The 20-20-20 Rule: What It Helps, What It Cannot, and How to Actually Keep It

Every page that scores for the 20-20-20 rule for eyes states, outright, that it works. The most-read page of all uses a heading for this rule. It includes an aside noting that breaks from screen work are beneficial, and doesn't cite a single study. There was a trial looking specifically for this effect, and it didn't find one, and aside from an optometrist trade journal, it doesn't appear anywhere on the first screen of results.

So let's begin here.

The rule as it is asked to be performed is a clinical guideline, not an empirically demonstrated optimum. The time units were not arrived at by experimentation, and the page which most directly addresses them is subtitled "Questioning the 20-20-20 rule" for this very reason.

Table of Contents

• Does the 20-20-20 rule actually work?
• What the rule can honestly do
• What it cannot do
• Why a screen dries an eye
• 30-30-30, 10-10-10, and the rest of the family
• How long does eye strain last?
• What the classical texts actually say about resting the eyes
• The conditions the rule was not written for
• Desk setup, which is duller and probably matters more
• What actually survives a workday

Does the 20-20-20 rule actually work?

The evidence is weak, as shown by the two trials that contradict each other. The first trial changed up the frequency of breaks and found no difference in any of the measured outcomes. The second trial saw improvement in symptoms while the reminder software was active, noted no change in one specific eye examination finding, and observed the benefits vanish within a week of discontinuing the reminders.

Take the negative one first, as no one else does. A 2023 crossover study published in Optometry and Vision Science asked 30 participants to perform a cognitively demanding forty-minute task of reading from a tablet on four separate occasions with twenty-second breaks every 5, 10, 20 or 40 minutes; the last of which effectively meant no breaks at all (PubMed). After the task, the participants who took breaks every 5, 10, 20, or 40 minutes all showed significantly increased symptoms (P < .001). Neither the break schedule had any significant effect on self-reported symptoms (P = .70) or on total task performance, including reading rate (P = .93) and accuracy (P = .55). The authors of this work found no evidence for breaks every 20 minutes being therapeutically beneficial for people suffering from digital eye strain, and furthermore note in the discussion section that there is very little peer-reviewed research on the matter.

Hold that at its real weight: thirty young subjects, one experimental trial, one device, and twenty-second breaks only, which means that the study actually only tested one particular time window described by the rule and cannot extrapolate to longer breaks. There were no objective measurements of eye conditions such as tear film, number of blinks, or accommodation, only self-reported questionnaire items and performance. A negative result in a study of only thirty participants is not enough to dismiss any meaningful clinical effects.

The supportive trial is the one everybody quotes, usually having not read the results section. Published in Contact Lens and Anterior Eye journal in 2023, it followed 29 symptomatic computer users who had laptops with bespoke software tracking them down via the camera and giving individual 20-20-20 reminders (PubMed). Behaviour did change, with the computer sessions being shorter and the number of breaks taken per day increased (p ≤ 0.015), while the dry eye and digital eye strain scores decreased (p ≤ 0.045).

These are the bits nobody talks about, though.

None of the ocular surface or tear film measures changed at all (p ≥ 0.089): not blink rate, not incomplete blinking, not tear meniscus height, not lipid layer thickness, not non-invasive break-up time. Symptom improvement did not last beyond one week after the reminders had stopped (p > 0.05). There was no control arm; all 29 were self-selected symptomatic users who served as their own control, so placebo response, response to being monitored by the software, or regression to the mean cannot be disentangled from the effect of the reminders in this setup. What was tested was not the rule of thumb, but reminder software which incorporated the rule. The paper generated a controversy, with a letter to the editor and a subsequent reply in the same issue.

Their conclusion states it best: two weeks was insufficient to demonstrate lasting benefit for either binocular vision function or objective signs of dry eye.

Place both of these papers one after the other and the answer is no longer debatable. The proposed rule has never changed any observable metric and the only study that varied the break interval by a factor of four found there to be no observable effect on either perceptual performance or the individual's ability to read.

What the rule can honestly do

I see two benefits in the rule. It provides structure to the workday where otherwise many would not have any breaks and provides time for blinking that many would otherwise forgo. Both points relate to a larger issue discussed in the next section.

Comfort is an undeniable effect and one not intrinsically inferior. The reminder trial did improve symptoms while it was occurring, and people notice when an hour feels better. Take this as what it is, an uncontrolled positive finding in the symptom-reduction domain, self-reported by 29 people, with a particular kind of software watching them throughout.

The more interesting property is as a enabler though. Nobody sets alarms to blink correctly and unhook their shoulders, but a break you've already decided on can carry those functions along with it. This is a behavioral design argument, not a clinical one, and it should be presented as such.

What it cannot do

There are four claims that come bundled with this finding that the paper does not support, and they should all be noted as such.

It is not an eyesight improver. There has been no trial examining the effect of breaks every twenty seconds on any measure of refractive error, so a page telling you about this rule is in fact inventing that it sharpens vision. It does not even help short-sightedness, since myopia research relates more to the amount of time spent outside, twenty seconds at a window is no substitute for hours outside.

It does not treat dry eye disease. The supportive trial looking at tear meniscus height, lipid layer, break up time and staining found no difference in any of them. If you have a dry eye condition rather than just tired eyes, unfortunately you have to take a step back and have different considerations, which are covered in our guide to eye drops for dry eyes, and the differences between them.

The three twenties are not scientifically based. They're a nice round number, which is a good quality in a health message, and a good quality of another sort, but not scientifically based.

Why a screen dries an eye

The commonly known fact about blink rate changing at the monitor is not what the study showed, and what it did show was far more interesting than the myth it corrected.

A 2020 pilot study published in Current Eye Research asked 12 young adults to read an excerpt from a novel on a smartphone for 60 min. The number of their blinks was counted every 10 min. (PubMed). The number of incomplete blinks increased from a median of 6 per minute at minute 1 to 15 per minute at minute 60 (p = .0049), while participants whose percentage of incomplete blinks increased the most also reported worse ocular surface symptoms and more tired eyes. The total blink rate increased throughout the 60 minutes, but this rise was not statistically significant (p = .08). Binocular accommodative facility decreased from a median of 11.3 to 7.8 cycles per minute (p = .01).

The study explains that the upper eyelid moves during a blink, but the motion is not completed, and thus the eyelid does not wipe across the surface of the eye to replace the tear film, leaving it partially dry.

The authors have chosen to call this a pilot study and the limitations are telling. Start with the sample size: twelve 18-23 year olds, prescreened to exclude anyone with dry eye or binocular vision issues, who were only tested once, with no controls or post treatment testing. Testing the effect of an intervention on the tear film instability of a group who is statistically least expected to demonstrate any tear-film change at all, in one session involving one intervention with no control group, can provide only anecdotal evidence at best. The link between incomplete blinking and the symptoms is a mere correlation observed in this particular sample.

Of additional interest is the fact that none of the other measures, including tear break up time, lipid layer or tear meniscus appeared to change over the course of the hour, as was also the case, according to the reminder trial, over the course of two weeks.

30-30-30, 10-10-10, and the rest of the family

The variations on the theme of distance – thirty minutes at thirty feet for thirty seconds or ten at ten for ten – are really no variations at all, since the underlying premise of the intervention has not been tested at all and found effective. Thirty feet for thirty seconds is simply an arithmetic mean of the other numbers, which in turn have been derived in turn from an unproven hypothesis.

There is, however, one piece of evidence that speaks directly to the whole family: the crossover trial. It compared breaks every five minutes with every 10, every 20, and none at all across the same task and found no difference between them. A broader range of intervals than the 10-10-10 versus 30-30-30.

That the one about the first number is not the one we should have been arguing about.

The second number gets a mention in passing, 20 feet being the distance at which focusing muscles fully relax, according to physics, and not actual science, with no citation and no consideration that for most people, the distance is closer. Being able to see the whole room is fine, but measuring it is not.

How long does eye strain last?

For most people, eyestrain eases over minutes to hours once the near work stops and is generally more akin to the tiredness of a worked muscle than some form of traumatic injury. What the results of these studies do not show is a recovery time. None of the three tracked participants' symptoms for more than a week after the treatment had ended, and none were designed to test for it.

What the results do show is how quickly such symptoms can emerge, which is right after the reading task in each of the crossovers, including the one using breaks every five minutes.

The question of whether the effects are reversible is at least partially answered by the same observation: that removing the reminder software causes scores to drop back to baseline within a week.

Persistent blur, pain, or symptoms that do not settle overnight are not a screen-break problem and should be looked at by an eye doctor. Uncorrected refractive error is a boring but common cause of eyes that ache at the desk, and no break schedule can correct a faulty prescription.

What the classical texts actually say about resting the eyes

The classical Indian texts allocate the eyes a fixed time in the daily routine rather than a periodic interruption. They also include a cautionary list of objects harmful to the eyes; one of them sounds remarkably like a desk job. Neither text is concerned with screens, and both are best read as practical guides.

The chapter on daily-routine in the Aṣṭāṅga Hṛdaya allocates the eyes a place in the morning ritual, between the toothstick and the nasal drops:

सौवीरमञ्जनं नित्यं हितमक्ष्णोस्ततो भजेत् ।
चक्षुस्तेजोमयं तस्य विशेषात् श्लेष्मतो भयम् ॥ ५ ॥
योजयेत् सप्तरात्रेऽस्रावणार्थं रसाञ्जनम् ।
 
sauvīram añjanaṁ nityaṁ hitam akṣṇos tato bhajet |
cakṣus tejomayaṁ tasya viśeṣāt śleṣmato bhayam || 5 ||
yojayet saptarātre 'srāvaṇārthaṁ rasāñjanam |

After cleaning the teeth, applying sauvīra añjana daily (which is good for the eyes) and every seventh night applying rasāñjana to remove the excess. That is the Sūtrasthāna, Second Adhyāya, 4b–5, as printed on page 25 in the 1939 Nirnaya Sagar edition with the commentary by Arunadatta and Hemādri.

The remarkable thing about this is not the substances but the pattern; one daily and one weekly practice fit into the day. The division between appropriate work and unwinding activities is less precise, but no one in that culture worked for eight hours at a time staring at a glowing rectangle. If you like the idea of fixed slots better than interruptions to whatever you are doing during the day, we can use the Ayurvedic daily eye care as a model.

The closest comparison is at the beginning of the eye section in Suśruta's work, which begins with a list of the various ocular pathologies.

शुक्तारनालाम्लकुलत्थमाष-
निषेवणाद्वेगविनिग्रहाच्च ।
स्वेदादथो धूमनिषेवणाच्च
छर्देर्विघाताद्वमनातियोगात् ॥
बाष्पग्रहात् सूक्ष्मनिरीक्षणाच्च
नेत्रे विकाराञ्जनयन्ति दोषाः ॥ २७ ॥
 
śuktāranālāmlakulatthamāṣa-
niṣevaṇād vegavinigrahāc ca |
svedād atho dhūmaniṣevaṇāc ca
chardervighātād vamanātiyogāt ||
bāṣpagrahāt sūkṣmanirīkṣaṇāc ca
netre vikārāñ janayanti doṣāḥ || 27 ||

Coming from soured and fermented products, horse gram and black gram; from the suppression of natural urges; from sweat; from an overexposure to smoke; from vomiting that is suppressed or performed in an excessive manner; from a stopped flow of tears, bāṣpagraha; and from continuous observation of small objects, sūkṣmanirīkṣaṇa. Coming from these factors, the verse says, the doṣas generate eye diseases. The verse cited just before this one on the same page also mentions plunging into cold water when heated up, observing things from a distance for too long, and sleeping incorrectly.

Suśruta Saṃhitā with the Āyurveda-tattva-sandīpikā Hindi commentary of Kaviraja Ambikadutta Shastri, Volume 2, Uttaratantra 1.27, printed folio 11.

Ambikadutta Shastri, while glossing over the word सूक्ष्मनिरीक्षण on that page, offers an occupational example of his own: घड़ीसाजी, watch-repair. Close, prolonged, detailed inspection as a livelihood. The screen is the latest addition to this list, which otherwise consists of old items. It is a parallel, rather than a projection. Two of the items on that list require suspension of tears and sleep, respectively, to perform in the contemporary workplace.

The conditions the rule was not written for

The 20-20-20 rule was not an Ayurvedic one, but arrived later, through office HR manuals and eye hospital brochures, and presumes a workspace that many readers of this article do not have.

Consider the actual day: an air conditioned floor heating dry air for eight hours, a window that won't open, and a monitor on a desk you didn't pick. Then a traffic jam where you can't get far, phone six inches closer than the monitor was, cranked up to maximum brightness to counteract the road lights. Then a warm dry evening at home.

None of that is in the trials, conducted in temperate laboratories. The reminder study's entire treatment was watching a laptop as someone else worked; there's no version where your eyes are crossing trying to make sense of the text through the dry air of a cabin at 24 degrees centigrade with a vent right above your head. During the hot months the dryness becomes an issue on its own, which is why we separately discussed the summer heating.

The mid-range Android needs its own post. Holding one's phone at arm's length is a significantly more effortful distance than keeping the monitor at 60 cm. Most reading in this country is done on the former.

Desk setup, which is duller and probably matters more

Screen position, anti-glare, and text size are all largely conventional rather than empirically dictated, and should be treated as such. They are also free, require no reminder, and never degrade as your working memory.

Mount the monitor so that the top of it is at eye level or lower, meaning your lids will have less travel to lower over the eye and reveal less of the window. Sit at arms length. Kill the glare from the window behind you before fiddling with the monitor's brightness controls and make the text larger than you think.

That last one is the one everyone can do right now, and which they're least likely to try.

What actually survives a workday

Nobody wants a twenty-minute timer. The reminder trial worked only because software was watching, and the benefits vanished after a week, which is the sort of insight that makes more sense to pursue than most findings in the paper on real-world behavior change.

Forget about running a clock; tether breaks to events that are already in your day. When a call ends, when the build finishes, when the kettle whistles, when your walk to fill the water bottle begins. There's no point in using an app to mark boundaries that are already there.

Make the breaks you take long enough to feel like a break (a minute at the window is not much but a minute at the window is something). Since twenty seconds has no evidence in its favor and was not supported in the trial that tested it, there is no reason to take breaks of any particular length.

The one action that the smartphone study actually evaluated is blinking properly while you are on the phone, with the lids fully closing over the eyes. That's it, and it's free.

If instead, by the end of the day, you feel less tired and more dry, that's a complaint this rule doesn't actually address. It deals with a different problem, one that responds to a cooler room, reduced exposure to the drying effects of the air blowing on your face, and something that soothes and relieves dryness, which is all our I-Care Eye Drops are made to do.

The rest is simply housekeeping; the rule is mostly sensible, loosely supported, easy to remember, but still completely lacking in any actual impact on the real-world physiology of your eyes. All in all, it's worth remembering, but not worth a single rupee.