Time to redefine normal body temperature? (2020)
health.harvard.edu
health.harvard.edu
We use both high-quality ear canal and anal thermometers, and while we have enough statistical grounds to unbias their results compared to each other, there remains significant variance. I.e. if we use an anal or ear canal thermometer makes temperature vary ±0.4 °C.
Then how sloppily you take the measurement with the same thermometer can result in a ±0.2 °C difference for the butt, and ±0.6 °C for the ear canal.
Of course, the time of day we take the temperature can give us a ±0.5 °C difference, and the activity level of the person something like a ±1 °C difference (we don't have as much data on these last ones, naturally.)
And all this is before you bring in the fever! So all of this we consider normal.
Basically, in our experience, anything below 38.4 °C could just as well be caused by any of a number of non-fever conditions at something like a 90 % confidence level, so unless there's a strong prior of fever, a "non-normal" temperature reading on its own is a fairly weak signal.
...which brings me to my final point: the prior ought to strongly affect what counts as normal temperature, but all of this discussion ignores it.
There's also the question whether you prefer false positives or false negatives when it comes to fever detection. My impression is that the medical community prefers false negatives, i.e. pretending that people with a low fever are perfectly normal -- but a lot of private persons I talk to have the opposite preference: they would prefer to think of someone as sick even though they are healthy.
Like any good medical test, you pick a threshold based on what probability you want for false positives and negatives, and you weigh the prior into this judgment. Just investigating what an average resting temperature with a particular measurement method seems to me silly.
But in the end, I want to get to the point where you are, using a rough rule of thumb. I'm just taking the evidence-based route there!
Indeed, this pseudo-accuracy seems pointless.
The only relevant conclusions that taking the temperature can deliver is whether, a) there is no fever, b) there is a mild fever not requiring major concern, c) there is a major fever and extra care needs to be taken. No one in their right mind would classify these differences based on sub-degree differences in temperature.
Instead of just giving temperature, a thermometer could say "there is a 70% chance that your immune system is more active than usual" etc
Of course, small additional data points like "i'm exhausted and coughing more than usual" increase the confidence of the result by orders of magnitude over just temperature...
This part is just known, though: the usual "close enough" standard for core temperature is an anal temperature and everything else you have to adjust. (Everything else you say I agree with, though.)
(For what it's worth, my local hospitals use ear canal temperatures for diagnostic purposes.)
1. If the ear temperature are consistently wrong (systematic error) it is very easy to debias them and they become equivalent (modulo random error) to anal temperatures.
2. Perhaps cephalic temperature is a better diagnostic tool than gut temperature to evaluate the effects of fever? After all, the brain is a more sensitive organ than the butt.
The more time I spend on the internet, the less certain of this I become.
The real problem appears to be that converting to Fahrenheit and reporting this as “98.6F” suggests an order of magnitude more precision than “37C”. 36.5C would still round to 37 but corresponds to 97.7F.
> The average normal temperature of the healthy human body in its interior, or in carefully covered situations on its surface, varies, according to the plan of measurement from 98.6° to 99.5° Fahr. (37° to 37.5° C.). It is about 98.6° in the well-closed axilla, and a few tenths of a degree higher (.5-1½ or 2° Fahr., .7° Fahr. average) in the rectum and vagina.
So, as far as I can tell it's not a sig-figs issue in conversion, but the actual original source that produced "98.6°". It almost seems like Wunderlich should've used 37.0° to 37.5°C to be consistent.
I'm a little surprised and disappointed that the translation also translates the units, though. That seems to be stepping just outside the bounds of what I would want translated (at least, in the context of a medical journal or scientific writing).
Thanks for that correction!
Funny that people argue over this and some guy named Wunderlich who measured a bunch of asses.
Did he write that translation or is Google Books fake news?
Actually makes it even more interesting, Wunderlich and his reaming to get some Reaumurs.
What were his actual measurements taken in and how accurate was his thermometer? And who's vaginas has it been in?
Curious I get flagged for pointing out the potential measurement inaccuracies and the errs of scientific reproducibility, all while having some humor.
I've had this happen as an adult: The doctors office had me take 800mg of children's ibuprofen and had me wait until the fever started coming down: If it didn't, they were going to call an ambulance. Sidenote: Children's liquid medicine tastes horrible.
40C (104F) isn't just a little fever, and is definitely in the danger zone for both children and adults. You should probably get to the doctor before that, in general - especially if you react as I do at a lower temperature.
Currently the top comment says "Basically, in our experience, anything below 38.4 °C could just as well be caused by any of a number of non-fever conditions at something like a 90 % confidence level" ... which is panicky territory for me. But anything under 38 is fine.
36-36.6 is considered normal, 37 is considered a mild alteration (or life threatening disease if you're a man).
Also, I was one of the men that complained a ton during cold. Turned out, at the age of 34, that I had a deviated septum and the level of nose congestion when I was healthy was comparable to when I was healthy. Even slight cold was absurdly doffucult for me. Since they fixed it, the colds are super light for me.
But I had to take my girlfriends’ laughs for 8 years each time I had cold.
I never miss an opportunity to mention this story when people laugh about some people having „man colds”.
What country/ language, if you don’t mind me asking?
Let me explain: measurements were usually taken externally from armpits, using analog mercury thermometers, which results in temperatures lower than the actual one by ~0.5 degrees Celsius.
If you read the instruction manual of modern digital thermometers, that are more accurate, it states that
Oral
Time: ~55 seconds
Typical temperature interval: 35,5~37,5 °C
Anal
Time: ~45 seconds
Typical temperature interval: 36,6~38 °C
Armpit
Time: ~70 seconds
Typical temperature interval: 34,7~37,3 °C
https://www.doctorpoint.it/images/deplian/docp-pdf10956.pdf
The science has always been sound, even in Italy.
My parents worked in healthcare, I never had the chance to skip school unless the thermometers said > 37.5 °C and after a double check with both palms on my forehead because I warmed it more than once on the radiator.
The problem is Italian paranoid moms, especially from the South, believing in spooky superstitions.
The contactless infrared ones, but also the contact digital ones have been proved to provide "false measurements", the stated accuracy is usually +/- 0.2 C, but there are many (ambient) factors that may influence greatly the result, the good thing being that they are very fast.
The gallium/galistan ones (1:1 replacement for mercury ones) are often hard to read (they are almost all of the "prismatic" kind) and more often than not they simply get "stuck", and they are very, very, slow.
they are not more accurate per se, the reading and calibration is.
many cheap thermometer were printed with the temperature lines out of scale or shifted.
they are also safer to use, it wasn't uncommon that the old ones broke while using them.
Imagine the danger of a glass tube filled with mercury breaking while measuring the temperature anally or from the mouth.
>Imagine the danger of a glass tube filled with mercury breaking while measuring the temperature anally or from the mouth.
Oww, comeon, they are of course safer, but it is not like "it wasn't uncommon".
It was (and is for the - still largely glass made - gallium ones) an extremely rare occurence that a termometer broke while in use, though they often broke when they fell accidentally and - at least according to EPA - that seems like a serious hazard:
https://www.epa.gov/mercury/what-do-if-mercury-thermometer-b...
I cannot find any statistics on actual cases, but judging from this article preview:
https://www.sciencedirect.com/science/article/abs/pii/S02782...
they counted at the time 43 cases of mercury poisoning of which 7 related to broken thermometers while in use, and of these 7, only 1 was about thermometer broken while in the mouth affecting the oral mucosa.
The reported "new" one was about a lady that had her cat striking her in the face while she was using the thermometer in her mouth.
Please also consider the fact that Italy is in the EU, EU made those kinds of thermometers illegal in the Union with the directive 2007/51/EC [1] that was ratified in Italy by the decree n. 245 on 18.10.2008 which basically made it illegal to sell measuring instruments containing mercury since April 3rd 2009. Mercury is considered toxic waste in EU.
[1] https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A...
Surely they were made illegal by the EU directive and the following Italian decree.
That doesn't mean that people in Italian hospitals having thermometers break in the mouth (or other orifices) was "not uncommon" (possibly we use different definitions for not uncommon).
Unless - of course - you can recall many cases in your 40 years of acquaintenance with hospitals in Italy.
Let's say that a hospital has 100 beds and average stay is one week, that should be in the ballpark of at least 52x100x2 (temperature checks) = 10,000 measurements per year x 40 years around 400,000 uses of thermometers (possibly many more).
How many of these accidents can you remember roughly (1, 10, 100, 1000)?
As I've said, I'm familiar with the specificity of Italian hospitals, and I'm saying to you it was not uncommon.
Does not mean very often, it means "something every person working in healthcare has been trained for, because it happens (happened) more often than you might think".
Especially if you think about the fact that I've already specified that they simply broke, not that "they broke in people's mouth" all the times.
> Let's say that a hospital has 100 beds and average stay is one week
Let's say that the average Italian hospital in big cities has hundreds of beds, at the times of the ban on mercury Rome had ~6 beds every 1,000 residents, US had 2.7 of them.
Let's also add that in Italian hospitals you don't pay for the bed.
My mom stayed in hospital for 6 months when she had tuberculosis.
Many of the patients in the same ward staid on average for a year.
> 10,000 measurements per year
That's per week here, probably more than that.
The hospital where my parents worked has > 1 thousand beds and > 5 thousands employees.
Not counting the ER and daily scheduled examinations.
And it's not the biggest one.
Good, so in a week, or 10,000 thermometers uses, how many broke, on average?
1, 10 or 100?
And we are talking only of those that you prompted to imagine the danger of:
>Imagine the danger of a glass tube filled with mercury breaking while measuring the temperature anally or from the mouth.
Is 98.6 an american-only thing?
eg:
> You probably have a fever if your temperature is 38°C or higher.
> A normal temperature is around 36-37°C, although it depends on your age, what you’ve been doing, the time of day and how you take the measurement.
( further notes on taking temp., children, etc. )
I personally think it really isn't that important though, because how people measure it is way more imprecise/fluctuant than the standard itself.
As a side note, I hate digital contact thermometers with passion. Mercury is dangerous as hell but if it wasn't for gallium alloy ones I'd smuggle them in the country.
It's elemental mercury encapsulated in glass not one of the organic compounds that cause all the acute and longterm health issues. Chlorine bleach is scarier stuff used in the household.
Mercury is mostly banned for being persistent pollutant, next after lead. There is no real safe level of mercury intake due to cumulative damage and hard for body to get rid of.
There is a safe, heck, required level of chlorine intake. (What is table salt?)
Mercury is not any superior at measuring temperature over a resistive electronic or a bimetallic thermometer, or alcohol capillary one. (Except the latter boils and freezes too easily, not a problem for body temperature.)
There was a spate of TikTok's recommending mixing them for cleaning purposes. To the horror of many.
Drink bleach and it'll probably kill you or at least give you terrible injuries.
I have noticed over the last decade, nurses checking my temp are more likely to say something like “You are 97.6°F. That’s normal.”
It’s not mentioned in the article, which I find odd, because it tells you something about averaging statistics and male body bias in science.
Sience is starting to wake up to this though, but slowly.
I know it's not exactly "thinking with statistics", but it strikes me as too coincidental that the obesity epidemic has only exploded in recent decades too.
Now, as an adult, my metabolism nosedived. I barely feel like eating once a day. It's not abnormal to get temps in the 96s, though that could just be the cheap covid thermometers.
But my temperature when measured is consistently below 36 celsius, often 35.
I'm curious how direct it is. If someone starts exercising from being sedentary for years does their body temperature raise appreciably?
"28.5, ok, next"
They were supposed to measure your forehead temperature with that infrared thermometer. But due to some insane rumor spread around the net about these thermometers somehow hurting the pineal gland or something like that, everyone starting measuring on the wrist, even though it's way less accurate.
That part resonates with me. I had a pretty bad tooth infection after delaying seeing a dentist for a cavity. (which is a mistake I will not make again), and during the few months I had a body temp probably a degree celsius higher than usual.
Given the state of dental health over 150 years ago and that a lot of data came from soldiers there may just have been a lot of inflammation going on.
It's one of my issues with biological sciences that they sometimes try to come up with precise answers as they do in physics or chemistry, when most of the time answers, such as this, should be phrased as "around 37C" instead of "it's 37C".
So 37 is +-1C, 40 is +-10, and 36.6 is +-0.1
You can't provide a scaled ranges with the digits alone like that.
Just use ± or a hyphenated range, and take it in context with adjacent measurements.
Yeah, but this doesn't work for human bodies when you round 36.5 either up to 37 or down to 36.
* 38C is surely fever, 37 is mild fever for most people;
* 36C is below-normal for most people, and 35C is surely hypothermia.
36.5 - 37C is the normal temperature range.
The real answer is standard deviation or something like that.
3.00 * 10^3
versus
3 * 10^3
Absolutely not. 47C is an emergency, not "hot".
Or as Population Australia put it:
> The state’s capital city is Perth. Perth has an estimated population of over 2.7 million people.
> Western Australia is Australia’s largest state. Its total area is 2,529,875 km² which accounts for 33 percent of Australia’s total land mass.
> Based on our research Western Australia population will reach 2.879 million by the end of June of 2023.
Most of Western Australia, of course, is not Perth. The density outside of Perth is... 0.2 people per square kilometer, using poorly-defined and chronologically mismatched statistics from the wikipedia sidebars. Perth has 80%+ of the people of Western Australia in 0.24% of the space. If we believe Population Australia about the size of Perth, it has more than 97% of the population of Western Australia in 0.3% of the space.
I feel strangely comfortable characterizing temperatures that prevent humans from living in a region as "emergency" temperatures.
I quite like a nice warm day.
Humans live in warmer climates than that and have done since before the invention of air conditioning.
You might feel uncomfortable if you're not used to it but you're not going to die if you stay hydrated
Or is there really an accurate and consistent calibration technique that's been in use since at least the 1800s, so we know for sure that it must be humans themselves getting cooler?
https://fireinabottle.net/this-is-your-body-temperature-on-v...
I measure my temperature in my armpit using a classic mercury thermometer. I tried an electronic thermometer but found it imprecise.
This way I want to defend 36.6°C as the normal temperature and to abolish 37°C being considered normal as well. Perhaps it can be normal but that would be an individual trait of a specific person. Unless such a temperature is proven normal for this specific person, a person with 37°C armpit temperature should not be allowed to visit the office, attend flights etc.
It's common myth people thinking their kids brain can get fried from infection, which is pretty much impossible, it can get fried from overheating in hot enviroment, but it's pretty much impossible just from infection.
https://www.wired.com/story/fungi-climate-change-medicine-he...
Edit: Found a link:
https://www.osfhealthcare.org/blog/whats-considered-a-fever-...
100.4 is their threshold.
I wonder if this can explain rising worldwide rates of obesity.