Spinal decompression exercises: this is what people search for when their back feels "squashed" at the end of the day and they hope a few minutes of unloading will put things right. Before giving you the list, however, I have to tell you the thing that almost all articles on this topic skip: that stretching your spine at home creates lasting space between the vertebrae is not proven. The professional version of the same idea, mechanical traction in the clinic, has been studied for a long time and has not stood up to the test.
This does not mean those positions are useless or that you are inventing the relief you feel. It means they work for a different reason from the one you were told, and knowing which one changes how you use them. In this guide you will find what the studies really say, which positions are worth it, what risks the "advanced" versions carry and what really makes the difference.
⚠ Important: read this first before going on
The exercises and advice in this article are general guidance and may not suit your situation. Everyone has a unique clinical history: what works for others could be counterproductive for you.
Want a professional's opinion? Start with the questionnaire: you register and see who will follow you, a physiotherapist on the official register chosen for your case, whom you write to in chat. If you would rather hear their voice first, the first consultation is free and we look at your situation together.
First of all: the signs that come before any exercise
Back pain is almost always benign, but there is a minority of cases where something else lies behind it. The four categories that must never be missed are always the same: fracture, tumour, infection and cauda equina syndrome. Though, to be honest, a review of 16 guidelines from 15 countries catalogued 46 different red flags, with no consensus among the guidelines on which to adopt and with diagnostic accuracy data almost never provided to support them: eight guidelines chose them on the basis of consensus or earlier documents. They are there to make you pick up the phone, not to make a diagnosis.
Go to the emergency department, today, if any of these appear: difficulty holding or passing urine or faeces, numbness in the area that rests on the bicycle saddle (inner thighs, perineum), loss of strength in a leg that gives way, tingling that involves both legs. This is the picture of cauda equina syndrome, and it is time-critical.
Get checked before starting any exercise if: you have had a diagnosis of cancer, you are losing weight without explanation, you have had a trauma, even an apparently minor one, but you have osteoporosis or have been taking corticosteroids for a long time, you have a fever or shivers, the pain wakes you at night and does not change in any position, or you are over 50 and this is the first significant episode of your life.
If none of this applies, you can keep reading with peace of mind.
What the research really says about spinal decompression
The reasoning behind spinal decompression is intuitive: if the pain comes from a compressed structure, moving the vertebrae apart should release it from the grip. It is so intuitive that it has been tested for decades, in clinics, with machines capable of applying forces far greater than anything you can do on a mat at home.
The Cochrane review on traction for low back pain with or without sciatica gathered 32 randomised studies with 2,762 participants in total, of which 16 studies at low risk of bias (57% of the participants). The authors' conclusion is clear: traction, alone or combined with other treatments, has little or no impact on pain intensity, functional status, global improvement and return to work. And in particular, in people with sciatica, that is precisely the group for whom the decompression idea seems most sensible, the evidence indicates that traction probably has no effect on pain, function or global improvement (on return to work, in this subgroup, no study reported data).
The authors themselves declare one limitation, and I report it even though it weakens what I am writing: the available evidence is of limited quality, from studies with small samples and a moderate to high risk of bias, and the literature search of that review stops at 2012. It is not "proven not to work": it is that, after decades of attempts, no effect that could be seen has ever emerged.
The picture has not changed over time. A 2025 Cochrane overview of non-pharmacological treatments for back pain reports that, in chronic pain, compared with sham traction there is probably no difference in pain intensity (mean difference -4 on a 0 to 100 scale, 95% confidence interval -17.7 to 9.7; 1 study, 60 participants, moderate-certainty evidence).
There is a fragment of signal pointing in a more favourable direction, but it needs reading in full. A meta-analysis on lumbar radiculopathy (8 studies included, 5 meta-analysed) found that mechanical traction in the supine position, added to physiotherapy, had significant effects on pain (g = -0.58; 95% CI -0.87 to -0.29) and disability (g = -0.78; 95% CI -1.45 to -0.11), but those results came from low-quality studies, while the analyses of the high-quality studies on prone traction were not significant. The effects, when they existed, were short-term, up to three months. The authors' conclusion, to be fair, favours supine traction added to physiotherapy in the short term: it is the most generous reading the literature allows, and it still applies to a treatment delivered in a clinic, not on the mat at home.
A second review by the same authors (16 studies, 5 in meta-analysis, people with back pain and radiculopathy) compared different types and doses of lumbar traction with each other, and here everything needs to be told. On one side: no significant difference in the short term between the different modalities, with evidence of very low to low quality. On the other, and this is the passage that cuts against what I am arguing, both high-force and low-force traction showed clinically significant pain improvements, and the authors conclude that the literature suggests a short-term effectiveness of traction on pain, regardless of type and dose.
The point is that in these comparisons all groups received traction: there was nobody who did not, so from there you cannot separate the effect of the treatment from the natural course and from expectation. That comparison is made by the other two reviews, and that is where the effect disappears. The authors themselves write that effects different from mechanical ones may be at play: if doubling the force does not change the result, the benefit, where it exists, probably does not come from having separated the vertebrae.
That is why an honest article on spinal decompression cannot promise what others promise. If a machine with measured forces does not beat a machine that is switched off, it is hardly plausible that your body weight on a mat will.
What really changes when you lie down
The fact that the vertebrae do not stay apart does not mean nothing happens. Something measurable does happen, and it is the load.
In a landmark study a pressure sensor was implanted in the nucleus of an L4-L5 lumbar disc and the values were recorded for about 24 hours during daily activities. Standing still the pressure was 0.5 MPa; lying prone 0.1 MPa; lying on one side 0.12 MPa. During the night the pressure rose from 0.1 to 0.24 MPa, consistent with the disc rehydrating. It should be said that this concerns a single volunteer and a single disc: these are numbers to read as orders of magnitude, not as a reference table.
From that same work, however, comes a finding that dismantles a piece of very widespread narrative. Standing bent forward measured 1.1 MPa, and lifting 20 kg with a rounded back 2.3 MPa: flexion increases the pressure inside the disc, it does not reduce it. So when you read that the cat stretch or the child's pose "creates space between the vertebrae", the true part is that they open the posterior space between the vertebrae and reduce muscle tension; the false part is that they are decompressing the disc.
The practical conclusion is less spectacular but more useful: what we call spinal decompression exercises are really unloading and relaxation positions. They give relief while you hold them and for a while afterwards, plausibly by reducing muscular guarding, and they let you move with less fear. They are not a structural repair, and they should not replace what does work.
The unloading positions to try at home
I keep them all, because they are comfortable, low-risk and free. Home spinal decompression only changes the expectation with which you use them, not the list of movements.
An important point: before trying these exercises, make sure you have received a correct diagnosis. People often assume they have one problem when the real cause of the pain is something else. Doing the wrong exercises can make things worse instead of better.
Knees to chest. Lying on your back, bring both knees towards your chest and hug the legs under the knees, not over them. Look for the position in which the lumbar area rests on the floor and breathing stays comfortable. It is the exercise almost everyone finds most immediate: if it gives you relief, keep it for as long as it feels pleasant and repeat several times a day.
Child's pose. Kneeling, sit back on your heels and bend forward, bringing your forehead towards the floor and your arms stretched out in front of you. If your knees complain, spread them wider and put a cushion under your buttocks. Breathe slowly: much of the effect here is the relaxation of the paraspinal muscles.
Cat stretch (cat-camel). On all fours, slowly round your back upwards and then let it drop, alternating without forcing the extremes. It is not an exercise of "space": it is an exercise of mobility and desensitisation, and one of the simplest ways to start moving a back that has stiffened out of fear.
Legs resting (partial inversion). Lying on your back with your calves resting on a chair or the sofa, hips and knees at around 90 degrees. It is the position that most reduces tension on the lumbar area without asking you for any effort. In the acute phase it is often the only tolerable position, and that is already useful.
Assisted hang on a low bar (unloaded dead hang). With a bar at pelvis height, grab it and bend your knees, leaving your feet on the floor, so that your arms support only part of your weight. It is the reasonable version of hanging: you can adjust how much load you take off and come out of it in a second.
For all of them the same rule applies: none should increase the pain or push it further down the leg. If a symptom you had in the buttock reaches the calf during or after the exercise, that position is not for you today.
Hanging from a bar and inversion tables: let's talk about the risks
The full dead hang and inversion tables are the point where home spinal decompression stops being harmless. There is no evidence that they cure back pain, and on the safety front there is something you should know.
A publication devoted to inversion table therapy reports that, according to the US FDA medical device reporting database, inversion has been associated with serious injuries including spinal cord injuries, fractures, lacerations and deaths, and that reports of injuries from unmotorised traction have been rising since 2011. The authors also stress that most manufacturers, by not declaring a medical use, remain outside FDA regulation.
I add the practical side, as a precaution rather than as a measurement I found in these sources: being upside down is not advisable in cases of uncontrolled hypertension, glaucoma or other eye problems, heart problems and during pregnancy. If you fall into one of these categories, inversion is not the place to experiment. And the dead hang on the bar, before it is a test for the back, is a test for grip and shoulders: most people stop because of their hands, not for the benefit.
If the idea appeals to you anyway, I discuss it in more detail in does hanging from a bar help back pain?.
When to do them during the day
I have no studies that say one time of day is better than another for spinal decompression exercises, so I will not invent one for you. It does make sense, however, to think about how a disc's day is structured.
At the end of the day the spine has accumulated hours of load, and unloading positions are usually the most welcome: it is the moment when most people find them useful.
On waking the disc has just spent the night rehydrating: in the in vivo measurement the night-time pressure rose from 0.1 to 0.24 MPa. That this makes the spine less available for maximal flexion in the first hour is a reasonable hypothesis, not a finding I have seen measured; when in doubt, once you are up it is best to start with small movements.
After training with axial loads (squats, deadlifts), unloading positions are a comfortable way to wind down. They have not been shown to prevent injuries, but they make a good closing ritual.
What does work: exercise
Here the tone changes, because here the evidence exists. A systematic review with network meta-analysis included 217 randomised studies with 20,969 participants with non-specific back pain lasting at least 12 weeks. Most types of exercise proved more effective than minimal treatment; the results were compatible with moderate to clinically important effects for Pilates, the McKenzie method and functional rehabilitation (for pain), and for flexibility exercises (for function), with mean differences compared with minimal treatment between -15 and -19 for pain and between -10 and -12 for functional limitation, on standardised 0 to 100 scales.
There is a sentence in the conclusions of that same work that is worth as much as the numbers: people with chronic back pain should be encouraged to do the exercise that they enjoy, to support adherence. This is exactly the opposite of an obsession with perfect movement.
So the right place for unloading positions is this: they are the entrance, not the programme. They take away enough pain for you to move, and then the real work is core strengthening and stability, progressive loading and activities you can keep up over time.
If you have a hernia or a protrusion
This is the case in which spinal decompression is proposed most often, and it is also the one that needs the most careful handling. The Cochrane review, as said, finds no impact of traction even in people with sciatica; and among the 32 studies included, seven reported adverse effects, including increased pain, worsening of neurological signs and subsequent surgery.
This is not a reason to stay still; it is a reason to choose the right criterion: the direction in which the symptoms rise towards the back instead of running down the leg. If bending forward makes the tingling drop into the calf, that is not your direction today, however comfortable that position may be. You will find more detail on this in the articles on disc protrusion and exercises and on when back pain radiates into the legs.
Precautions
Avoid traction positions and inversion altogether in cases of recent vertebral fractures, severe osteoporosis, known or suspected spinal tumours, infections, documented vertebral instability, pregnancy, uncontrolled hypertension and eye or heart problems.
Never force through pain. The criterion is not the sensation of stretching: it is that the symptoms do not increase and, above all, do not move further away from the back.
Go step by step. If you decide to try the suspended versions, start from the low bar with your feet on the floor, not from the full dead hang.
Do not use them as a substitute. The most concrete risk of these exercises is not getting hurt: it is spending months looking for passive relief and postponing the part that really has the evidence behind it.
Clinical traction and home unloading
It is worth distinguishing them, even though the conclusion puts them together. Clinical traction uses machines with controlled, measured forces and supervision; home unloading uses body weight and is much gentler. The difference in intensity is enormous, while the difference in demonstrated results is minimal, because neither has been shown to change the course of back pain.
The cost-benefit ratio does change, however. Lying with your legs on the sofa costs nothing and carries practically no risk, so if it gives you relief, do it. Paying for a course of mechanical traction sessions in the hope that it will fix things is a different matter, and on that the literature is clear.
Online physiotherapy: what it is really for here
On a topic like this, a physiotherapist is not needed to teach you to lie down with your knees to your chest: you already know how to do that. They are needed for three concrete things.
The first is ruling out what must be ruled out and working out whether your pain has a mechanical behaviour or not. The second is finding your direction of movement, the one in which the symptoms come closer to the back, which is individual information and cannot be worked out from an article. The third is building the progressive loading programme that comes afterwards, with doses that increase when they need to.
With Ri-Hub you can be followed by experienced physiotherapists for 29,90 € per session, by video call, with exercises reviewed together and corrected when needed.
In summary
Spinal decompression is an intuitive idea that research has not confirmed: traction, which is its most powerful version, has shown little or no impact on pain and function, and changing its type or dose does not change the result. The unloading positions you find here remain good allies anyway: they cost nothing, give relief and get you moving again, as long as you call them by their proper name.
What really alters the course is exercise done consistently, chosen also according to what you can keep up. If after a few weeks of self-management your back is where it was, or if the symptoms run down the leg, the next step is not to look for a more effective spinal decompression position: it is to be assessed.
Sources
- Traction for low-back pain with or without sciatica, Cochrane Database of Systematic Reviews, 2013
- Non-pharmacological and non-surgical treatments for low back pain in adults: an overview of Cochrane reviews, Cochrane Database of Systematic Reviews, 2025
- Effectiveness of Mechanical Traction for Lumbar Radiculopathy: A Systematic Review and Meta-Analysis, Physical Therapy, 2021
- The effects of the addition of mechanical traction to physical therapy on low back pain? A systematic review with meta-analysis, Acta Orthopaedica et Traumatologica Turcica, 2023
- Some types of exercise are more effective than others in people with chronic low back pain: a network meta-analysis, Journal of Physiotherapy, 2021
- New in vivo measurements of pressures in the intervertebral disc in daily life, Spine, 1999
- Recommendations for inversion table therapy, Disability and Rehabilitation, 2023
- Red flags presented in current low back pain guidelines: a review, European Spine Journal, 2016
Sources last checked: August 2026.
You might also be interested in:
Lifting Weights Without Hurting Your Back: A Complete Guide Knock Knees Exercises: 10 Corrective Movements for Knock Knees Core Strengthening: Exercises for a Solid and Stable Base
Stay in touch with us Join our dedicated Facebook group: we reply to all patients, give practical advice and clear up the most common doubts.
Also read: Smart Working and Back Pain: How to Prevent It.
Do you need a consultation?
Start with the questionnaire on Ri-Hub. A few questions about your case, you register, and you see who will follow you: a physiotherapist on the official register, whom you write to in chat whenever you like. If you would rather talk it over by voice, the first consultation is free, lasts 20 minutes and commits you to nothing: they arrive having already read what you wrote, and together you decide whether and how to go on.
