Neck pain: it is probably not your posture
The desk gets the blame and the evidence does not support it. Here is a test you can do on the floor, and the thing that does work.
Get the one-page sheetWhat it is
Neck pain with no injury behind it, building through the day and easing overnight. It is the commonest thing we see in people who work at a screen and it is rarely dangerous.
Most of it settles a long way in the first six or seven weeks. What catches people out is the tail. Measured across large groups, pain had dropped by about half by week seven and then barely moved for the rest of the year. That is not a reason to worry. It is a reason not to expect a clean finish by Friday.
The posture question
You have been told to sit up straight. The research has not been kind to that advice.
When more than a thousand seventeen year olds had their sitting posture measured and were then followed up, how they sat made no difference to whether they got neck pain or headaches. In office workers tracked over time, the only two things that consistently predicted new neck pain were being female and having had neck pain before. Posture did not make the list.
Text neck is in the same position. Twenty one studies have gone looking for a causal link between phone posture and neck pain. None of them found one.
That does not mean sitting still all day is fine. Low variety in what you do at work is a risk factor in its own right. The problem is staying in one shape, not which shape it is.
What helps
Strengthen it rather than stretch it. Pooled across the trials, strengthening the neck, shoulders and upper back helped. Stretching on its own did not.
The dose is smaller than you would think. Two minutes a day of band work, five days a week for ten weeks, cut headache frequency by more than forty per cent in office workers with neck and shoulder pain. Twelve minutes a day did a little better. Two minutes is not nothing.
Change what you are doing, not how you sit. Get up, take a call standing, switch tasks. Variety beats posture.
Keep going past the point it eases. The trials that worked ran for ten to twelve weeks, and this tends to come back in people who stop the moment it feels better.
What to stop spending on
A new chair, a standing desk, an ergonomic gadget. Pooled across the trials, workstation adjustment and sit-stand desks did not reduce neck pain.
A scan, as a first move. Close to nine in ten people with no neck symptoms have disc bulging on an MRI. It will find something, and finding it rarely changes what we do next.
Try this first
- Do the hold test with someone and write both numbers down
- Two minutes of band work for the neck and upper back, most days
- Change what you are doing every half hour, rather than how you sit
- Give it ten weeks before you judge whether it is working
Get it looked at if
- Clumsy hands, dropping things, or your walking has changed
- Electric shock sensations down the spine when you bend forward
- Dizziness, double vision, slurred speech or facial numbness, same day
- Neck pain after a fall or a collision
- A history of cancer, weight loss, or pain that is worse at night
What we would do
We measure it rather than guess: the hold test, how far you turn each way, and what the shoulder blades are doing. You leave with a programme built on those numbers and a date to retest.
Want this as a one-page sheet?
The version we hand out in clinic. Print it, stick it on the fridge, work through it.
Related: Neck pain physiotherapy · All patient guides · Our blog
This guide is general information about a common problem, not a diagnosis of yours. If something is not settling, or it does not match what is written here, come and get it looked at.