Not sure if this is the right forum, but here goes.
I work for a gym. We’re currently closed but putting in measures ready for reopening. We’re considering temperature checking every user coming in using a contactless IR thermometer. This would be an initial check, an unusual reading would have the user pulled to one side and asked further questions.
Are these things accurate enough? Worried about too many false-positives here.
And if the are good enough, can anyone recommend a decent model?
I have one that work gave me. We also have one at work by the front entrance. While most of us are working from home, if we need to go into the office for any reason we are required to check our temperature and record it. If it is too high, you have to go home. No exceptions.
The one we have seems to read a little low, but it works well enough.
I just picked up the box to get the brand name off of it, and there is no brand name whatsoever on the box. The only identifier on the box is “Item No. BLIR-3”.
I can’t compare it to any others since this is the only one that I have ever used.
Work also gave me an Osport pulse oximeter. Seems to work pretty well, but they printed the instructions in micro-text that I could not even begin to read even with my glasses on.
I’m not sure how accurate they are. But if you’re going to buy one, I recommend a unit that is designed to measure forehead temperature. Something like this.
We use one of these and it works well for us… https://www.thermoworks.com/wand
Forehead temperature is roughly 1 degree F lower than oral temperature. The thermometer I linked makes that conversion. Someone else said they thought theirs read a bit low. It could be that theirs is perfectly accurate but not making the “forehead to oral” conversion.
Disclaimers and all that, but if your intention is to screen out people with an elevated body temperature, I would worry that, no, it is not at all accurate.
First problem is the calibration and correction for emissivity value of different types of skin. Secondly, skin temperature can vary between 33.5 and 36.9 depending on where you measure it, and also depends on the air temperature, patient’s gender, and all sorts of other factors (eg see here). Are you prepared to deal with all that? Last time I was in a hospital, they measured under the tongue.
Infrared thermometers can be made far more accurately and precisely than would be needed for any medical purpose. They can also be made cheaply with far inferior precision and quality. You’ll have to do research on the brands and models that you’re looking at.
It’s certainly not accurate for a definitive reading, and you are correct that values can vary by people, time of day, etc. But they can still be a good solution for initial screening. I visited a hospital last week, and was scanned on my neck with a touchless IR thermometer on entrance to the building and again at the entrance to the lab. The medical staff doing the scanning trust this method well enough to use it.
Did a little research, and there are some who believe more consistent and accurate results could be obtained if the IR thermometer measured the temperature of the inside of the mouth. The person-under-test would open their mouth, and the tester would point the LED at the back of the throat.
That might be true, but having people open their mouths and breathe in your face is not a good idea with a disease that can be transmitted by aerosols.
And if you’re using an IR thermometer designed for foreheads, it has a temperature correction built in so it shows the presumed internal temp based on the forehead temp. You’ll get a less accurate reading if you use that on a spot inside the mouth.
Since it measures skin temperature, it’s probably designed assuming the person being tested is indoors and calm. It may be thrown off if someone’s skin is abnormally hot from something external like standing in the sun or working out. I would think this might be a concern at a gym where people might have been standing outside waiting to get in or ran to the gym as part of their exercise routine. You should do some test runs with your trainers in the gym to see if that’s true. Take the trainer’s temp and then have them work out hard and see how their temp changes.
I just used a kitchen IR thermometer on both spots. I got consistent readings of only .2-.3 difference.
And the forehead reading was 0.2-0.3 degrees more accurate. 
In general, a forehead temperature is 0.5-1.0 °F (0.3-0.6 °C) lower than an under-tongue temp. The tongue surface or back-of-mouth temp is probably between those two, so your measurements make sense.
I just bought a Braun BNT400CA no-touch thermometer. We have to take our temperatures each day before going to work (when we are recalled). You aim it between the eyebrows. There is a light to help aim. The manual said that it reads about half a degree (Celsius) low.
My employer just started testing us with THIS bit of awesomeness. The TI480 from Fluke. The powers that be seem to think it’s an impressive bit of kit.
Under 10 grand. What a bargain!
We played around with a thermal imager when I was in architecture school. People heat up in the most interesting places when they get embarrassed…