Kim Burton is an osteopath and the Director of the Spinal Research Unit at the University of Huddersfield.He is Associate Professor of Clinical Biomechanics at the British School of Osteopathy and Editor‑in‑Chief of the International Journal of Clinical Biomechanics He served on the Royal College of General Practitioners working group developing clinical guidelines for the management of acute low back pain, and is co‑author of the systematic evidence review forming the basis of the Faculty of Occupational Medicine's acute low back pain guidelines. He is a member of the management committee of an EU COST programme of action to develop European back pain guidelines. He is also a national clinical co‑ordinator for the UK Back Pain, Exercise and Manipulation trial (BEAM) being funded by the Medical Research Council.
Osteopathy: Science and Evidence
This paper was originally provided by Professor Kim Burton to supplement the GOsC's response to the House of Lords Select Committee's call for evidence on complementary and alternative medicine. The intention was to give some background to the relationship between osteopathy and science and to summarise what science has been able to contribute. The focus is on the clinical efficacy of osteopathy. It is published here for consideration by all practitioners.
Science and osteopathy
There is a fundamental problem facing Complementary and Alternative Medicine (CAM), which centres on the way these disciplines define themselves. For instance, osteopathy has been defined in terms of its holistic ‘system of diagnosis and treatment’. Such a definition implies that osteopathy is a distinct philosophical entity that is somehow separate from mainstream medicine. It may be seen (by its practitioners and patients) as being complementary to mainstream medicine, yet offering a form of patient care that is intrinsically different. This does not, a priori, invalidate it as a medical philosophy, but it makes it difficult for science to answer pertinent and reasonable questions about its contribution to health care.
The problem for science stems from the all‑embracing nature of the profession's philosophy, which implies that osteopathy possesses some special property or approach that is lacking in other forms of medicine, and that this approach may be used successfully to treat a multitude of conditions. The scientific method, on which evidence‑based medicine is founded, is not a suitable tool for investigation of such philosophies, concepts or theories. What science requires is the advancement of specific and testable hypotheses. To ask science to test the 'value of osteopathy' is like asking it to test the value of mainstream medicine; this does not constitute a 'scientific' question. What science can do is test the strength of specific hypotheses concerning the value of a specific treatment for a specific condition; this is quite different from attempting to establish the validity of a theory or philosophy. The route for assessing the clinical value of osteopathy (or any other CAM) is to test its value in the treatment of specific conditions, using appropriate outcome measures, and comparing it with competing treatments. Stated simply, for osteopathy to be considered an appropriate treatment option for a given condition or disease requires that it be tested against the competing treatments for that condition or disease. This testing process (usually by means of randomised controlled trials) does nothing to prove the value of a philosophy, but simply identifies those conditions or diseases for which a particular form of treatment offers superior or equivalent benefits. The question of whether osteopathic treatment may be cost-effective is another matter and requires a different set of parameters.
Osteopathy and low back pain ‑ the evidence
Although osteopathy is defined as a system of health care applicable to many diseases and disorders, the greatest proportion of patients attending osteopaths do so for musculoskeletal complaints, in particular low back pain (Burton 1981). It is in the field of back pain that most evidence about the effectiveness of manipulation (and osteopathy) exists. The most convenient and comprehensive way to assess the strength of the evidence about a treatment is by reference to existing systematic reviews of the literature (where available) and to supplement this with examination of individual studies where appropriate. This process has recently been undertaken by a working group appointed by the Royal College of General Practitioners in respect of low back pain (Waddell et al. 1999). The conclusions of that review have been incorporated in the RCGP's Clinical Guidelines for the Management of Acute Low Back Pain (Royal College of General Practitioners 1999, www.rcgp.org.uk ). The RCGP guidelines conclude that there is strong evidence that manipulation can provide short‑term improvement in pain and activity levels in low back pain patients, and higher patient satisfaction. However, there is no firm evidence that it is possible to select which patients will respond or what kind of manipulation (osteopathy, chiropractic or physiotherapy) is most effective, or what may be the optimum timing for this intervention. The evidence suggests that the risks of manipulation for low back pain are very low, provided patients are selected and assessed properly and the treatment carried out by a trained therapist or practitioner. Another review (Assendelft et al. 1996) came to similar conclusions, but indicated that certain manipulative techniques should be avoided when treating the neck. Because chronic back pain was outwith the remit of the RCGP working group, they did not consider the evidence in respect of manipulation in chronic cases. A recent systematic review (van Tulder et al. 1997) concluded that there is strong evidence for the effectiveness of manipulation for chronic low back pain, especially for short term effects. Not included in these reviews are two recent randomised controlled trials of osteopathic management of low back complaints. The first is from the USA and concerns back pain of a duration between three weeks and six months (Andersson et al. 1999). This found that osteopathic management and standard medical care have similar clinical results in these patients, though the use of medication was greater with standard care.
The second is currently awaiting publication (Burton et al. 2000). The participants in this trial were a group of patients with a specific diagnosis ‘sciatica due to lumbar disc herniation’ with an average duration of about eight months. Osteopathic manipulation was compared with another treatment of known efficacy for this diagnosis (chemonucleolysis). The 12 month outcomes for pain and disability were equally good from osteopathic manipulation as from chemonucleolysis, but the manipulated patients experienced greater improvement in back pain and disability (though not leg pain) at two weeks and six weeks. Crude cost analysis found a financial advantage for the osteopathic arm of the trial.
Rigorous economic evaluations of osteopathic treatment for back pain have not been reported to the best of my knowledge. However, the above two studies (and at least one other (Burns & Lyttelton 1994)) do suggest there are areas where cost‑benefits from the use of osteopathy could accrue. This particular issue will be investigated formally in a current national primary care trial (the UK Back Pain, Exercise and Manipulation (UK BEAM) trial being funded by the Medical Research Council (http://www.york.ac.uk/depts/hsce/ukbeam.htm).
However, when considering costs it may be necessary to follow patients for a considerable period of time. It is now known that back pain patients generally rarely consult their GP beyond three‑months, but some 75% continue to experience pain or disability (Croft et al. 1998). A recent longitudinal study looking at osteopaths' patients reported similar findings at four years from consultation (Burton et al. 1997). This study also found that patients frequently returned to their osteopaths over the follow-up period for treatment of recurrences. This latter feature perhaps displays some measure of satisfaction with the care received (a feature of numerous investigations of manipulative care). The mechanisms through which osteopathic treatment may benefit back pain patients (and indeed those with other disorders) remain undetermined; research in fields such as biomechanics and physiology has thus far provided only inconclusive explanations. It is likely that osteopathic treatment, in common with other treatments, for back pain is currently sub-optimal; there is early evidence that appropriate patient educational material (addressing specific psychosocial factors) when used in conjunction with usual care, can have a beneficial influence on outcomes (Burton et al. 1999).
Summary and discussion
The balance of the scientific evidence provides strong support for the efficacy of manipulative treatment (including osteopathy) for relief of low back pain, at least in the short term. It seems likely that osteopathic treatment is especially effective for certain sub‑groups of patients, but there is currently insufficient evidence accurately to define these groups. There is currently only suggestive evidence that osteopathic treatment of back pain may offer economic advantages compared with usual medical care. There is emerging evidence that the efficacy of management of low back pain (including osteopathic approaches) may be enhanced by incorporation of specific patient education. There is no conclusive evidence that offers an explanation for the mechanism by which osteopathic treatment exerts its effect on low back pain. Osteopathy, in common with other CAM professions, continues to rely largely on experience to justify its contribution to health care. Whilst the available scientific evidence does support consideration of osteopathic treatment as a treatment option for low back pain, there remains a need for careful scientific investigation to establish its optimal role. There is considerably less evidence to support its effect for other disorders, but a lack of evidence should not be confused with evidence of no effect. From a scientific perspective, osteopathy can most usefully be considered a form of treatment that should be tested through the, ‘gold standard’ of the randomised controlled clinical trial.
This will need to be done for all those conditions (or their sub-groups) that osteopathy claims to be able to treat successfully. Such trials may usefully be conducted in a variety of environments (eg primary care and industry as well as private practice) and will need to encompass formal economic analysis. The results of this would contribute to the medical literature at an accepted level, and will be available to inform both patients and practitioners in their choice of treatments, as well as providing guidance to health care purchasers. Randomised controlled trials are expensive and require considerable expertise. The CAM professions do not have a strong tradition of scientific research, and apparently find it difficult to formulate their research questions in a scientifically testable form. However, some of the studies cited above illustrate that when the osteopathic profession adopts a scientific stance, and seeks appropriate collaboration, the expertise and finance to advance our knowledge can be secured.
Andersson GBJ, Lucente T, Davis AM, Kappler RE, Lipton JA, Leurgans S 1999. A comparison of osteopathic spinal manipulation with standard care for patients with low back pain. The New England Journal of Medicine,. 341: 1426‑1431.
spinal manipulation: a comprehensive review of the literature. Journal of Family Practice; 42: 475‑480.
Burns K, Lyttelton LK 1994. Osteopathy on the NHS : one practice's experience. Complementary Therapies in Medicine; 2: 200‑203.
Burton AK 1981. Back pain in osteopathic practice. Rheumatology and Rehabilitation; 20: 239‑246.
Burton AK, Tillotson KM, Cleary J 2000. Single blind randomised controlled trial of chemonucleolysis and manipulation in the treatment of symptomatic lumbar disc herniation. European Spine Journat, (in press)
Burton AK, Tillotson KM, McClune TD, Clarke RD, Main CJ 1997. Four year follow‑up of low back pain patients in primary care. Singapore, June 2‑6, International Society for the Study of the Lumbar Spine.
Burton AK, Waddell G, Tillotson KM, Summerton N 1999. Information and advice to patients with back pain can have a positive effect: a randomized controlled trial of a novel educational booklet in primary care. Spine,. 24: 2484‑2491.
Croft PR, Macfarlane GJ, Papageorgiou AC, Thornas E, Silman AJ 1998. Outcome of low back pain in general practice: a prospective study. BMJ, 316:1356‑1359.
Royal College of General Practitioners 1999. Clinical Guidelines for the Management of Acute Low Back Pain. London, Royal College of General Practitioners (www.rcgp.org.uk).
Van Tulder MW Koes BW, Bouter LM 1997 Conservative treatment of acute and chronic nonspecific low back pain: A systematic review of randomized controlled trials of the most common interventions. Spine; 22:2128‑2156.
Waddell G, Mcintosh A, Hutchinson A, Feder G, Lewis M 1999. Low Back Pain Evidence Review. London, Royal College of General Practitioners.