The model predicted Monday that the virus would kill 81,766 people in the United States over the next four months, with just under 141,000 hospital beds needed. That’s about 12,000 fewer deaths – and 121,000 fewer hospital beds – than the model estimated on Thursday.
A “massive infusion of new data” has led to the adjustments, according to model maker Dr. Christopher Murray, who is director of the University of Washington School of Medicine’s Institute for Health Metrics and Evaluation.
Additional data on the trajectory of the pandemic – in the United States and around the world – have always been awaited, as well as methodological changes to refine the forecasts. And from the start, the IHME researchers, who built the model, stressed that it was going to change.
But the new version of the model highlights how important social distance continues to be: it assumes that these measures – such as closing schools and businesses – will continue until August, and it still predicts tens of thousands of deceased.
Although the analysis was cited several times by Dr. Deborah Birx, White House coronavirus response coordinator, current administration guidelines only recommend social distancing until April 30.
The flood of new data triggered adjustments
The model essentially predicts how social distancing measures will affect the trajectory of the coronavirus in the United States. Everyone agrees that social distancing will save lives, but how quickly distancing works – and how it significantly reduces infections – is unclear.
When the model was first published, the only place that had reached its “peak” for coronavirus was Wuhan, China, according to IHME researchers. But on Monday, seven sites in Spain and Italy also appear to have peaked, providing a flood of new data to be analyzed by the model.
These regions appear to have peaked faster as a result of social distancing, the researchers said. This means that some states – such as Florida, Virginia, Louisiana and West Virginia – are now expected to reach their peaks earlier than expected, which will potentially give them less time to prepare.
Beyond infusing the model with new data, the researchers refined its methodology to better predict the spread of the virus in states that have seen few cases. And they also refined their analysis of social distancing measures after noting that some measures – such as school closings – appear to have more impact in some places than in others.
Each state, and even each state’s region, has examined the White House guidelines on non-essential travel differently, Murray said at a press conference on Monday.
Based on data on mobile phone mobility, for example, researchers have found that there are[s] in the way mandates are interpreted. “In the future, researchers plan to explore whether integrating this data will further improve forecasts,” said Murray.
A wave of hospitalization data
Early versions of the model contained little data on the plight of patients after hospitalization in the United States, but the version released Monday includes more detailed information now available from state governments.
Researchers have reviewed more than 16,000 hospital admissions, for example, and nearly 3,000 deaths related to Covid-19. They then estimated that fewer hospital resources – such as the total number of beds, intensive care beds and ventilators – will be required during the peak of the virus.
The length of hospital stay has also changed: intensive care patients should now stay longer than expected, while those with less severe cases should have shorter stays.
On Thursday, for example, the model predicted that patients requiring intensive care would only stay in hospital eight days before discharge. Now they should be hospitalized for 20 people. But patients who didn’t need intensive care originally needed 15 days, compared to just over a week now.
As more data become available, these estimates – like all projections in the model – will change. Most importantly, they are based on the continuing assumption that social distancing measures will continue for months and will be implemented in places that have not yet done so.
According to Murray, the creator of the model, the consequences could be disastrous if social distancing measures are relaxed or ignored: “The United States will see more deaths, the death toll will be later, the burden on hospitals will be much larger and the economic costs will continue to grow. “
New York Governor Andrew Cuomo, whose state is at the epicenter of the American epidemic, said on Saturday that the situation was “turning around” and “the rate of infection is going down.” This is consistent with the IHME model, which predicts that the state will reach “peak resource use” – day hospitals will be the closest – on Wednesday.
But while New York cases have generally followed IHME forecasts so far, Cuomo said there was a “danger” in being “too confident”. Other entities have made this mistake, he said, “and we are not going to make this mistake.”