Infrared thermometers to screen for Covid: more marketing than medical evidence
cebm.net
cebm.net
Here in the Portland, OR area I've noticed a definite change in how many people are wearing masks. At the grocery store, 2 months ago there were about 5% of the people not wearing masks and they all tended to be men in their 30s and 40s. At convenience stores in more suburban area, I was the only one in a mask. Now it seems like 100% of the people are wearing masks.
Wearing masks, not temperature checks, is what seems to be our best option.
> Good to see an article calling this out. Temperature-screening might catch a few people who know that they have a temperature (or are just not feeling well) and are going to "tough it out".
The article says that 30000 people were tested. 4 people were false positives, 0 people were true positives. This is called ‘many false positives’. Yet the people who actually had high temperatures were supposed to be at home anyway, right?
4 out of 30000 is not many, they are not real false positives and the design is flawed anyway because people are going to self select, as stated. So the article is just bullshit. Let’s not even go into the rest which tells some unrelated anecdotes about Ebola and other diseases that are not the same diseases and serve nothing but to indicate vague trends under the guise of evidence.
> Here in the Portland, OR area I've noticed a definite change in how many people are wearing masks. At the grocery store, 2 months ago there were about 5% of the people not wearing masks and they all tended to be men in their 30s and 40s. At convenience stores in more suburban area, I was the only one in a mask. Now it seems like 100% of the people are wearing masks.
This is a completely unrelated tangent. According to this anecdotal non evidence mask use has increased among some stereotyped group and in some establishments. Even if taken at face value this tells nobody anything about the effect of the masks.
> Wearing masks, not temperature checks, is what seems to be our best option.
That’s an opinion, based on nothing, which even if assumed true means nothing. To the writer something seems to be the best option.
That’s fine and maybe it even is the best option.
But does it really help to announce they jumped on the mask bandwagon before ‘men in their 30s and 40s’ and people in ‘convenience stores in more suburban area’?
I should hasten to add that I'm not an epidemiologist, so I don't know whether this is true. But my intuition suggests that even if this is just partially effective it might still be worthwhile in the context of national efforts to control the spread, which is a numbers game.
I think this is a crucial distinction between "security theater" and "hygiene theater" -- security risks are not contagious and don't have the same potential for exponential growth that virus transmission does.
Last week I rocked up to operate at a hospital I irregularly work at which has recently introduced checks (as opposed to all public hospitals which have had this in place since March). In theatres we were talking about how useless it was. One of the nurses said someone tested high last week. They made the person sit in a chair for half an hour and then re tested them and granted them entry. To some degree I get it. I mean, my state is only recording 15-20 positives a day across almost 20,000 tests so the pre-test probability is small. On the other hand, what’s the point of a policy if it doesn’t change anything.
Another aspect of ‘hygiene theatre’ - when the apple stores reopened in sydney I happened to be one of the first back in (by coincidence). I visited late on the first day. I was initially impressed - security guard scanning, handing masks out. Inside the sales rep who served me had a mask continually falling down his face and kept grabbing at his face and readjusting. I eventually capitulated and told him he was wearing it upside down (metal bit across bridge of nose needs to be squeezed). Mask hygiene is important but pretty much everyone is so slack with them that, whilst they may be decreasing aerosolisation, they would be actively increasing spread of body fluids on surfaces, since so many people put a mask on and then immediately give up hand hygiene.
It’s all a bit depressing
While 30 minutes seems excessive, this makes sense. Your surface temperature is heavily biased by your environment’s temperature. Letting one settle in a controlled environment to test is reasonable to see if it’s really a fever (temperature will stay elevated) or due to environmental factors.
An issue is that anyone could ice their forehead in advance and beat the test. Dunno if too low of a temperature triggers a fail. In theory, everyone should have to sit in a controlled environment before forehead testing, but at that point, may as well do tympanic checks.
(Among many other issues, like the massive variance that’s probably accepted because of unreliability of surface temperatures as a proxy for internal temps)
Right now where I am, air conditioning isn’t common, but it was slightly above body temperature at the peaks of each day (I guess 100F). I wonder how that worked out with scans.
Given that the time from infection to symptomality is as short as 48 hours, this means it may be possible to be infected by someone before either they or the person that infected them has started to show any symptoms.
You may be very infectious when symptomatic, but you're likely bedridden and the few people around you are taking some precautions.
On the other hand you may be less infectious while presymptomatic, but live a normal life, effectively infecting more people right before you "get sick" yourself.
1. Store reused masks in a designated spot to avoid cross contamination.
2. Wash hands after putting on and taking off used masks.
3. Dispose of masks which get wet or N95 masks which get folded.
This care can be achieved by PSAs in countries with functional governments.
There are good reasons they are called disposable masks.
Disposable masks that the majority of people are using are are not for reuse, and it says right on the box "not for viruses". The "oh but large particles they (aerosol experts) didn't think of that" thing is surreal to see people say with a covered face.
Yet those same masks are reused multiple times per day. Taken on and off. The multi-day exponential hand wave fails on reuse, the exhale re-hydrates the medium. Masks have a finite capacity before they are a net negative, and cloth IRL insanity is a growth medium.
The psychology of "wear a fake mask for others" is next level propaganda. It's right in line with the drastically more dangerous social contagion that person X knows better than person Y and therefore is righteous to force their ideas (medical procedures) on person Y.
Conditioning people to make medical decisions for others is too obvious, but here we are, fake masks and all.
It's a deadly religion, which repeats over and over in human history.
Line up. Right? https://www.clickondetroit.com/news/2020/08/03/oakland-unive...
"There is no benefit and there may even be negative impact." https://www.dailymail.co.uk/news/article-8583925/The-land-no...
-Seriously-: Why not wear the masks forever?
yet...
"Every single claim you make here requires scientific evidence as in actual studies that demonstrate such effects."
Boyle's Law: http://v6y.net/maskop2
Your turn, maybe you can start with "Reuse of medical masks is fine". This issue clearly is generating interest.
Can you say it? Why not wear the masks forever?
Obviously medical professionals or other people who come in close contact would be better off using medical-grade masks in a hygienic and consistent way- but for the majority of grocery shoppers and such, I thought the purpose was simply reducing your ability to spread the germs.
And cause others with symptoms to stay home in the first place with the fear that they'll be caught, wasting the trip (e.g. to LAX in LA traffic), keeping them out of the secured area & also off transportation to that area.
Now that likely won't work if you're in the throws of a Covid infection, but in the first few days when you're not sure if you're really sick? It probably does a great job of masking fever.
All that said, a low-grade fever isn't all that harmful. I still tend to take something. I used to try to tough it out before the experience above, but also found that without the body aches and headache that comes with the fever, I slept better while sick, even though I still would sleep for hours. I also tended to feel better a little sooner. Of course, this isn't data, and I've read the stuff about the beneficial fever before, but until general medical advice tells me to do differently, I'll stick to comforts.
[1] https://www.mayoclinic.org/diseases-conditions/fever/symptom...
[1] https://www.businessinsider.com/cdc-recommends-different-fac...
They’re no better than a surgical mask if you have a lot of gaps: they’ll filter nothing but your coughs. So little inhalation will be filtered because of the path of least resistance.
At the very least they provide false sense of protection. If people want to be protected, they should use a real mask designed for viruses.
In situations where 100% mask compliance exits, they are still not effective.
Fort Benning: "according to Elizabeth Howe of Connecting Vets. About 640 recruits from the 30th AG Battalion and 2nd Battalion, 29th Infantry Regiment arrived at Fort Benning in May for training and were immediately tested for COVID-19. Four recruits tested positive and were removed from the group while the remaining soldiers were placed in isolation for 14 days without any training exercises. After the 14-day quarantine, they were all retested and every single one was negative. Only then did the training commence – with the full panoply of obsessive social distancing measures, including mask wearing. You can imagine that there will never be greater compliance to these rules than during military training. Yet, just eight days later, after one recruit exhibited symptoms, 142 of the trainees tested positive. That is 22% of the entire group isolated and quarantined together. As they were young, none of them were hospitalized and most were asymptomatic."
https://www.theblaze.com/horowitz-from-fort-benning-to-japan...
And then there is the psychology, which is... more troubling to say the least. To the people promoting this pseudoscience, why not wear the fake masks forever?
https://stpauls.vxcommunity.com/Issue/Us-Experiment-On-Infan...
Why are you framing this as if there is a single "best option", as opposed to a plurality of complementary techniques?
IMO catching/discouraging a few people who have a fever but aren't isolating is worthwhile.
In my experience, I've only ever seen temperature checks at doctor's offices and hospitals, where they also have mask policies so "thorough" that I sometimes have to argue to keep my P100 non-glasses-fogging respirator on.
Future historians will wonder why various medical records show people with acute idiopathic hypothermia in the middle of July 2020.
You underestimate how many people in the database (from any time period) will have a recorded body temperature of 0, thanks to weak typing and buggy null coalescence.
There might be stuff we can do indoors, but state government agencies aren't funding fast research into it - efficacy of filters and alterations to HVAC systems like installing UV-C lamps in them. Instead it's just "wear masks and take temperatures and get mad at people when they don't wear masks," which is perfect shirking of responsibility that large organizations love to do.
In contrast, a plain old filter is likely to be extremely effective. A HEPA filter will catch almost all particles big enough to contain a virus, and a MERV 13 or better filter is quite good. All you need is a filter with enough surface area that a good filter has low pressure drip.
Almost all HVAC systems already have filters, so upgrading them is usually straightforward.
(I used to have a fancy electrostatic precipitator in my HVAC system. What a waste of power and effort in keeping it clean.)
The problem with installing a MERV 13 filter in a normal HVAC system is that the pressure (usually) isn't rated for it and can cause serious damage to the motor(s) as you noted, which is more common than people think (definitely the case in my home system).
UV-C installation is fairly simple; though I agree they are power hungry and you need to keep them on most of the time with maximum one power cycle per 24 hour period. They're cheap upfront and do work on normal coronaviruses, though.
Still the real problem is very little state-funded research on the topic, or development of some mitigation techniques for businesses that must run indoors. It's once again fairly useless bureaucracy.
I liken using UV-C (potentially) and a good filter as a defense-in-depth technique, just like masks, social distancing, sanitizer, temp checks (all which do very little on their own, but in the aggregate help).
With a few exceptions of more complex units all of the recirculated air spends time passing through the coil.
UVC doesn't damage metal parts, but does effect some materials that would be common in residential installations
https://customer.honeywell.com/resources/techlit/TechLitDocu...
https://www.usnews.com/news/national-news/articles/2019-04-0....
And those who don't die suffer all kinds of ill effects like asthma, reduced cognitive performance, infertility.
If we ignored COVID, but spent 2020 replacing all coal and oil fueled cars, trucks and powerplants, we would save 10x as many lives as would have been lost to COVID this decade, and make significant progress on climate change. A shameful missed opportunity.
It's possible to have multiple disasters at once.
CDC Influenza and pneumonia deaths by influenza season and age: United States, 2008–2015: https://www.cdc.gov/nchs/data/health_policy/influenza-and-pn... (these are not estimates, see the footnote)
2015-2016
Flu: 7,961
Pneumonia: 131,858
All: 1,769,940
Curious, where did you see this? I can't remember seeing anything like that (but that's not to suggest it didn't happen)
There were already publications from China in early February stressing that tracing the virus via symptomatic cases alone would not be enough to stop the spread. Here’s one from Feb 11. https://www.medrxiv.org/content/10.1101/2020.02.07.20021154v...
China suppressed the virus not by IR scanning but by a well publicized system of travel restrictions and contact isolation/quarantine of anyone who had been in contact with a known case, regardless of whether they had a fever or other symptoms.
All this was public knowledge in February and can be verified by reviewing preprints and news articles from then.
It isn't hard to understand his difficulty with normal infrared thermometers: they don't work through fur. But in all animals (us included) temperature can vary many degrees depending on where it is measured. The face/forehead is not the best place if you are looking for 'core' temp. People with reduced blood flow (ie heart disease) will have much cooler foreheads. They may have a fever but their foreheads could appear rather normal.
Why is that? Is there any danger?
Here is a similar device, with a picture of it being used on a dog: https://www.reichert.com/product_details.cfm?skuId=3125&skuT...
The scatter plots of eye/ear/rectal temperature in this article are on point to the covid situations. Variations of +/- several degrees are observed, showing that IR temperature scans might be less than useful in detecting mild fevers. (Also, stubborn cat refusing to open eyes)
https://www.theveterinarynurse.com/research/article/investig...
(i) Many thermal cameras and temperature screening products were originally designed for non-medical purposes, such as for building or site security. Businesses and organisations need to know that using these products for temperature screening could put people’s health at risk.
(ii) Temperature readings from temperature screening systems will measure skin temperature rather than core body temperature. In either case, natural fluctuations in temperature can occur among healthy individuals. These readings are therefore an unreliable measure for detection of COVID-19 or other diseases which may cause fever. Furthermore, infected people who do not develop a fever or who do not show any symptoms would not be detected by a temperature reading and could be more likely to unknowingly spread the virus.
[0] https://www.gov.uk/government/news/dont-rely-on-temperature-...
There is a subset of the population that wants to be told that whoever is in charge is "doing something."
Compliance with that "something" never seems high on the list. Honestly, some people seem to like that as it gives them something extra to gripe about.
On the other hand, I have visited a DHL warehouse to ship some things in the Bay Area recently. Before I can enter, a guard at the door points an infrared thermometer at my forehead. "Eighty-five degrees! Come on in, sir!"
sounds like they used a general purpose infrared thermometer rather than the ones designed for medical use.
I remember having a hell of a fever during H1N1 going through a airport being scared they would catch me and I'd have to quarantine. Because it was continue or get stuck I continued, if it was just walking into a store....
https://www.richtech-ai.com/product-page/aatss-v2-pro-enterp...
Sadly, we never got to the facial recognition stage because the client decided they didn't want to support it.
Of course, that would require people to stop long enough to take a reading, and would be useless for the products trying to scan crowds with an unattended camera and AI.
And, of course, nothing is really going to help you if the thermometers are inaccurate to begin with.
Forehead temperature is highly variable with ambient conditions - sweating while walking in reduces surface temperature, but sitting in a low-humidity air-conditioned car with your face in the morning sun increases it. After as little as 60 seconds in a conditioned space, though, skin surface temp is pretty much normalized.
The camera scanners are using Flir microbolometers, which have a huge amount of noise/calibration variation, even on a per-pixel basis. The handheld guns and kiosks which require alignment are using thermopile sensors work much better for employee check-in, they have like a 1:10 spot size so you need to be within a foot or so of the sensor and intentionally align yourself with the sensor. The reading takes about 2 seconds to align and measure, which is fine for clocking in but, as you said, doesn't work in a crowd.
Old-fashioned oral thermometers warn that they'll be inaccurate if you've drunk a hot or cold beverage recently. I imagine back of the throat thermometers would have the same flaw?
Generally those readings can vary a few degrees even if you are using it well. It's just not made to measure fractions of a degree.
So I was wondering how infrared could possibly work... Not really surprised it doesn't, but the stats at how bad it has been in the past (h1n1, Ebola) are shockingly bad. I guess I was hoping there was some medical variant of these thermometers, which is accurate in the 95-105F range, and the cooking thermometers just sacrificed accuracy for range. But it sounds like that's not the case.
They compensate.
I have an iHealth unit that Wyze was selling a few months ago. I check my temperature several times a day. I believe it's quite accurate. In the mornings it'll read 98.2˚F, afternoons I'm at 98.4˚F, and just before bed it'll read either 98.4˚F or 98.6˚F.
The fact that it's this consistent with the intraday variation gives me some confidence in it. But, granted, I have yet to have a fever. (That I know of!)
I'm not sure what it uses to arrive at the answer. It could be as simple as adding a fixed constant or multiplier to the forehead surface temperature. Or it could separately measure the ambient temperature and factor that into the radiant heat coming from my forehead to arrive at an answer.
Fortunately, the cruise operator was giving everyone a forehead IR thermometer scan, before letting them board the ship.
Unfortunately, at least one person ahead of them in line happened to scan in, and be let on board, sporting a cool temperature of ~85 degrees. I'm not sure if 'body temperature of a two-hour-old-corpse' is a symptom of COVID, but this did not ease my friends' anxieties for the duration of the trip.
Most common problem: operator error. Are you reading the temperature correctly? Tip, it should not be read directly from the middle of the forehead https://www.seattlechildrens.org/conditions/a-z/fever-how-to...
If the temperature reads anything lower than 36C, you're doing it wrong and it's very easy for people to get a non-medical IR thermometer, start pointing at people and "checking temperatures"
That being said, it seems Covid-19 fevers are usually more intense than flu fevers (even H1N1 ones)
Alright, 15bucks later, I tried it on myself... red alert, fever-fever, when in-ear. Hmmm, roll around forehead -- ambient temp.. The old-fashioned palm-to-forehead -- obviously, no reason for alert. Finally, test with the old in-mouth thermometer- all normal.
Reading the instructions only confirmed the unreliable nature of the instant readings. We kept it anyway for "backup", but for practical use continued with the "slow" in-mouth beep-beep, it also helped that it looked like a froglet.
Apparently, a success of in-ear measurement is a direct view of the eardrum by the sensor, which is not always possible, esp. with babies.
Everyone else in this thread seems to be focusing on the problems with the probability of fever given that someone's infectious, when the real question is the probability that someone is infectious given that they have a fever, which might be due to anything else.
But medical discrimination is okay in 2020 I guess.
Wearing masks on the other hand is not a solution if people does not respect spacing. The disease is stoppable only by vaccines at this moment. Most of us will catch it before that. This is the reality.