Typing Ergonomics: Set Up Your Desk to Prevent RSI

Keep your wrists straight and level, put the keyboard just below elbow height, set the top of the screen at eye level, and take short breaks often. The UK Health and Safety Executive gives these points in its workplace guidance, and this guide measures each one against the research. The equipment most people buy for typing helps less than product pages suggest. Wrist rests, ergonomic keyboards and workstation changes have weak or inconclusive evidence for preventing symptoms, while breaks have the clearest signal, and the trial we found that compared break intervals reported that the shorter interval reduced discomfort most. Fewer keystrokes is the change you control most directly.

Work-related musculoskeletal disorders are common. In Great Britain in 2024/25, an estimated 511,000 workers had one ¹, and 41 percent of those cases mainly affected the upper limbs or neck ². Typing alone has not been proven to cause carpal tunnel syndrome. A systematic review found insufficient evidence that computer work causes carpal tunnel syndrome ³, and MedlinePlus says studies have not proved that typing, mouse use or other repeated movements cause it ⁴. Carpal tunnel syndrome is excessive pressure on the median nerve at the wrist ⁵. The goal of a desk setup is therefore to keep that pressure low while you work, not to prove that a particular chair or keyboard prevents a diagnosis. For the symptoms themselves, see our guides on RSI from typing and wrist pain from typing.

A neutral wrist, straight and level, is the goal of every setup

A neutral wrist means your hand continues the line of the forearm, without bending it up, down or sideways. It is the posture that official guidance describes, and it is the one the laboratory evidence favours. In a 2008 study of 20 healthy volunteers ⁶, typing raised carpal tunnel pressure compared with holding the same posture still ⁷. The same study found that bending the wrist in extension or toward the thumb side was associated with higher carpal tunnel pressure ⁸.

Two cautions. The study measured pressure during typing in a laboratory, not symptoms over months or years, so it explains why a straight wrist is the right target without proving that it prevents injury. The HSE's setup guidance does not give a wrist angle, so a practical test is our own rule of thumb: if the back of your wrist visibly bends up or down while you type, check whether the keyboard is too high, too far away or at the wrong slope.

Put the keyboard just below elbow height and the top of the screen at eye level

The UK Health and Safety Executive's standard setup gives two rules that set the height of the whole desk. The keyboard sits just below elbow height, with shoulders relaxed ⁹. The top of the screen sits level with your eyes, about an arm's length away ¹⁰. If the keyboard is too high, your shoulders rise and your wrists are more likely to bend up; if the screen is too low, you tilt your head forward. Those are the two common failure points, and the rules are designed to avoid both.

Use these as a numbered check whenever you set up a desk:

1. Sit so that your elbows are about level with the keyboard or slightly above it, with your shoulders down and not shrugged.
2. Lower the keyboard, or raise the chair, until the keyboard is just below elbow height and your wrists are straight.
3. Raise the screen, using a stand or a stack of books, until the top of the screen is level with your eyes.
4. Put the screen about an arm's length away so that you do not lean forward to read it.
5. If your wrists still bend up when the keyboard is at the right height, add a slope adjustment or a thicker chair cushion rather than a higher keyboard.

Workstation changes are not a guaranteed fix. A randomised trial of 376 people who used keyboards for more than 15 hours a week tested two workstation and posture programmes over six months ¹¹. The trial concluded that these programmes are unlikely to reduce the risk of upper limb symptoms ¹², and only 25 to 38 percent of people managed to follow every part of them, mostly because their workstations could not be set up to fit ¹³. A 2018 Cochrane review found only one study each of workstation adjustment and of sit-stand desks, and neither had an effect on upper limb pain or discomfort compared with no intervention ¹⁴. These results do not say that a good setup is useless. They say that a setup is a starting point, and that the trials could not show a large benefit from changing it.

Laptops force a bad compromise unless you add a separate keyboard

A laptop fixes the screen and the keyboard to one hinge, so you cannot meet both HSE rules at once on a desk. If the screen is at eye level, the keyboard sits too high and too far forward. If the keyboard is at elbow height, the screen is too low and the neck bends. The only way out is to separate them: raise the laptop on a stand so its screen is at eye level, and type on a separate keyboard set just below elbow height, with a separate mouse or trackpad.

We could not find a trial that compares a laptop with a separate keyboard on symptoms, so treat this as a setup rule derived from the two HSE rules rather than a measured result.

Split and tented keyboards change wrist posture, but the evidence on symptoms is mixed

Split and tented keyboards change the angle of the hands, and the lab evidence on that angle is consistent. In a 2001 study, 11 experienced operators typed on a split keyboard set up conventionally and in three split configurations ¹⁵. All three split configurations reduced ulnar deviation, the bend of the wrist toward the little finger, compared with the conventional setup ¹⁶. That is a measure of posture, not of symptoms. The three configurations did not differ from each other in ulnar deviation ¹⁷, so the practical question is whether a split layout suits your hands, not which exact angle you pick.

The evidence on symptoms is much thinner. A 2022 Cochrane review found only two trials with 105 participants that compared ergonomic keyboards with placebo keyboards ¹⁸. In one small trial of 25 people, an ergonomic keyboard reduced pain after 12 weeks, but the other trial found no difference in pain at six months ¹⁹. The review concluded that there was insufficient evidence from randomised trials to say whether ergonomic equipment is beneficial or harmful for carpal tunnel syndrome ²⁰.

So a tented or split keyboard is a reasonable choice if it lets you keep your wrists straighter and it feels comfortable after a week of use. It is not proven to prevent RSI, and it does not replace the height and break rules above. For a guide to the research on each design, see ergonomic keyboards compared.

Wrist rests are for pauses, not for leaning on while you type

No trial we found tests a wrist rest against symptoms. The Cochrane review of ergonomic equipment for carpal tunnel found none that compared equipment with no treatment ²⁰, and the 2018 Cochrane review on office workers found only inconsistent evidence for arm supports, which were studied for the mouse arm rather than the keyboard wrist ²¹.

The mechanism is the reason to be careful. Wrist extension was associated with higher carpal tunnel pressure ⁸, so a rest that makes you bend the wrist back while you press keys works against the goal. The practical rule that follows from this is to rest the heel of the hand between bursts of typing and to keep the wrist level while the keys are being pressed. This is our reading of the mechanism, not a finding from a trial of rests. For the wider question, see do wrist rests help carpal tunnel.

Short, frequent breaks matter more than any single piece of equipment

Breaks are the one intervention with a signal in the trials. The HSE advises that 5 to 10 minutes every hour is better than 20 minutes every two hours ²². It also says there is no legal guidance on how often breaks should be ²², and that the right pattern depends on the work.

A randomised study of computer keyboard workers compared microbreaks taken at their own discretion, every 20 minutes and every 40 minutes. Microbreaks reduced perceived discomfort in every area measured, and the benefit was greatest with breaks every 20 minutes ²³. A 2003 trial of software that prompted breaks over eight weeks found that 55 percent of workers in the intervention groups reported recovery from their complaints, compared with 34 percent in the control group ²⁴. The same trial found that severity and frequency of the complaints did not change significantly between the groups ²⁴, so the recovery figure is the strongest part of the result, not the whole story.

The 2018 Cochrane review on office workers reached the same cautious place. It found very low-quality evidence of an effect of supplementary breaks on upper limb discomfort ²⁵, and it found no evidence of an effect from training and multifaceted programmes. So breaks are the best-supported habit, but the evidence is not strong enough to set a precise rule. Use the HSE pattern as the baseline, and use the 20-minute interval from the trial if you want a more frequent schedule. For the broader question of how often to stop, see how often to take typing breaks.

The changes ranked by how well the evidence supports them

Not every change has the same evidence behind it. Official guidance tells you what a good setup looks like, but guidance is not the same as a trial that tests whether a change prevents symptoms. The table separates the two.

Read the strength column with care. "Guidance only" means the advice is official but was not tested on outcomes. "Mechanism only" means the change follows from a measured effect, such as pressure, without a trial of symptoms. Neither is a reason to skip a change, but neither justifies buying equipment on the promise of prevention. Where the Cochrane reviews say "very low-quality" or "insufficient", that is their rating of how much the trials can support, not a verdict that the change fails.

Fewer keystrokes is the change you control most directly

Every keystroke adds load to the hand, and typing itself raises carpal tunnel pressure ⁷. The most direct way to reduce that load is to type less. No trial we found measures symptoms before and after cutting keystrokes, so this is reasoning from the mechanism, not a measured result. It is still the change you control completely, and it does not depend on buying anything.

The savings can be concrete. Using the speaking and typing rates from a controlled phone study, a 1,000-word draft saves about 12.7 minutes when it is spoken rather than typed on a phone ²⁶. A text expander saves less per use but adds up: replacing a 60-character sign-off with a short abbreviation saves about 13 seconds each time, at a keyboard speed of 51.56 words per minute ²⁷. That figure does not include the time spent finding the expansion. Dictation replaces typing with speech. Yap, which we make, is an on-device dictation app for Mac and Windows, and it is one option. The guide on dictation for RSI and carpal tunnel covers when it helps and when it does not. If you type a lot without looking at the keys, touch typing is worth a look, though it is not a substitute for the setup rules.

Five checks cover most of the advice

Once a week, run through this list. Each check points back to the evidence above.

1. Measure the keyboard height against your elbows, and confirm that your wrists stay level while you type ⁹.
2. Check that the top of the screen is level with your eyes and about an arm's length away ¹⁰.
3. Set a timer for a short break about every 20 minutes ²³, or stand up for 5 to 10 minutes every hour ²².
4. If you use a laptop for long sessions, add a separate keyboard and raise the screen.
5. Count the phrases you retype every day, and replace the longest with an expansion or a dictated draft.

When a desk setup is not enough, see a clinician

Numbness, tingling, weakness or pain that does not settle needs a clinician, not a new keyboard. The trials above describe groups of people, and none of them can tell you what is causing your symptoms. Our guides on carpal tunnel syndrome and carpal tunnel treatment explain the options, and the definition of repetitive strain injury explains the wider condition. For broader support at work, see dictation as a workplace accommodation, and for the whole subject, see our accessibility overview.