Coronavirus: The state's blueprint model could possibly cause thousands of deaths in the Bay Area in June.

Coronavirus: The state's blueprint model could possibly cause thousands of deaths in the Bay Area in June.
Coronavirus: The state's blueprint model could possibly cause thousands of deaths in the Bay Area in June.

Experts warn against reading too much in numbers

Deaths in the Bay Area from the new coronovirus will double by at least June and may increase by 10 to 36 times according to the model by the state of California to project the likely number of victims of the epidemic.

It may seem surprising that in 10 Bay Area counties, where 241 COVID-19 deaths had occurred by Friday, the state model estimated to reach 8,727 in early June. But how can that be?

Experts warn that such estimates can be quite misleading. Nevertheless, the bleak outlook on the state model provides a window into the Governor's thinking. Gavin Newsome, the first to be ordered to live in a house to control the spread of the virus, as well as Bay Area officials, issued the first shelter-in-order order at the county level. You can explain your reluctance to say when those restrictions will end.

"It gives you an idea of ​​what the potential outcomes are," said Santa Clara County Executive Jeff Smith, who has a medical degree and whose administration led the Bay Area's push for a regional asylum request in mid-2015 did. March. The state model is one of three that the county uses to guide its response. "They are all useful in the sense that they make it clear that the virus is spreading very dangerously."

But such models can be controversial if their estimates seem out of place. Critics say that this could give the authorities the answer that they consider the threat justified and undermine public confidence.

And the state's data shared with counties on April 14 shows just how dubious such estimates can be.

Professor of Epidemiology and Biostatistics at UC San Francisco, Drs. George Rutherford said after reviewing the figures, "This is far from the many, many, many sighted figures." He said model estimates for deaths in San Francisco ranged from 59 to 960, with a median of 248. There were 21 coronovirus deaths in the city as of Friday.

Rutherford said, "It's always difficult to estimate. UCSF modeling for short-term estimates for hospital days proved reliable enough to send staff to help outbreaks in New York. And in the United States Largest country, Navajo nation. Reserve "I think it's important to do things well". I think they co Beds are hard. "

The state uses an open model from Johns Hopkins University that Smith said is widely used and respected. Roger Butler, considered for the California Health and Human Services Agency, said the state "reviewed several models to inform our decision-making, but only used it."

The state model used data from 24 March and took into account the order of stay in the home and other distance and sanitation measures from 20 March.

Asked if the one-month estimates are still relevant, Butler said, "California is focusing its response on the actual case and hospital data." But he said the counties requested modeling data.

Among the parameters that had an infectious period of 2.6 to 6 days in the model-associated state was a transmission rate of 2–3 people per confirmed case, an average incubation period of 5.2 days, and a mortality rate of 1%. They also believed that 10% of those infected would be hospitalized, 32% of that group would be admitted to intensive care, and 15% of them would need a ventilator to help them breathe.

If any one of them proves to be closed, it will affect the accuracy of the projection.

"All models are based on the assumptions you make," Smith said.

Estimates of the state model have a wide range, and even hospitalization, intensive care needs, and death at the lower end.

Statewide, the median median projection for the model was that hospitals would reach 10,711 by April 12 and about 44,000.

On April 12, the actual confirmed and suspected hospitalization numbers were less than half at 5,048, and sat down at 4,929 on Friday, even below the model's low-end launch of 7,506, with trend lines suggesting they first Have reached their peak. On Friday, 1,531 cases required intensive care treatment, significantly lower than the maximum projection of 2,885 low-end.

The projected statewide death pattern for June will range from 2,521 at the high end to 39,713 at the low end. There were 1,529 COVID-19 deaths across the state as of Friday, the largest single-day jump in the state in a week.

In California, June deaths have increased by 65% ​​to allow for low-end projections.

In Santa Clara County, which had 94 deaths through Friday, the estimated total for June ranged from 117 to 1,938. For Alameda County, which had 45 deaths by Friday, the projection ranged from 119 to 1,977. For Contra Costa County, which had 22 deaths by Friday, the estimated range was 80 to 1,311.

In contrast, in Contra Costa County, the top health official said in late March that between 2,000 and 14,000 deaths could occur by the end of the year.

The state's high-level projection of Santa Clara County deaths is similar to the "best-case" forecast for San Jose officials that was released publicly in late March that June 2,000 with a shelter-in-place order in the county Will die Smith was angered by that prediction, who said it was "absurd" and would destroy public confidence because people see very few numbers, who say it is only the result of restrictions that have prevented new infections.

"It scared a lot of people," Smith said. "It gives people a false sense that things have improved, and that's not true."

Santa Clara County also uses a popular model at the University of Washington and a custom model from Stanford University, Smith said. The University of Washington has reported 1,719 total deaths in June in California, which is lower than the state's low end.

But Smith said in a recent discovery that a Santa Clara County woman who died on Feb. 6 had changed the image of COVID-19 and questioned previous forecasts.

"A large pool of viruses is ready to attack," he said, "and a large susceptible population is ready to be infected."

Post a Comment