* Caption from the state via the Sun-Times…

Illinois Dept. of Public Health death projections due to COVID-19. The gray line indicates range of deaths, which at its peak could hit between 50 and 150 deaths. The black line is reported deaths. The green line is the University of Chicago median line; the orange line is the median line projected by the University of Illinois-Urbana Champaign researchers.

Graph…

* From the Sun-Times article…

And based on the projections, the peak “should be somewhere starting now through the first week of May,” [Cameron Mock, who oversees the state’s COVID-19 projections] told the Sun-Times.

“I think it’s tough to say, but I think a couple of weeks,” Mock said of the range.

According to a projection obtained by the Sun-Times, the top of the peak could see 150 deaths a day. But during that peak period, the daily range of fatalities could fluctuate anywhere from 50 to 150. […]

Another graph Pritzker plans to show will estimate that the death rate would go up ten times if the stay at home order was lifted Friday – a little less than a week before the scheduled April 30 expiration.

The model shows deaths in the state slowly flattening into August.

* I asked the governor’s office about the article’s focus on death rates because it’s a lagging indicator of about 3-4 weeks and because the governor has said he’s looking at a wide range of things, including hospitalization, case numbers, etc. I told subscribers the response, but I think it’s important to make that public to prevent confusion…

You are correct that deaths are lagging behind infections, but we don’t have enough testing right now to have a good sense of the true infection rate, so you have the option of choosing a lagging but reliable indicator, or a current but unreliable indicator. Like the governor has said, a model is just one tool he looks at.

Also, like the Sun-Times story mentions and like the governor has said, models are constantly changing when you put new data into them. This model has changed, changes, and will change every day as we put new data into it. The state could see the ‘peak’ within a span of time from now through May and then we could continue on that peak, which will probably be more of a plateau for a period before we see the decline.

       

24 Comments - TheInvisibleMan - Thursday, Apr 23, 20 @ 10:20 am: who needs facts and models. Plainfield announced just yesterday they are keeping their schedule for June to have their cruise nights event every Tuesday, where dozens of people gather in large groups downtown. Who were the decision makers you might ask. Perhaps doctors or the county health department. Nope. The local chamber of commerce made the decision, alone. - Pundent - Thursday, Apr 23, 20 @ 10:28 am: Unfortunately until the testing gets to a larger sustained level these models aren’t going to be particularly useful. Pritzker’s goal of 10K a day is likely not nearly enough and has to be increased by several multiples. Until then these assumptions will be a moving target. I’m not sure talking about when the peak will be even makes sense without it as it will only set expectations that may not be met. And the more that occurs the less confidence in how the situation as a whole is being managed. - Dee Lay - Thursday, Apr 23, 20 @ 10:44 am: The gov. needs to extend the shelter in place now for two weeks, and talk about reassessing every two weeks and making adjustments. - Todd - Thursday, Apr 23, 20 @ 10:47 am: why are they using the death rates when it appears they are being manipulated. If a person is in hospice dying from cancer and contracts the virus, I understand that cause of death can be attributed to raspatory failure , but was it the same as someone who caught the virus ended up on vent then died? If you test positive for the virus but die of a gunshot wound in Chicago, are they going to use this to lower the weekly body counts? “During Gov. JB Pritzker’s health briefing on Sunday, Dr. Ngozi Ezike, the Illinois Department of Public Health director, said anyone who had COVID-19 at the time of death, even if the person died of other causes, is counted among the COVID deaths.” - Anon221 - Thursday, Apr 23, 20 @ 10:55 am: Watched the Frontline special on COVID-19 last night. Deaths have been found attributable to COVID-19 or in conjunction with much earlier than recorded due to forensic investigations much later after death. This and the desperate need for ramping up reliable tests ( along with tracking and tracing), are vital to get a handle on the entire COVID-19 picture. https://amp.cnn.com/cnn/2020/04/22/us/california-deaths-earliest-in-us/index.html - Pundent - Thursday, Apr 23, 20 @ 10:56 am: =why are they using the death rates when it appears they are being manipulated= There’s a comorbidity element in most of these deaths. It may be hypertension, diabetes, obesity, etc. That’s consistent with certain populations being “at risk.” But to suggest that the data is suspect because somebody may have a gunshot wound is a stretch. - Excitable Boy - Thursday, Apr 23, 20 @ 10:57 am: Wow Todd, you and secondcitycop are in lockstep today. I guess you can all add public health expert to your resumes. - Chambanalyst - Thursday, Apr 23, 20 @ 11:04 am: Most interested to hear in the developments on the 2nd “T” of the 3 T’s: Testing, Tracing, and Treatment. How are we going to effectively contact trace? How can we accurately and reliably identify: 1. Where a person has been? 2. Who they have been within X feet of? 3. How do we notify every individual that’s been within X feet of an infected person? Even as ubiquitous as cell phones are these days, I’d think there’s still a decent size segment of the population that doesn’t have that technology, and demographically, they may be more vulnerable than those that do carry devices capable of providing that data. And lastly, i’m not well versed enough on privacy laws regarding that kind of data. Would people who CAN transmit that data have to consent to doing so first? Is there an expiry on that consent? - notsosure - Thursday, Apr 23, 20 @ 11:07 am: It seems to me comparing deaths with the same time the prior year might help ease the minds of the paranoid that they’re not overcounting. - VerySmallRocks - Thursday, Apr 23, 20 @ 11:12 am: -why are they using the death rates when it appears they are being manipulated— Manipulations can work both ways, such as no testing done on a death, or it inadvertently or deliberately omitted as a cause of death on the certificate. - Todd - Thursday, Apr 23, 20 @ 11:13 am: Pundent — comorbidity would make sense for not life threatening illness like the things you described. Those would seem to be reasonable attributions. But if someone is in ICU or hospise and then gets COVID while near death, attributing it to as a COVID fatality seems to be inflating or manipulating the numbers. EB — where di I claim to be an expert on any of this? I asked a question and cited the quote from JB’s point person. If you fail to get the sarcasm in the latter part of the post, 1. the mayhem in Chicago doesn’t seem to be slowing down, 2. Chicago reporting on crime is so bad the FBI won’t use their numbers, 3. look squirrel - Rich Miller - Thursday, Apr 23, 20 @ 11:16 am: ===attributing it to as a COVID fatality seems to be inflating or manipulating the numbers=== Oh, stop already. Nothing can be perfect in a human-based system and especially during an international crisis. And you can bet your gun stash that there are plenty of unreported cases in deaths. Again, stop it. This isn’t your Facebook page. - Pundent - Thursday, Apr 23, 20 @ 11:19 am: =But if someone is in ICU or hospise and then gets COVID while near death, attributing it to as a COVID fatality seems to be inflating or manipulating the numbers= And the same argument could be made if these deaths were excluded. There’s no evidence to suggest that the numbers are being significantly skewed for any of the reasons you’re inferring. There have likely been cases where no COVID-19 testing was done but likely played a role in a death as well. Nobody said that this was going to be perfect. - Demoralized - Thursday, Apr 23, 20 @ 11:28 am: ==If a person is in hospice dying from cancer and contracts the virus, I understand that cause of death can be attributed to raspatory failure , but was it the same as someone who caught the virus ended up on vent then died?== The answer is yes. Coronavirus contributed to both deaths. - Nagidam - Thursday, Apr 23, 20 @ 11:28 am: I guess in Todd’s world someone’s grandma that maybe would live 6 months to a year longer but died because she contracted COVID doesn’t count. Let’s not worry about people with underlying health conditions, they don’t matter as long as I can go shoot at my range and hand out with buddies. - Demoralized - Thursday, Apr 23, 20 @ 11:30 am: ==when it appears they are being manipulated== And when you say stuff like that you seem to be placing yourself in the camp of the people who thinks things aren’t as bad as people are making them seem. And those people are just ignorant. - Pundent - Thursday, Apr 23, 20 @ 11:43 am: Harvard is saying that going forward we will need to test somewhere in the magnitude of 5 million a day to upwards of 20 million a day to fully reopen the economy. Everyday that we spend ignoring that reality and debating how good or bad the disease and fatality counts are is a day lost. - Pundent - Thursday, Apr 23, 20 @ 11:44 am: Should have clarified that’s number of tests needed daily. For perspective since February we’ve administered 4.2M tests in total. - fs - Thursday, Apr 23, 20 @ 11:49 am: While imperfect, if the methodology of how people is consistent, it can give at least a minimally reliable trend line. It’s at least more reliable, and a good additional metric to use than just going off of only raw numbers of positive cases, when the number of tests administered per day can vary wildly. - Anon221 - Thursday, Apr 23, 20 @ 12:20 pm: Influenza related total death estimates 2018-2019 flu (entire- my emphasis) season (Table 1)- 34,157 https://www.cdc.gov/flu/about/burden/2018-2019.html CDC COVID -19 deaths to date recorded- 23,358 (Table 1)(My Note-these numbers are lagging behind actual reports) https://www.cdc.gov/nchs/nvss/vsrr/covid19/index.htm John Hopkins data shows over 41,000- https://coronavirus.jhu.edu/us-map Worldometer shows 48,788 deaths- https://www.worldometers.info/coronavirus/country/us/ It.Is.Not.The.Flu. We’re only a few months into the COVID-19 pandemic, and learning more and more about how it infects, spreads, complicates health. I’m getting so tired of people downplaying the risks this disease poses to all of the population. I have a family member recently diagnosed with cancer. Had one surgery just prior to the COVID-19 outbreak. Needs another one to make sure they got it all. Now we’re in a waiting game to seen when that might happen. If this was “just like the flu”, it won’t be as much of or maybe an issue at all. They would have the surgery and start treatment. But it isn’t, and they are not willing (nor is their doctor) to roll the dice that COVID-19 wouldn’t cause a complication and possible be a factor in co-morbidity should they get infected while in hospital and die because of complications from COVID-19 and not the cancer (which is one that is highly treatable). TEST.TRACK.TREAT. and ENFORCE those SBA loans to help the businesses that really need them to get through this. I want liberty as much as the next person. I miss getting out just to browse in the book store or wander the aisles of the hardware store. I want to support local businesses. But I’m not willing to take away the life and liberty of others when we can still invest some time to get things in order for the three T’s- especially the Test and Trace. we’re still a long way from Treatment, but the other two Ts are vital to get to that next step (banned punctuation). And when I do feel safe enough to go to local businesses, I will only go to those that are taking this pandemic seriously. - Groundhog Day - Thursday, Apr 23, 20 @ 12:22 pm: This virus has been very difficult to comprehend. As far as death counts go, I suspect we are significantly under-counting rather than over-counting. Data from around the world, and in NYC, comparing “normal” death rates to this year’s death rates demonstrate a much larger number dead this year. I have also heard from physicians caring for nursing home patients about a person who died suddenly. For some reason they were tested for COVID, and found to be positive. The first objective sign of it was the death. Further investigation revealed other facility patients asymptomatically infected as well. Scary times. - Dotnonymous - Thursday, Apr 23, 20 @ 12:47 pm: Stranger and stranger… https://www.theguardian.com/world/2020/apr/22/french-study-suggests-smokers-at-lower-risk-of-getting-coronavirus - Teve Demotte - Thursday, Apr 23, 20 @ 12:54 pm: My concern with JB’s handling of this is what intellectual resources is he tapping into to get a handle on testing, testing kit distribution, the number and frequency of testing, the purchasing of masks and other vital equipment and the management of data as to when the covid 19 curve will flatten. You have a lot of experienced private sector people in Illionis for instance in the area of logistics and supply chains, particularly as it relates to acquiring masks that are not defective (China) or test kits that work (South Korea). I don’t know if JB’s Deputy Governors are the point people managing this and there are teams of private sector professionals assisting them, as a all hands on deck mentality thing. If not, I think that is a mistake. Cameron Mock is really talented, my concern however is he the lead heading up a team of experienced data experts or essentially a solo actor? I think JB has done a good job. I do think however, bending the curve requires a greater effort on the state’s part to scale up. We all know the federal government has failed here. However, with unemployment rising and people getting more anxious by the day due to a loss of income, we cannot stay on lock-down through the summer. The state must find a way some way some how to exponentially test more people than what is currently happening. That is the key. No offense to the Governor’s staff, they are doing the best they can. However, is JB marshalling all the resources and talent he can to address the size and scale of the convid 19 challenge. - No Longer A Lurker - Thursday, Apr 23, 20 @ 1:21 pm: My wife was listening to Bob Sirott yesterday on his WGN morning show when his guest started discussing the Rt value. The value does change daily and is a key measure of how fast or slow the virus is growing. It’s an interesting graph and I thought I’d post it here. Illinois value is still pretty high, not good by the way, relative to other states. Here is the link. https://rt.live/#learn-more Sorry, comments for this post are now closed.