 26 million people who are treated for flu, out of as many as 55 million Americans who experience flu-related illnesses). We don’t shut the country down for that.

The question of when government officials will reopen the country they have shut down for coronavirus presses, as does the question of whether some less-draconian measures than the ones in place could suffice. Hopefully, officials will have a better answer than, “Well, our models say …”

Most Popular

Elections

About That ‘Broken Algorithm’

Of all the loopy assertions made at the press conference President Trump’s legal team conducted last Thursday, the one that has been most roundly derided is the since-ousted Sidney Powell’s claim that the national popular vote was such a landslide for President Trump “that it broke the algorithm that had ... Read More
Elections

About That ‘Broken Algorithm’

Of all the loopy assertions made at the press conference President Trump’s legal team conducted last Thursday, the one that has been most roundly derided is the since-ousted Sidney Powell’s claim that the national popular vote was such a landslide for President Trump “that it broke the algorithm that had ... Read More
Elections

Election 2020’s Endgame

Today on the menu: States are certifying their votes, and Trump's legal team is in turmoil. The 2020 Election Starts to Wind Down Michigan and Pennsylvania are set to certify their election results today, and Nevada is scheduled to do the same tomorrow. Saturday night, U.S. district judge Matthew W. ... Read More
Elections

Election 2020’s Endgame

Today on the menu: States are certifying their votes, and Trump's legal team is in turmoil. The 2020 Election Starts to Wind Down Michigan and Pennsylvania are set to certify their election results today, and Nevada is scheduled to do the same tomorrow. Saturday night, U.S. district judge Matthew W. ... Read More