When your back suddenly goes out, the first instinct is usually to lie down for a few days. Before I looked into this, I assumed that was at least the safe option, even if it was not the fastest one.
Then I went looking for trials that actually randomised people between the two pieces of advice. The results turned out to be more delicate than the usual telling of this story.
① Being told to lie down was the standard advice for a long time
Bed rest was the default prescription for acute low back pain for decades. Not a few hours, but days of it, getting up only for the bathroom.
What unsettled that was not a theory but a trial. In 1995, researchers in Finland randomised 186 employees of the city of Helsinki into three groups.
▸ Two days of bed rest (67 people)
▸ Back-mobilising exercises (52 people)
▸ Carrying on with ordinary activities as far as the pain allowed (67 people)
At three weeks and again at twelve, the third group had recovered best. That held for how long the pain lasted, how intense it was, disability scores, and days off work.
The slowest recovery was in the bed rest group.
② But not every trial gave the same answer
Writing "so lying down sets you back" on the strength of one trial would be inaccurate. A larger French trial in 2002 compared four days of bed rest with normal activity in 281 people, and pain and function came out similar in both groups at one week, one month, and three months.
What did differ was something else. Of those prescribed bed rest, 86% took initial sick leave, against 52% in the normal activity group.
They did not end up in less pain. They just spent more days off.
③ Pooling ten trials — the direction favours activity, the size is small
A Cochrane review pulled these trials together in 2010. The review covered ten trials and 1,923 people in total, but the pooled estimate for pain and function in acute low back pain rested on two of them, 401 people.
Staying active came out slightly ahead. Pain 0.22 (95% confidence interval 0.02 to 0.41) and function 0.29 (0.09 to 0.49) — a clear direction, a small size, and a lower bound sitting almost on zero.
The authors graded this as moderate certainty and added that further research is very likely to change the estimate.
So this post cannot get as far as "moving speeds up recovery". What it can say is that the advice to spend several days in bed is not backed by the evidence.
④ Pain that travels down the leg was a different story
In the same review, sciatica — pain running down below the buttock — came out differently. Pain was -0.03 (-0.24 to 0.18) and function 0.19 (-0.02 to 0.41), so no difference between the two pieces of advice could be detected.
That is not the same as "you may as well lie down". It means this comparison cannot tell you which is better.
And if pain running down your leg is new, or getting worse, there is something to check before you pick between rest and activity. I wrote about that separately.
→
barosit.com/community/p/e24258f4-8806-4275-b65f-5452a93cc58f
⑤ "Stay active" does not mean "go and exercise"
This is worth pausing on. In the Finnish trial, the group assigned back exercises did worse than the group who simply carried on. The Cochrane review likewise found no clear difference between staying active, exercise, and physiotherapy, though the certainty there was low.
So what the evidence points to is not "start exercising when it hurts". The instruction the trials actually gave was far more modest — keep doing what you were doing, as far as the pain allows.
Current guidelines sit in the same place. A 2026 review of low back pain guidelines worldwide found that ten of the seventeen guidelines covering non-drug care for acute low back pain recommended self-management, centred on staying active, including at work.
⑥ So if your back hurts right now
▸ If you have been lying down for a day or two — that is not a mistake. What lacks support is stretching it into several days.
▸ If the pain has you doing nothing at all — the test is what the pain allows. Restarting with whatever you can tolerate from your ordinary routine is exactly what the trials asked people to do.
▸ If there is pain or numbness running down your leg — this comparison does not answer your question. The post linked above is the better place to start.
Changing the question helps. Not "should I rest or move?" but "how far can I go right now without making this worse?"
One more thing from the same guideline review: of the thirteen guidelines that covered prevention, ten recommended regular physical activity. What to do while it hurts and what to do so it happens less often are different questions, and the answer to the second one is more consistent.
BaroSit works on that second question. It notices when you have been in one position too long and says so — what to do while you are in pain is not something this app can answer.
→ Have a look at
barosit.com if you are curious.
The evidence on sitting time and movement is collected at
barosit.com/science.
Sources
• Dahm, Brurberg, Jamtvedt & Hagen, 2010 · Cochrane Database Syst Rev (6):CD007612 — ten randomised trials, 1,923 people. For acute low back pain, advice to stay active gave standardised mean differences of 0.22 for pain (95% CI 0.02 to 0.41) and 0.29 for function (0.09 to 0.49), from two trials totalling 401 people, at moderate certainty. For sciatica, pain -0.03 (-0.24 to 0.18) and function 0.19 (-0.02 to 0.41), no difference. Comparisons with exercise and physiotherapy were low certainty. The authors state that further research is very likely to change the estimate of effect
• Malmivaara et al., 1995 · N Engl J Med 332(6):351–355 — 186 municipal employees (two days of bed rest 67, back-mobilising exercises 52, ordinary activity as tolerated 67). The ordinary activity group did better at three and twelve weeks on pain duration, pain intensity, disability score, and days absent. Recovery was slowest in the bed rest group
• Rozenberg et al., 2002 · Spine 27(14):1487–1493 — 281 people, four days of bed rest versus normal activity. Pain and function were equivalent at 6–7 days, one month, and three months. Initial sick leave was 86% versus 52%. People with pain radiating below the buttocks and work-related injuries were excluded
• Oliveira, Koes, Pinto et al., 2026 · Lancet Rheumatol 8(6):e470–e485 — a review of 32 clinical practice guidelines worldwide. Ten of the seventeen guidelines covering non-drug care for acute low back pain recommended self-management, mainly advice to remain active including at work. Ten of the thirteen covering prevention recommended regular physical activity
This post is about acute low back pain of less than six weeks with no identified specific cause. It does not carry over to chronic back pain or neck pain. The literature search behind the Cochrane review above ran to 2009 and no updated version has been published since, so the evidence is old and I would rather say so.
This is general information, not medical advice. If the pain is getting worse or dragging on, or if you have weakness in a leg, trouble controlling your bladder or bowels, a fever, or a fall or other injury behind it, please see a doctor without delay.