Hanging from a pull-up bar for back pain is one of the most widespread tips online: you hook on, let go, gravity "decompresses" the discs and the pain goes away. It is a convincing image. The problem is that the central part of that promise, the claim that traction makes back pain go away, has been studied a great deal, and the results are not what the story suggests.
This article does three things. It tells you what physically happens when you hang and what has not been shown. It reports the evidence on traction for low back pain and sciatica as it stands, uncertainties included. And it explains what hanging is really good for, because there is an answer, it is just not the one you might expect.
Important before continuing. The exercises and advice below are general guidelines and may not be suitable for your situation. If you want a professional's opinion, you can start with the questionnaire: you register and see who will follow you, a physiotherapist registered with the Albo chosen for your case, whom you can write to in chat. If you would rather hear their voice, the first consultation is free.
First of all: the signs that do not wait
The rest of the article is about exercise and patience. There are however pictures in which looking for the right exercise is exactly the wrong move, and they need reading now, not at the end.
Go to the emergency department immediately if even one of these appears:
- difficulty urinating, retention, or loss of control of bladder or bowel;
- numbness in the "saddle" area, that is the perineum, genitals and inner thighs;
- weakness or tingling that appears in both legs.
These are the possible signs of cauda equina syndrome. It is rare, but it is a genuine emergency, and how quickly it is treated affects recovery.
Contact a doctor within a few days if you have:
- a loss of strength that is getting worse over the course of days: a foot that gives way, difficulty climbing stairs, walking on your heels or on your toes;
- fever together with back pain;
- pain that appeared after a major trauma, such as a fall or an accident, especially over the age of 50 or with known osteoporosis;
- a personal history of cancer;
- weight loss that you cannot explain;
- pain that progressively worsens even at rest or at night.
A review published in the European Spine Journal compared 16 guidelines from 15 countries and found 46 distinct red flags, grouped into four categories: neoplasia, fracture, cauda equina syndrome and infection. The authors also note an honest limitation: diagnostic accuracy data for these signs are rarely reported.
These are not infallible tests; they are reasons to have someone look at you.
If any of these points applies to you, stop looking for the right exercise and book an appointment. Hanging is not the answer, and in at least one case it can make things worse: we come back to this shortly.
If any of these signs applies to you, the urgent symptoms page explains where to go and how quickly.
What really happens when you hang
The spine lengthens: this part is true
Your height changes over the course of the day, and this is a measured fact. In a study published in Spine in 1985 on eight young adults, the average circadian variation in height was 19.3 mm, that is 1.1% of height. So yes: in the evening you are about two centimetres shorter than in the morning, because the intervertebral discs lose water under load.
The same study, however, contains two details that the popular story always skips. The first: 54% of the loss in height happens in the first hour after waking. It is not a slow build-up that accumulates during the sedentary day; it is a very rapid phenomenon, largely complete by the time you sit down at your desk.
The second: about 70% of the height is recovered in the first half of the night, simply by lying down.
Put directly, nocturnal decompression happens on its own, free of charge, every night. You do not need a bar to get it.
The logical leap nobody declares
When you hang, the weight of the pelvis and legs does apply a traction force to the lumbar spine. So far the physics holds.
The logical leap comes afterwards, in three steps taken for granted: that this force reaches the lumbar discs to a significant degree, that the separation achieved lasts beyond the seconds you remain hanging, and above all that a momentarily "longer" spine is a spine that hurts less.
The last step is the one that matters, and it is also the only one that can be checked directly: take people with back pain, apply traction, and measure the pain. That is exactly what has been done, dozens of times.
What studies say about traction
The Cochrane review: no meaningful difference
The Cochrane review of traction for low back pain with or without sciatica included 32 randomised trials with a total of 2,762 participants, considering 16 of these studies, equal to 57% of the participants, to be at low risk of bias. It is not a small body of literature.
The authors' conclusions are clear: traction, alone or combined with other treatments, has little or no effect on pain intensity, functional status, global perceived improvement and return to work. For people with low back pain and sciatica, with evidence of low to moderate quality, traction probably makes no difference to pain, function or global improvement.
For chronic low back pain without sciatica, with evidence of moderate quality, traction probably makes no difference to pain compared with simulated treatment.
The same review also reports an uncomfortable finding for anyone who presents traction as harmless: adverse effects appear in 7 of the 32 included studies, and include increased pain, worsening of neurological signs and subsequent recourse to surgery. Four studies, by contrast, reported no adverse effects.
The picture has not changed over time.
The 2025 Cochrane overview, which re-examined 31 Cochrane reviews covering a total of 644 studies and 97,183 adults with low back pain, concludes that in chronic low back pain there is probably no difference between traction and simulated traction on pain intensity, with a mean difference of 4 points on a 0 to 100 scale and a confidence interval running from -17.7 to 9.7.
That is: the uncertainty is larger than the effect.
The case of sciatica: where the evidence is weaker, not stronger
This is the part that deserves attention, because this is where the biggest promises circulate.
A systematic review with meta-analysis published in Physical Therapy in 2021 examined mechanical traction in lumbar radiculopathy, including 8 studies and analysing 5 of them.
The result, read in full, is this: supine traction added to physiotherapy shows significant effects on pain and disability, but those results come from low-quality studies; when the authors look at the high-quality studies, on prone traction, the effects on pain and disability are not significant.
And everything that is seen concerns, in any case, the short term, up to three months.
There is also a review that concerns exactly the movement we are talking about, because hanging is, technically, vertical traction. It found just three randomised trials, too heterogeneous to be combined in a meta-analysis. Vertical traction added to bed rest shows a large effect, but with evidence of very low quality and a confidence interval that nearly touches zero.
Added to medication, a large effect with low-quality evidence. And added to physiotherapy, that is the comparison that most resembles the real question, the effect is tiny: g equal to -0.14, with an interval from -1.03 to 0.76, so compatible even with a worsening.
The authors write that the long-term effects are unknown and that higher-quality research will very probably change these conclusions.
It is not "traction works for sciatica". It is "on three weak studies, perhaps something in the short term, and when you compare it with normal physiotherapy it adds nothing".
And the herniation? It shrinks anyway
A double-blind randomised trial compared a course of physiotherapy with and without non-surgical spinal decompression in patients with a lumbar disc herniation, measuring disc height and herniation thickness on MRI. Both groups improved in pain, function and the size of the herniation. No significant differences emerged between the two groups.
It should be said that there were only 20 patients, so the study is small, but the direction is consistent with everything else.
The point is a different one, and it has been documented for a long time: most radiculopathies caused by disc herniation resolve spontaneously, without surgery. The herniation shrinks on its own, over weeks or months. If you hang every day during those months, in the end you will credit the bar. It is an error of attribution, not evidence.
Why so many swear it works
Because the relief, often, is real. It is just short-lived and comes from somewhere else.
Staying suspended changes the position of the pelvis, lengthens the lumbar and back muscles, interrupts the posture you have been in for hours and forces you to breathe more slowly. Many people come down from the bar feeling better. This is an experiential fact, and there is no reason to deny it: that would be the opposite extreme, equally dishonest.
What does not hold up is the next step, that is "I feel better for thirty seconds, so it is treating my back". Systematic reviews exist precisely to distinguish the two things, and they are clear on this distinction.
A global comparison of 22 recent clinical guidelines on low back pain shows what is recommended in practice for the chronic form: therapeutic exercise, non-steroidal anti-inflammatory drugs, spinal manipulation and acupuncture. Traction is not among them.
If you are looking for the lever that really moves the needle, the same 2025 Cochrane overview points to it: exercise-based therapies, compared with no treatment or usual care, reduce pain in chronic low back pain by 15.2 points out of 100, with moderate-certainty evidence from 35 studies and 2,746 participants.
And in the acute phase, advice to stay active beats advice to rest, both for pain and for function. It is not as spectacular as "I decompress my discs", but it is what holds up under scrutiny.
What hanging is really good for
Now the good part, because there is one.
For grip strength, which is a serious indicator of health. A prospective study of 502,293 participants in the UK Biobank, published in the BMJ, found that every 5 kg less of grip strength corresponds to a higher risk of all-cause mortality: hazard ratio 1.20 in women and 1.16 in men.
Similar associations have emerged for cardiovascular and respiratory diseases and several cancers. Be careful, however, in how this figure is read: it is an association, and grip is above all a marker of overall strength and health, not a switch. Training your grip does not lengthen your life in itself.
That said, hanging is one of the simplest ways to train it, and with age it is a capacity that is lost quickly.
For shoulder mobility and shoulder health. Raising the arms above the head and holding a load there is a movement that many people have stopped doing altogether. The dead hang puts the shoulder into full flexion under tension, involves the scapulae, latissimus dorsi and the shoulder girdle muscles, and maintains a range of motion that a desk-based life silently erodes.
If you introduce it gradually and in the absence of pain, it is a reasonable exercise for the shoulder, much more so than for the back.
For feeling better right now, knowing that is what it is. As a sensation of stretching at the end of the day, it is worth about as much as a good stretch: pleasant, temporary, with no therapeutic claims. If you frame it that way, it is not harming you. If you use it instead of having a weakening leg assessed, then it is.
Who should not hang
This list is not a formality and should be taken seriously.
- Shoulder instability or previous dislocations. Hanging with straight arms puts the joint in a position of maximum traction: it is one of the worst conditions for a shoulder that tends to come out.
- Symptomatic rotator cuff. If you have active pain raising your arm or at night on your side, suspended load is not the place to start. A targeted progressive programme is needed first.
- Severe osteoporosis or known vertebral fractures. A fragile spine should not be subjected to improvised traction, and the risk of falling when coming down from the bar is real.
- Advanced pregnancy. It changes the centre of gravity, changes ligament laxity, and the grip gives way sooner than you expect.
- Uncontrolled hypertension, or significant cardiovascular problems. This applies all the more if you are thinking of inverted variations.
- Any neurological symptom that appears or worsens while hanging. Tingling, electric shocks, weakness: come down and get assessed, do not push on.
A final note on a close relative that is often confused with the dead hang: the inversion table. It is not the same thing, but it says something useful.
An analysis published in Disability and Rehabilitation examined the FDA database of medical device reports and documents serious injuries associated with inversion, including spinal cord injuries, fractures, lacerations and deaths, with reports rising since 2011; most manufacturers do not claim a medical use and therefore remain outside FDA regulation.
A clinical case published in 2024 describes a central cord syndrome that appeared in a 64-year-old man during normal home use of an inversion table. "Severity" is not a synonym for "harmless".
How to do it, if you are among those who can
If your back is not the reason and your shoulder is fine, hanging is a simple exercise and it is worth doing it well.
Grip the bar with an overhand grip, slightly wider than the shoulders, thumb wrapped around the bar. Do not completely let go of the shoulder girdle: keep a minimal activation, as if you wanted to move your ears a finger's width away from your shoulders, without stiffening up.
Feet free, or resting on the ground or on a raised surface if you want to reduce the load: this is the most sensible progression for starting, not a second-rate version.
No swinging, no momentum, no twisting to "feel it more". Breathe slowly. Start with 10-15 seconds, two or three times, and build up over the following weeks as your grip holds, up to 30-45 seconds. Come down in a controlled way, placing your feet down, not letting yourself drop: much of the injuries linked to this exercise come from the exit, not from the hold.
And the rule that outranks all others: if it hurts while you do it, or if the next day your shoulder complains, it is not the right exercise for you at the moment.
What to do instead, if the problem is your back
If you came here looking for something for back pain, the honest answer is that the bar is not the right lever, but there are better levers and they are within your reach.
In the acute phase, staying in motion within what the pain allows beats resting. In the medium term, a gradual exercise programme is the treatment with the strongest evidence behind it, provided it is given weeks rather than days.
If the pain runs down the leg, the pathway is specific and it is worth building it with someone: we have covered this in exercises for sciatica and, for the disc case, in the guide to exercises for a disc herniation.
The hard part is not choosing the exercise: it is understanding what is going on in your case, ruling out the pictures that deserve a medical look, and then adjusting the load week after week while the pain goes up and down. With Ri-Hub this work is done by video call: the physiotherapist assesses your clinical history and movement, sets a sustainable programme and corrects it as you go.
In two lines
Hanging from a pull-up bar does not cure back pain, and systematic reviews of traction have been saying so for years, in fairly clear terms. The relief you feel is real and it does not last long.
As an exercise for grip and shoulders, on the other hand, it is excellent, and that is enough to keep it in your week, provided you do not have one of the conditions for which it is better avoided, and provided you are not putting off an assessment that is needed.
You might also be interested in:
Rotator Cuff: Exercises and a Complete Guide to Strengthening Back Stretching Exercises: A Complete Routine
Sources
- Wegner I, Widyahening IS, van Tulder MW, et al., Traction for low-back pain with or without sciatica, Cochrane Database of Systematic Reviews, 2013. PubMed
- Rizzo RRN, Cashin AG, Wand BM, et al., Non-pharmacological and non-surgical treatments for low back pain in adults: an overview of Cochrane reviews, Cochrane Database of Systematic Reviews, 2025. PubMed
- Vanti C, Panizzolo A, Turone L, et al., Effectiveness of mechanical traction for lumbar radiculopathy: a systematic review and meta-analysis, Physical Therapy, 2021. PubMed
- Vanti C, Turone L, Panizzolo A, et al., Vertical traction for lumbar radiculopathy: a systematic review, Archives of Physiotherapy, 2021. PubMed
- Demirel A, Yorubulut M, Ergun N, Regression of lumbar disc herniation by physiotherapy. Does non-surgical spinal decompression therapy make a difference?, Journal of Back and Musculoskeletal Rehabilitation, 2017. PubMed
- Benoist M, The natural history of lumbar disc herniation and radiculopathy, Joint Bone Spine, 2002. PubMed
- Tyrrell AR, Reilly T, Troup JD, Circadian variation in stature and the effects of spinal loading, Spine, 1985. PubMed
- Celis-Morales CA, Welsh P, Lyall DM, et al., Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all cause mortality: prospective cohort study of half a million UK Biobank participants, BMJ, 2018. PubMed
- Verhagen AP, Downie A, Popal N, et al., Red flags presented in current low back pain guidelines: a review, European Spine Journal, 2016. PubMed
- Zhou T, Salman D, McGregor AH, Recent clinical practice guidelines for the management of low back pain: a global comparison, BMC Musculoskeletal Disorders, 2024. PubMed
- Kassay A, Soliman MAR, Jhawar BS, Recommendations for inversion table therapy, Disability and Rehabilitation, 2023. PubMed
- Michels I, Jacobson L, Central cord syndrome after fall from inversion table, Wisconsin Medical Journal, 2024. PubMed
Last source check: August 2026.
Wondering how much it would cost to be followed? The spend depends on how many sessions are needed, not on the single rate: you will find the figures in How much physiotherapy for low back pain costs.
Need a consultation?
Let's see whether a course makes sense for you: a few questions about your case, you register, and you see who will follow you: a physiotherapist registered with the Albo, whom you can write to in chat whenever you like. If you would rather talk it through, the first consultation is free, lasts twenty minutes and commits you to nothing: they will arrive having already read what you wrote, and together you decide whether and how to go on.
