قالب وردپرس درنا توس
Home / Health / State COVID-19 model cannot predict what will pass next week

State COVID-19 model cannot predict what will pass next week



Information obtained through the public record request shows that state health officials have COVID-19 models of top Arizona researchers, but they cannot predict what will happen next week.

ABC15 asked them to receive any new models from the modeling team from the Arizona Department of Health Services, which were suspended earlier this year.

ADHS responded to our request and provided six models, which can predict the forecast until November 2, 2020. These models include the following information: death prediction, recovery, infection and ICU beds.

ABC1

5 hopes to see some predictions in the coming months, but these models are not forward-looking.

Since the Governor of Arizona said at the end of a public conference on Thursday that a “storm” is about to occur, the news may have been very insightful, but did not give an explanation or his insights into the future.

The Arizona modeling team consists of dozens of researchers from all over the state, who use data and mathematics for analysis and prediction.

ABC15 asked Joshua LaBaer, ​​executive director of the Institute of Biological Design at State University (ASU), when the state government asked for the model for the last time, “Well, we have been working with the state government on a regular basis.”

“The last time we made a formal model was in the summer. Since then, we have never made a formal model again. I will say that our current model or the model we built before is really not applicable now, because the situation has changed with COVID. It’s more complicated.”

LaBaer said the team is still very active, but because they can use the same model as before, the information is difficult to predict.

He said: “The numbers in the last few days seem a bit scary to me, they are rising faster.”

There is no new model, and it is not clear if there will be a new model or when it will be seen. “The team is collecting ideas on how to build the next model. I don’t want to say that we don’t know how to do it, I just want to say it. It is much more complicated than the original model.” LaBaer said.

LaBaer knows that there are no contracts with ADHS and statewide researchers, so it is not clear whether the state is willing or even able to request it.

“I know that the government requires anyone to do anything, the way the rules are made, I think they are obliged to provide some payment, I don’t think they can ask you to do something for free, so at some point, they may need to pay We, but now, this is more of an ongoing dialogue than a formal contract.” Thorne said.

An ADHS spokesperson said that modeling is just a tool in the toolbox, and they said they are still working on other public models.

An email read: “We reviewed all models provided to us, including models proposed by federal partners and universities and other public model sources.”

However, ADHS officials stated that they are focusing on other data, “Over time, some of the modeling predictions we have seen have not been confirmed by actual data. Therefore, our current main focus is to analyze real-time data. We found these Indicators are a good indicator of potential future trends, and they have been used in our business benchmarks, education benchmarks, and other planning processes. Modeling is particularly valuable when determining the worst-case scenario to guide the planning process and resource requirements .”

To view the model, click the link below:

  • Recovered projection score, model fitting: 7/23, final data: 9/23
  • Hospital patients, updated on 9/23/20
  • Estimated cumulative deaths, model fitting: 7/23, final data: 9/23
  • It is expected to be carried out in the hospital, model fitting: 7/23, final data: 9/23
  • Projection in ICU, model fitting: 7/23, final data: 9/23
  • Estimated total number of infections, model fitting: 7/23, final data: 9/23




Source link