Models project less severe toll from virus, but are stalked by unknowns
A nurse checks someone’s temperature outside the emergency room at Integris Baptist Medical Center in Oklahoma City in early April as the hospital prepares for an influx of coronavirus patients. WHITNEY BRYEN/OKLAHOMA WATCH
Oklahoma Watch is a nonprofit, nonpartisan media organization that produces in-depth and investigative stories on important issues facing the state. For more Oklahoma Watch content, go to oklahomawatch.org. COURTESY
The early models on the coronavirus spreading across Oklahoma were alarming: hundreds dead, thousands in hospitals, tens of thousands infected and an overloaded health-care system struggling to keep up.
Three weeks later, the projections remain grim and many people are bracing for a blow. Many politicians and public health experts are optimistic, however, though not without reservations. They say the limits on nonessential businesses, social gatherings and personal interactions are working but must remain at least until the end of April.
Gov. Kevin Stitt, facing pressure to relax restrictions, has taken on the role of cheerleader-in-chief. His public messages in recent days have focused on expanded COVID-19 testing capacity and assurances that hospitals and health care workers are prepared to handle the coming peak of hospitalizations and deaths. He has drawn criticism from some leaders and residents for not ordering more restrictions, such as a stay-at-home mandate for all Oklahomans, as opposed to the old and vulnerable, and not moving swiftly enough. But he has stuck to messaging that individual voluntary efforts are a key.
“We are making tremendous progress, but like I’ve said before, now is not the time to take our foot off the gas,” Stitt said at a news conference on April 15. “If you remember, I said earlier this month, what we do together over the next three weeks will greatly determine our outcome of what we can do after April 30th. That is still so true today.”
The state’s model, developed by interim State Epidemiologist Aaron Wendelboe and other public health experts, show Oklahoma’s peak will hit on April 21, with a daily toll of 22 deaths and 436 new infections. The model does not project daily or total deaths or cases beyond May 1.
That model differs in the peak’s timing from one developed by the University of Washington’s Institute for Health Metrics and Evaluation. Its latest version, released April 13, shows Oklahoma hitting its peak on May 1, with 24 new deaths that day and 470 cumulative deaths from COVID-19. The model projects further out than Oklahoma’s, with total deaths of 929 by Aug. 1. That number is lower than what the model projected a few weeks ago, when more than 1,100 deaths by August were forecast.
Public health experts caution that no model is perfect and each relies on many assumptions and ever-changing data points. They are helpful for planning but are not crystal balls. One reason is that because there’s a lag in the virus’s symptoms and their severity, multiple peaks could occur depending on the timing of cases, hospitalizations or deaths. The state’s model and the institute’s model are also limited in scope. Each looks at how hospital systems might cope at the peak, not necessarily how or where the virus is spreading outside hospital walls or in hot spots like nursing homes.
However, both models project Oklahoma will have the capacity in intensive care and other beds, as well as ventilators, to handle the peak.
“We’ve talked about how the models get better over time, like the hurricane models get better as they get closer to shore,” Stitt said. “Our models in Oklahoma are continuing to be revised and get better with more testing.”
The first iteration of the IHME model in late March showed Oklahoma needing 5,000 hospital beds at the peak, Stitt said. Later versions of the model fell to 3,000 and now 1,100.
Patrick McGough, executive director of the Oklahoma City-County Health Department, said local officials are paying attention to several models. They are looking at a peak at the end of April and a surge by mid-May.
“It’s a novel virus. People are doing their best to figure it out. Some of the best scientists and public health officials we’ve ever had are looking at this and trying to project. But they’re all models,” McGough said April 14 in a virtual town hall with U.S. Rep. Kendra Horn, D-Oklahoma City, and Oklahoma City Mayor David Holt.
Holt said his goal as mayor is to beat the models.
“Models are usually going to err on the side of a worst-case scenario or at least a less-than-ideal scenario,” Holt said. “Any time we have numbers that are better than what the models predicted, I view that as a success. That is evidence that the people of Oklahoma City are doing a good job in flattening the curve.”
Tulsa officials are expecting a local peak of new infections to hit next week, said Bruce Dart, executive director of the Tulsa Health Department. The daily hospitalization rate is expected to peak in early May.
“This time is crucial for staying at home and limiting social contacts,” Dart said at a news conference April 14.
Monica Rogers, information systems manager for the Tulsa Health Department, said the county’s modeling continues to be adjusted as new data comes in.
“The primary focus of the models is to predict infections and deaths in Tulsa County,” Rogers said in an email. “For infections we attempt to estimate all infections, including asymptomatic and mild cases that would be unlikely to be tested. This is important because in public health, we want to know all potential community spread, not just observed/positive test results, or for only those people experiencing symptoms.”
As new data was collected, Tulsa lowered its R-naught value, the epidemiological number for the reproduction rate of the virus. Rogers said that value can be as high as 3.8 for travel-related growth to less than 1 for periods when city-mandated social contact orders are in effect.
“We are currently working on models where the R0 value will increase at different points, assuming social distancing is no longer being strictly adhered to and/or executive orders are modified,” Rogers said.
Oklahoma still hasn’t hit its peak or surge. But the national conversation has shifted to the gradual reopening of coastal states that were hit earlier than the country’s interior, said Tulsa Mayor G.T. Bynum.
“We’re definitely keeping an eye on that as cities begin to ramp down their social distancing activities. But there isn’t a light-switch flip and we just all go back to February,” Bynum said at news conference April 14. “That would be extraordinarily reckless and could risk everything we’ve done for the last month. Just as we’ve ramped up restrictions, I think we’ll see a gradual ramping down of them.”
Elected officials are also feeling pressure to relax restrictions on business activity from some residents and interest groups. A few dozen vehicles converged at the State Capitol April 15, some with signs calling for businesses to be reopened. But there’s no indication that is imminent.
Stitt said Wednesday he was cautiously optimistic about a downward trend in the state’s daily hospitalization rate for COVID-19. He said that metric peaked in late March. It includes the number of positive cases in the hospital and the number of people under investigation for virus infection in the hospital.
State data shows 562 people with COVID-19 or possible COVID-19 infections were in the hospital on March 31. That daily hospitalization number had fallen to 370 by April 15.
Meanwhile, Stitt said hospitals across the state have taken financial hits with the moratorium on elective surgeries. He said the moratorium would lift on April 24. The moratorium was intended to make sure hospitals had enough personal protective equipment and beds to treat COVID-19 patients.
“Our governor solutions task force is still continuing to be proactive and preparing for anything COVID might throw our way,” Stitt said. “We will continue to monitor our hospitalizations and our PPE from around the state day and night.”