risk assessment tool

[Amanda]

I think maybe some of the confusion lies in what this assessment tool means in terms of chance of "exposure". I think it is only predicting what the chances are in a group of a particular size whether or not at least a single person would test positive for covid? I don't think it takes into consideration any mitigation. So if I am in a group of 100 outside my "risk" is going to be quite a bit different in terms of getting covid from an infected person than if I was in a smaller inside space. Am I interpreting right?

 

[skschlag]

Mod note - I have deleted two posts for "off topic discussion".

I almost removed another one, please keep this civil, please keep this about the risk assessment tool. If we cannot do this then yet another thread will be locked.

I have removed another one. Please keep this on topic and civil.

 

[mommyof1]

To expand on this and questions for clarification?

Does this take into account mitigation?

Specifically the following;

Ventilation?
Barriers?
Indoor vs Outdoor gathering/activity?
Type of activity?
Masks vs not?
Sanitizing?

The tool only gives the probability that an infected person is present, based on the overall infection rate and the size of the event. It does not calculate the probability that the infected person will transmit the virus to others at the event. That probability will obviously depend on the measures you list.

 

[mommyof1]

Mod note - I have deleted two posts for "off topic discussion".

I almost removed another one, please keep this civil, please keep this about the risk assessment tool. If we cannot do this then yet another thread will be locked.

I am concerned that a post challenging the tool’s validity was left up but the response to that post, pointing out valid and invalid reasons for challenging the tool, was removed. I will rephrase and hopefully that response will be acceptable.

 

[mommyof1]

My county is listing near 100% exposure rate for an event of 100 people... Yeah... No. Like I said (and had my post deleted for it b/c supposedly it was not "on topic" ) the tool is inaccurate.

Check out the technical information on the data source and how the estimates are calculated. Is there a specific concern with the methodology, as opposed to the results?

 

[doublestrike]

I’m too dumb for this, I have no understanding what this predicts. Lol.

 

[Deleted member 18037]

The tool only gives the probability that an infected person is present, based on the overall infection rate and the size of the event.

Again for clarification.

So 14 % risk assessment means it 14% likely 1 covid person is present?

Which would mean its 86% likely, there is no one Covid positive?

And that’s not including any mitigation?

Do I have that right?

 

[mommyof1]

Again for clarification.

So 14 % risk assessment means it 14% likely 1 covid person is present?

Which would mean its 86% likely, there is no one Covid positive?

And that’s not including any mitigation?

Do I have that right?

Yes. If there is a 14% probability that at least one infected person is present, then there is an 86% probability that no infected person is present. This is where individual risk preference factors in. I may not be willing to take a 14% chance of being in the same space as an infected person, but someone else might not find that same risk off-putting.

We have to disentangle mitigation at the community level from mitigation at the event. These estimates are based on the current observed infection rate. So community mitigation is already baked in to these estimates—if your community is doing more mitigation, the infection rate will be lower and the probability of an infected person’s being present will also be lower. The probability that an infected person will transmit the virus during the event is beyond the scope of these estimates, but will obviously depend on mitigation measures in place at the event itself. So you could imagine a community with very few mitigation measures. This community will have a high infection rate, and hence the tool will calculate a high probability of an infected person’s being present at an event. If, however, strong mitigation measures are in place at the event itself, that person may not spread the virus. Conversely, a community with strong mitigation measures will have a low infection rate, and the tool will calculate a low probability that an infected person will be present. However, if the event does not have strong mitigation measures in place, in the unlikely event that an infected person does show up, that person will probably transmit the virus to many others.

 

[Deleted member 18037]

These estimates are based on the current observed infection rate.

So then. the infection rate pretty much does the same thing, as expressed as a percentage?

The site is apparently down due to high traffic.

However our counties positive rate currently is 1.2 % Of those tested.
Add in the extra mitigation at my kids gym.

Risk assessment nearly 0

 

[mommyof1]

So then. the infection rate pretty much does the same thing, as expressed as a percentage?

The site is apparently down due to high traffic.

However our counties positive rate currently is 1.2 % Of those tested.
Add in the extra mitigation at my kids gym.

Risk assessment nearly 0

The infection rate is not the same thing as the probability that an infected person is at an event of a particular size, but it is the starting point for that calculation. The bigger the event, the higher the probability of an infected person’s being present given the same infection rate. The site explains the arithmetic.

 

[mommyof1]

Since the site is down, let's DIY just for fun. @Deleted member 18037, let's use your example of a reported infection rate of 1.2 percent. We'll assume 50 kids in the gym.

To find the probability that any person (one or more persons) in the gym is infected, we first need to find the probability that no one is infected. This is the probability of a repeated event: the probability that the first person, and the second, and the third, and so on all the way through the fiftieth person, is not infected. To do this, we take the probability that any given person is not infected (.988 if the infection rate is 1.2 percent) and raise it to the power of the number of people in the group. So we raise .988 to the 50th power, which gives us .547. So given an overall infection rate of 1.2 percent, there is a 54.7 percent chance that no one in a group of 50 is infected. Subtract from 100 percent, and we see that there is a 45.3 percent chance that anyone (one or more persons) in that same group is infected. If there are 100 people in the group, the probability that no one is infected falls to .299, and the chance of someone's being infected rises to 70.1 percent. So we can see that even with a low overall infection rate, the probability that someone in a large group is infected is quite high.

If infections are undercounted, the risk of an infected person's being present is greater. If there are 10 times as many actual infections as reported infections, the infection rate is 12 percent. In that case, the probability that a 50-person group includes at least one infected person is greater than 99 percent. @gymgal, that's probably why you are seeing such high numbers if you are looking at the county-level estimates--the county-level estimates assume a true infection rate 10x the reported rate. The tool only seems to use the range of assumptions (1x, 5x, 10x) at the state level.

Fun, right?

 

[gymgal]

Check out the technical information on the data source and how the estimates are calculated. Is there a specific concern with the methodology, as opposed to the results?

yes, there is. The numbers used are wrong. There are not that many active cases compared to the the population. Looks like they used total cases, not active.

 

[mommyof1]

yes, there is. The numbers used are wrong. There are not that many active cases compared to the the population. Looks like they used total cases, not active.

They are using "circulating" or current infections.

 

[novagymmom]

My county is listing near 100% exposure rate for an event of 100 people... Yeah... No. Like I said (and had my post deleted for it b/c supposedly it was not "on topic" ) the tool is inaccurate.

Does this take into account mitigation?

Specifically the following;

Ventilation?
Barriers?
Indoor vs Outdoor gathering/activity?
Type of activity?
Masks vs not?
Sanitizing?

I'm not sure that mitigation factors are relevant for this tool. The tool isn't suggesting that there is close to a 100% chance that the virus would be transmitted at a 100 person event in your county. Just that there is a (close to 100%) chance that at least one person at the event HAS the virus (and I believe the county figures are the 10x CI). So, depending on the current population infected in a given county, it is very possible that at least 1% (or 1 in a 100 person event) has it. That's how I had interpretted the model at least.

 

[Deleted member 18037]

I'm not sure that mitigation factors are relevant for this tool. The tool isn't suggesting that there is close to a 100% chance that the virus would be transmitted at a 100 person event in your county. Just that there is a (close to 100%) chance that at least one person at the event HAS the virus (and I believe the county figures are the 10x CI). So, depending on the current population infected in a given county, it is very possible that at least 1% (or 1 in a 100 person event) has it. That's how I had interpretted the model at least.

I am less concerned with who has it. I look at what is the chance of transmitting it to others when think of risk. Mitigation decreases the risk that person who has it will transmit it.

 

[novagymmom]

I am less concerned with who has it. I look at what is the chance of transmitting it to others when think of risk. Mitigation decreases the risk that person who has it will transmit it.

That's correct. The tool calculates the chance that someone there HAS COVID and that, along with mitigation factors at the event, would impact the risk of transmission.

 

[txgymfan]

It’s hard to have a thread going when you’re obviously on the extreme side of advocating masks for all, lockdowns, and quarantines. It’s ok to take a breath and let people have a discussion without being berated and shamed and then having threads locked. We aren’t here for a lecture every time we have a discussion. Thank you.

I have started a herd immunity thread for you.

 

[bogwoppit]

I have given out a warning, and will give out others if I feel it is necessary. Please stay on topic, and please be civil.