I've never been a fan of the daylight colors and this just goes to prove why you should not use it.  2700-4100K or bust, baby! 
https://medicalxpress.com/news/2019-05-eyes-health-authority.html

I've never been a fan of cold light in general, specially with CFL and LED. My eyes feels weird with almost all LEDs and CFLs, even with some warm white ones. Because of that I prefer incandescent, halogen or fluorescent for home use.

I, unfortunately, had made that mistake when I chose 5000K LEDs for almost the entire house (except for the bathrooms, kitchen, and my room).

Now, because of LEDs dying and a clearance deal, most of my home lighting is now 2700K and 3000K.

I, unfortunately, had made that mistake when I chose 5000K LEDs for almost the entire house (except for the bathrooms, kitchen, and my room).

Now, because of LEDs dying and a clearance deal, most of my home lighting is now 2700K and 3000K.

The eye damage has not to do with the color temperature being 5000K. It can do the damage even if it is 6500K like most daylight LED lamps. Most LED lamps don't have actual CCT at all, as their light is outside the blackbody locus.

Wait 'till the author of that article finds out what color temperature daylight is. The lights at work are also 5000-6000k, and of much higher higher intensity than a living room table lamp. Also, most modern TVs, computers, and phones. I'm relatively certain the dose from a day at the beach would dwarf even a year's worth of a LED in the home.

What am I supposed to do - close the blinds, never go outside, and light my home with candles? 

I've never been a fan of the daylight colors and this just goes to prove why you should not use it.  2700-4100K or bust, baby! 
https://medicalxpress.com/news/2019-05-eyes-health-authority.html

But the thing is, there is no difference in this matter from any other source with distinct spectral lines. This includes triphosphor fluorescents, MHs, MV and many other.

The only difference is, the LEDs were the first one entering homes en masse.
Mainly bedroom areas, where this problem applies.

The spectral power distribution of LED (driven by blue LED chip at 450nm) delivers higher percentage of power at the blue light hazard band than any discharge light source, at least for comparable CCT source (Not including special lamps). The discharge light sources simply have more spectral bands, outside of the 450nm band, that the power divides between....

Discharge lighting had been used at workplaces for nearly a century with no adverse health effects related to the light spectrum. People might have not liked it's visual appeal, but actual headaches etc. had been caused by flicker, not by the spectrum

To this i want to add, that i have read the most complaints about FL headaches from US based sources - Which leads me to think of a possibility (at least for some of the cases), that the cause might be not the 120Hz flicker of an intact lamp, but 60 Hz rectification flicker of failing lamps on RS ballasts. As RS is the only magnetic FL ballast type (in common use) that can keep a rectifying lamp lighting continuously, so the user might keep being under the lamp without realising that the lamp is EOL and needs replacement

The spectral power distribution of LED (driven by blue LED chip at 450nm) delivers higher percentage of power at the blue light hazard band than any discharge light source, at least for comparable CCT source (Not including special lamps). The discharge light sources simply have more spectral bands, outside of the 450nm band, that the power divides between....

Discharge lighting had been used at workplaces for nearly a century with no adverse health effects related to the light spectrum. People might have not liked it's visual appeal, but actual headaches etc. had been caused by flicker, not by the spectrum

The spectrum related problems do not come from work places, but from relaxation places. And it depends on how much of the blue is excited - if too much, the brain "thinks" it is mid day, so maintains you awake. Good for working (it keeps you awake and alert in a natural way, like the daylight does), but very bad if you are about to go to sleep (when the 1000's of years of human fire use made us "programmed" so the red/oange warm glow without any blue means home cave fire safety for a good sleep, but bluish moonlight means "you are unprotected in the open so need to be ready for fight or flight").


Accusing the LEDs of being anything different is just a hoax, originated when only very high CCT LEDs were available for lighting and when these were used by some as bed lights (which is wrong; the hoax part is the generalizing, to either all LEDs, or to all activities).

The part of power LEDs emit around the 450nm band exceeds that of any other light source, including all 4000...6500K discharge lighting and direct sunlight. In the working hours of the day it is not a matter of sync of the biological clock as much, but of retinal damage that would accumulate over years of exposure

The part of power LEDs emit around the 450nm band exceeds that of any other light source, including all 4000...6500K discharge lighting and direct sunlight. In the working hours of the day it is not a matter of sync of the biological clock as much, but of retinal damage that would accumulate over years of exposure

It does not matter if it is exactly 450 or 480 or 420. All have the same effects and the effect is related to the sum of the power in the whole blue band. And you need to emit in the blue band if you want the high CCT, regardless of the lamp chemistry. So nothing is special to "only LED".

The LED does have everythng in a single line, but so does e.g. the MV or many MH, but it stlii plays no effect. Everything in the blue band, regardless if concentrated in one line or spread over the whole band in a continuum, has the same strong effect for the "biological clock".
And if that effect is beneficial vs detrimental to humans health depends on how the exposure to the light strong in blue is synchronized with your intended activity cycle.

Nothing from this is any new in any way, it was just not well known in the public in the past, because normal people just did not use any high CCT lights at homes, mainly around bed. The problem came only with first LEDs with very high CCT (so rich in blue), which because of being low intensity and low ciurrent consumption, were tempting to be used as nightlights. And that is the real problem. There was an attempt to notify the public about this problem, but it was worded not in a really clear way to be understood by "an ordinary Joe".
As a result it got completely misunderstood, then misinterpreted over and over and turned into the hoax about the "extremely dangerous LEDs" started spreading.
And this hoax is still being spread, even by people that should have the qualification to understand it correctly if they pay at least a bit of attention. But the laziness (instead of really verifying the basis they just repeat the misinterpretations) and fame hunger (need to be "the one to warn all the others") took over their good judgement instead.

So yes, you should be careful with where to not use daylight LEDs at home and no, it is not LEDs what makes the problem, it is the daylight (high CCT so rich in blue) light color, you should be the same way careful with any other daylight light source.