Traction (“spinal decompression therapy”) doesn’t help lower back pain

By: Lindsay Davey, MScPT, MSc, CDT
October 1, 2013
Editors: Ryan Davey, PhD and Lindsay Davey, MScPT, MSc, CDT

Father with back pain carrying son on the beachThe latest available clinical data does not support the use of traction (either manual traction or machine-based “non-surgical spinal decompression therapy”) for treating patients with lower back pain with or without sciatica, other than for providing temporary pain relief.

Lower back pain is a major source of disability, work absenteeism and health care expenditure worldwide.  In addition to standard rehabilitation practice which relies principally on exercise-based physiotherapy to treat low back pain, a variety of additional therapeutic modalities and alternative therapies exist.  For the latest research on which lower back pain therapies appear work, and which don’t, check out our new post: “New research for preventing recurrence of lower back pain: 2 do’s and 5 don’ts“.

Many back pain therapies enjoy widespread popularity irrespective of supporting clinical evidence.  Traction, and in particular “non-surgical spinal decompression therapy” (an automated form of machine-based traction) is one of the more prominent and controversial treatments.

Traction for lower back pain

What is it?

While “non-surgical spinal decompression therapy” sounds high-tech and modern, forms of traction have actually been used for thousands of years to treat back pain.  Traction involves physically manipulating the body in order to lightly stretch the spine and increase intervertebral space (the space between the vertebrae in your spine).  Traction is believed to work by relieving nerve compression, decreasing disc compression, increasing disc hydration, improving intervertebral joint mobility, or by a number of more esoteric mechanisms, all of which lack conclusive evidence.

Traction is often incorporated into physiotherapy and chiropractic back treatment as a manual (hands-on) approach where the therapist uses their body weight to manipulate the patient’s spine. It is also routinely used by some as a stand-alone therapy, and is sometimes delivered through mechanical or motorized traction tables, including those controlled by a computer (called “non-surgical spinal decompression therapy”).   In fact, here in Ontario up to 30% of people with acute lower back pain with sciatica have received traction therapy (ref 1).

But does the available clinical evidence support its use for treating the most common form of lower back pain, called “non-specific low back pain”, where no specific underlying cause (such as vertebral fracture) is apparent?

Is traction therapy supported by clinical research?

A large number of studies have examined the effectiveness of various forms of traction for acute, subacute and chronic forms of lower back pain, with or without concomitant sciatica.  The majority of lower back pain studies were small in size and inadequately designed to draw meaningful conclusions.

Far from providing clarity, the abundance of low quality back studies have instead muddied the waters with conflicting accounts and low-quality data, making it possible for proponents of traction and non-surgical spinal decompression therapy to find independent clinical research that appears to bolster their claims.

In the absence of large and appropriately designed clinical trials, a systematic analysis of the body of currently published trials is the most valid approach to generating robust conclusions.  Systematic reviews of the likes published in the Cochrane Library are well recognized as the most reliable.  Last month a new Cochrane review of the use of traction for lower back pain was published (ref 2).

The authors included 32 of the best available randomized controlled trials (a robust experimental design) in their review, and excluded the plethora of studies that provide primarily anecdotal evidence.  Unfortunately, even with stringent selection criteria, the quality of evidence in these 32 trials was mediocre at best. The trials involved the use of manual or mechanical/motorized traction to treat non-specific low back pain that was either acute in nature (duration of less than four weeks), subacute (four to 12 weeks) or chronic (more than 12 weeks).  The trials included patients with or without sciatica – an irritation or compression of nerve roots in the lower spine that can result in pain, numbness, weakness and or tingling in the low back, or leg.   Traction was investigated as a stand-alone treatment compared to sham (fake) treatment, other modalities, or in combination with physiotherapy.

Four primary outcome measures were examined by the authors: pain intensity, functional impairment, overall improvement,and return to work.  The authors concluded that:

“There are some randomized controlled trials (RCTs) showing benefit of traction, but the limited quality evidence from these small studies [which have a moderate-to-high risk of bias] show very small effects that are not clinically relevant. In summary, to date the use of traction as treatment for non-specific low back pain is not supported by the best available evidence.”

Further to this, the authors also point out that one of the highest-quality studies they examined (ref 3) demonstrated that traction in patients without sciatica is not better than sham (fake) treatment.

Traction was also found to be associated with some adverse effects in some cases.  Seven of the studies reported adverse effects including exacerbation of pain, nerve irritation, and subsequent surgery, while four of the studies reported that there were no adverse side effects. The remaining 21 trials did not report on side effects.

Should traction be eliminated from lower back pain treatment protocols?

The systematic review discussed above concludes that traction should have no primary role in clinical practice as a treatment for non-specific low back pain, a conclusion that we support, and that is also supported by other recent review studies (ref 5,6).  This conclusion strongly conflicts with the practice of numerous clinics where spinal decompression therapy is provided as the primary “treatment” for low back pain, and sometimes at an exorbitant cost.

However, many people receiving traction do experience temporary relief from lower back pain, as many of my own patients have anecdotally reported.  Should the value of this outcome be entirely discounted?  Of course not.

If temporary relief from lower back pain is desirable, and can be delivered quickly, safely, and at low cost through a few minutes of traction, then it is warranted.

The handful of studies included in the review that compared the effectiveness of physiotherapy with and without the use of traction suggested that there were no differences in patient outcomes.  Nevertheless, it seems reasonable to hypothesize that temporary relief of back pain could still be beneficial for decreasing muscle tension (not to mention elevating mood and optimism) and therefore aide the delivery of clinically-validated treatments such as exercise physiotherapy.  Consequently, in our practice we continue to offer limited manual traction to interested patients suffering from non-specific low back pain as a secondary treatment alongside a clinically-validated exercise physiotherapy program.

Providing temporary pain relief from low back pain is a worthwhile endeavor as long as it doesn’t come at the expense of evidence-based treatment, or significantly increased patient fees.


  1. Li L.C., Bombardier C. Physical therapy management of low back pain: an exploratory survey of therapist approaches. Physical Therapy. 2001; 81(4):1018-28.
  2. Wegner I., Widyahening I.S., et al. Traction for low-back pain with or without sciatica. Cochrane Database Syst Rev. 2013; Aug 19;8.
  3. Schimmel  J.J.P, de Kleuver M., et al. No effect of traction in patients with low back pain: a single centre, single blind, randomized controlled trial of Intervertebral Differential Dynamics Therapy. European Spine Journal. 2009; 18:1843-50.
  4. Chou R., Huffman L.H. Nonpharmacological therapies for acute and chronic low back pain: a review of the evidence for an American Pain Society/American College of Physicians clinical practice guideline. Annals of Internal Medicine. 2007;147(7):492-504.
  5. Gay R.E., Brault J.S. Evidence-informed management of chronic low back pain with traction therapy. Spine. 2008;8(1):234-42.


  1. Nima Pardisnia PT,DC Nima Pardisnia PT,DC says:

    I have to disagree with this article. As a chiropractor and a physiotherapist I have used spinal decompression/traction for years. On subgroup of patients I see great results, including my own neck. I would like to bring your attention to what I think about Spinal Decompression Therapy.

    • Ryan Davey Ryan Davey says:

      Hi Nima,

      Thanks for commenting. The best available evidence currently doesn’t support the use of spinal decompression therapy for treating low-back pain, but it’s possible that in the future stronger data may come out in favor of its use for some specific sub-population(s) of patients that have not currently been identified. That would be great, we would love to have another evidence-based therapeutic modality for treating low-back pain!

  2. Nima Pardisnia PT, DC Nima Pardisnia PT, DC says:

    Hi Ryan. The clinical prediction rule CPR for lower back suggested by Fritz JM, Cleland JA, Childs JD (Subgrouping patients with low back pain: Evolution of a classification approach to physical therapy , Journal of Orthopaedic & Sports Physical Therapy 2007; 37(6): 290-302. ) does include traction as a sub group. This study has been out for long time. Not all lower back pains are the same and therefore when you look evidence you could be miss lead that there is no evidence for a particular treatment.

    In the above study there is a subgroup “4” that is Traction with signs and symptoms of nerve root compression
    and no movement centralizes symptoms
    . Although this is small group regardless they do exist. This is what exactly traction/spinal decompression shines. Small subgroup that did not get better within two weeks without therapy, did not get better with usual physiotherapy non-sense of TENS and heat and some generic exercises, that did not get better with McKenzie centralization, that go worse with over doing wrong core stabilization and increase of intrathecal pressure, did not get better with med (the evidence is out on Tylenol with no effect), and etc. This subgroup respond well with proper care to spinal decompression when and only when coupled with proper education on ADL’s, use of proper modalities and proper core stabilization. Consequently these are the patients that are desperate and come to internet to get confused by your article or to be taken advantage by over priced spinal decompression scam clinics. So yes as Fritz suggests there is a small subgroup. They do respond and you and your colleagues can visit my site or come to my clinic any day to see for yourself. If you want to play the evidence game you need to throw out 80% of physiotherapy modalities out the window and prey that exercises and mobs work. These patients are steps away from surgery sometimes. As long as spinal decompression/traction is indicated like the CPR suggests, they should try it.

  3. Ryan Davey Ryan Davey says:

    Hi Nima,

    Thanks again for the discussion, it is wonderful to engage with interested and knowledgable people on such topics. 

    We agree, there may be a small subpopulation of low-back pain sufferers that receive particular benefit from traction. Such subpopulations can be hard to effectively identify and therefore investigate, and thus may be insufficiently represented in the literature, or their results masked in studies that include overwhelming numbers of subjects that tend to be non-responsive to traction.
    We are very comfortable standing behind Lindsay’s post on this topic, which discusses the best available research to date. Far from scaring patients away from traction, Lindsay acknowledged that we do see value in traction for low back pain, and we do use manual traction in our clinic. The literature does support the use of traction to provide temporary relief. But it doesn’t support it as a primary treatment modality, as it is often marketed. We think that educating interested people to be skeptical of health claims and realistic about their expectations is a good thing, and woefully absent in the public sphere. 

    As Lindsay said:

    “If temporary relief from lower back pain is desirable, and can be delivered quickly, safely, and at low cost through a few minutes of traction, then it is warranted.”


    “Providing temporary pain relief from low back pain is a worthwhile endeavor as long as it doesn’t come at the expense of evidence-based treatment, or significantly increased patient fees.”

    The best evidence supports the use of exercise therapy with education for rehabilitating low back pain without surgery. TENS, traction, heat, laser, drugs and the like may be useful for providing temporary relief to some patients, and may also aide in the delivery of exercise therapy.

    Thanks again for the discussion. It’s great to know that there are therapists out there who share our interest in providing evidence-based care. 

  4. Joseph Lawrence DC DACNB FACFN FABBIR Joseph Lawrence DC DACNB FACFN FABBIR says:

    Ah yes the debate over traction/decompression therapy carries on.As a firm believer of evidence based care,I think we can to some extent rely on the current literature,but to a point.What is consistently overlooked in the referenced papers is the concept of patient selection and treatment position bias.With respect to low back pain for example,there is little if any reference to the protocol one must utilize to classify whether the patient is amenable to traction,let alone whether prone or supine positioning may be most effective.Furthermore,in a certain number of cases,patient positioning needs to change depending on their progress during care.Failure to consider these factors will tip the scales towards relegating traction as a modality with little if any benefit,which may not be the way to categorize it.As one whom is involved in research on traction at the faculty of engineering at a Canadian university for the last two and a half years,we may be seeing some new tricks from an old dog so to speak that can benefit all of our patients in the years ahead.

    • Ryan Davey Ryan Davey says:

      Thanks for your thoughtful comment, you make some good points. Patient selection issues are definitely rampant in the literature, and if the treatment in question is only beneficial to a subset of the study participants, treatment efficacy can be masked.

  5. Stephane Stephane says:

    Am I wrong to infer that L1 to L4 Disk degeneration with 3 small herniations with sciatica would qualify as falling the small subcatagory that may see benifit to adding or including lumbar traction to exercise therapy for rehabilitating low back pain without surgery. As a patient and proud owner of the above diagnosis living in a small town have not been able to find anyone that is even set up to try to include traction/decompression in my recovery. I have gone to 2 physiotherapists and one chiropractor (small town) and still need medication to be able to sit or stand for any normal amount of time. They all sit on their hands and spout the similar findings to your so called study to which you say there is no data to refute or confirm the validity of spinal decompression therapy.

    I have spent 3000$ in the last 2 years on Physio, Acupuncture and Chiro and not to mention the the hundreds on medications (Lyrica and Morphine) and this is why no one seems to want to help. There is no profit if the treatment that you offer truly reduces the number of total visits required to manage pain or improve mobility..

    Your article is inconclusive at best. I agree with Dr Lawrence “What is consistently overlooked in the referenced papers is the concept of patient selection and treatment position bias.With respect to low back pain” You need to qualify your patients by actually diagnosing and defining specific illness criteria so you can target treatment and not dump all lower back pain to one group. The use of sub groups are a way of saying in the study… ” oh damn there are some that are seeing a decrease in pain … lets add a sub group and not treat them as different in the first place.” A study will always say what you intended it to say.
    I agree more study with a more targeted aproach to classifying patients is needed.

    Remember we in pain do not want to be in pain.

    • Lindsay Davey Lindsay Davey says:

      Hello Stephane,

      I’m so sorry to hear that you have had such a challenging time with your back, and that your pain has persisted for so long with so much expenditure in treatment to date. Given you have likely had an MRI or other imaging to know the details of the extent of your herniations, I hope that your team members have pursued a referral to an orthopaedic/neuro surgeon to weigh in on your surgical candidacy. Not that we don’t always hope that surgery can be avoided, and in most cases, conservative approaches do work, however, as you say, it seems that you are in a subclass of patients whose issues may require escalated intervention when conservative measures have not helped sufficiently.
      On the topic of mechanical spinal decompression therapy, I agree that this can potentially be of benefit to patients such as yourself, in that it can temporarily off-load the herniated discs and adjacent emerging spinal nerves, thus alleviating symptoms. (Certainly I would presume that manual decompression via manual traction is something that has likely been attempted already in your care, unfortunately, it appears, with minimal lasting benefit). In your case, with so much effort to treat your condition using other measures, I agree that mechanical decompression therapy is something that I could see having validity in trying if it was available to you. The only thing I would say to my patients in your state would be that while I am hopeful for relief of symptoms using spinal decompression (a very meaningful goal indeed), I would want them to know that I would not expect there to be a reversal of the herniations per se, and that it would provide most benefit if it allowed for a window of reduced symptoms so that additional core and stability exercises and dural flossing techniques could be used. If with reduced symptoms and irritation of the herniated discs/nerves you could perform a more progressive exercise program and tolerate other avenues of manual therapy, then I would certainly support the use of spinal decompression in this way.
      I wish you the best of luck with your pain management and ultimate resolution of your back condition Stephane, using whatever therapeutic approach(es) it takes to do so.
      Warm Regards,
      Lindsay Davey

  6. Macanandha Butros Macanandha Butros says:

    Thanks for the article. As a patient consumer such as Stephane, who is obviously like more and more patients taking charge of educating themselves as much ad possible because hey they have most at stake, are the ones who’s life and functionality, along with financially is taking the risks. That is the hit and miss modality of treatments offered.  The article to me as a consumer, seems to say 1) this at the tine is the latest evidence based study, that is still limited 2) that there  still be a more narrower subset of lower back sufferers that need to be identified properly that spinal decompression can have positive benefits 3) that it seems by many practitioners such ad Nima see success with the protocol-along with PT etc. 4) The point of being low cost ? which with the newer computer based therapies does not seem to most consumers like me a low cost treatment, given it seems to be a gamble risk, worth every penny if it works or even significant long term improvement. Less than surgery -yes, insurance covered-no. 5) Marketed as a breakthrough cure all- kind of. I can go to ten different practices online and clearly see as perceived by me and probably Stephanae as a consumer perceived as finally something that will have an 70 to 90 percent chance of positive results. Bit am I and others that 10 to 20 that haven’t got relief  with other traditional treatments fall into that 10 to 20 percent, and that represents tens of thousands of low back pain suffers. Where as the other 70 to 80 percent  respond  to traditional medical, PT or chiropractic care or simply time. 6) shows as Lawrence stated the de bate continues because of lack of specific unbiased  evidence based studies, and larger and long term follow-up studies. So while many consumers like me are desperate, we are also skeptical,  because we have been disappointed  with relief and results and newer uninsured treatments are expensive to just roll the dice, or possibly do more harm.

Leave a Reply

Your email address will not be published. Required fields are marked *