Iowa Still Hasn’t Sent Data To Create A State COVID-19 Model

There was a bit of an awkward moment in Gov. Kim Reynolds’ daily press conference today. When asked why the state hadn’t entered into a contract to develop a model for Iowa’s coronavirus spread until April 7, Reynolds pushed back by noting that there were still conversations happening before that with the University of Iowa College of Public Health, now contracted to develop the model, before the agreement was signed.

“I assume we’ve provided the data for them, but Sarah, do you want to add,” Reynolds said, turning to the deputy Iowa Department of Public Health director Sarah Reisetter.

Turns out, they haven’t yet.

“So, Ryan, we are just getting ready to provide data,” Reisetter said from the podium. “One of the things about modeling is you need a sufficient baseline of data, so that any model can actually be informative of what you’re trying to do.”

The question came from Ryan Foley of the AP, who broke the news this morning that Iowa had only recently finalized a contract to develop a predictive model for coronavirus in Iowa. How the state is modeling its own outbreak has been a frequent point of concern in recent press conferences.

As the AP reported, the contract states the university will “produce predictive models within two weeks of receiving the department’s patient data, or on another mutually agreed upon schedule.”

Back on April 3, the Des Moines Register reported that Reynolds’ chief of staff had told legislators that the state would release the state’s modeling in the next week. It’s now clear that the framework for developing that model wasn’t even set up on that date, and as of today, 36 days after Iowa’s first positive case, the state hasn’t even sent the data over to get it started.

Reisetter said that they wanted to make sure there was enough data to send over first. The number of days into the outbreak has not, of course, stopped other health agencies or universities from developing their own models.

“We have been talking to the University of Iowa in the weeks prior to when that contract was signed. We’re getting ready to finalize the data provision to them. But we wanted to make sure that we have enough data so that any modeling they did would actually be meaningful based on what’s actually happening here in Iowa,” Reisetter said.

She also noted that models can be helpful but not necessarily prescriptive of what actions the state will take.

For much of April, Reynolds and IDPH has said that the virus will hit its peak in Iowa in an ill-defined weeks to come, usually mentioned as near the end of April. How they are determining that without a model is not clear.

What’s also interesting about the current situation is how much Reynolds and Reisetter have spent the past few weeks downplaying what other models have predicted for Iowa. They were critical of the projections that the IMHE coronavirus model that the White House is utilizing, pointing out (correctly) that it didn’t take everything Iowa was doing, like school closures, into account. The model has since downgraded the severity of the pandemic’s impact on Iowa.

But it they were so concerned about what other models are predicting in Iowa, it’s curious that they still haven’t set the wheels fully in motion to developing a state-based one for Iowa.

Will we ever see what the model says, whenever it’s created?

“We will release it at some point,” Reynolds assured the public.

So, we’ll probably get it as soon as we were originally promised that the model exists.

 

by Pat Rynard
Posted 4/13/20