The latest computer projections from the University of Washington’s Institute for Health Metrics and Evaluation deliver a double dose of discouragement about the course of the coronavirus outbreak, especially for those in the institute’s home state.
IHME’s April 22 assessment estimates that conditions could be acceptable for Washington state to loosen its current social distancing restrictions on May 28 — which is 10 days later than the April 17 estimate. Moreover, that assessment assumes that public health officials will have adequate resources for testing patients, conducting contact tracing and isolating those who become infected — which is not assured.
The other discouraging word is that the projected U.S. death toll through Aug. 4 has been raised, from 60,308 on April 17 to 67,641 on April 22. There’s a wide interval of uncertainty to that figure: The institute said it could end up as low as 45,375 or as high as 124,120. (The actual death toll was already nearly beyond that lower bound when IHME made its projection.)
Epidemiologists at the IHME based their projected “threshold” dates for easing current social distancing policies on the date when the model suggests the rate of new daily infections will fall below 1 per million residents in a given state. They acknowledged that the dates are later, on average, than they were on April 17.
Coronavirus Live Updates: The latest COVID-19 developments in Seattle and the world of tech
“The main driver of these later predictions is the increase in reported deaths since the last release, as well as predictions of longer (and flatter) epidemic peaks for several states,” they wrote in a status report.
The latest state on the list is North Dakota, with a threshold date of July 19.
Like every computer model, IHME’s projections make a lot of assumptions about the extent of shelter-at-home orders and how closely they’re followed. “These projections could change as new data become available or different policies are implemented,” the institute said.
If some states loosen their current restrictions before the threshold date, that would presumably change the model’s assessments for the eventual number of cases and deaths — and probably not in a good way.
Some experts have criticized the IHME’s methods and its models, complaining that the projections vary too widely from week to week and may leave policymakers either confused or complacent. The institute’s director, Christopher Murray, acknowledged last week that the projections were likely to shift earlier or later depending on the data.
“As the quality and quantity of our data increase, we will offer policymakers refined views of the pandemic’s course,” Murray said.
For the record, the April 22 version of the model projects 813 total deaths in Washington state due to COVID-19 by Aug. 4, compared with 23,232 deaths in New York and 67,641 deaths in the nation. Johns Hopkins University’s coronavirus tracker reported 824,147 confirmed COVID-19 cases and 45,039 deaths in the United States as of April 21. Over the past few days, the cumulative death toll in North America has been growing by roughly 2,000 daily.
Update for 5:15 p.m. PT April 24: This report has been updated with minor changes to the projections as of April 22. As of the time of the update, the JHU coronavirus tracker reported 51,017 deaths.
