Complementary therapies and the NHS

BMJ 2005; 331 doi: http://dx.doi.org/10.1136/bmj.331.7521.856 (Published 13 October 2005) Cite this as: BMJ 2005;331:856
  1. Trevor Thompson, clinical lecturer,
  2. Gene Feder, professor of primary care research and development (g.s.feder{at}qmul.ac.uk)
  1. Academic Unit of Primary Care, University of Bristol, Bristol BS6 6JL
  2. Centre for Health Sciences, Barts and the London, Queen Mary's School of Medicine and Dentistry, London E1 4AT

    Uncertain evidence of cost effectiveness should not exclude complementary medicine from reviews and guidelines

    In the early 20th century, scientific medicine emerged as the dominant model for health care in the West. Yet, despite the successes of scientific medicine, people have continued to seek treatments outside mainstream services.1 In the United Kingdom about one in 10 of the adult population consults a CAM (complementary and alternative medicine) practitioner every year, and 90% of this contact happens outside the NHS.2

    Why do people turn to these therapies? Persistent symptoms and the real or perceived adverse effects of conventional treatments are the main reasons.w1Patients value complementary practitioners viewing their predicament “as a whole” and not through the fragmenting lens of clinical specialisation or within the time pressured environment of primary care.w2

    The popularity of a clinical method should not, however, be confused with its value. The popularity of CAM may simply reflect the limitations of conventional treatments. In the past 20 years there has been substantial research on its effectiveness. By March 2004 the Cochrane Collaboration had 145 completed reviews of randomised controlled trials of complementary and alternative therapies: a third showed a positive or possibly positive effect, although over half found insufficient evidence to make such judgments.3

    This work has met with resistance from CAM practitioners.w3Many of the methodological objections they raise (the individualising of treatments, the integrity of the practitioner-patient relationships, the subtle and long term outcomes expected) are shared by complex interventions for chronic conditions within mainstream health care.4 Methodological responses have included pragmatic trial designs, nested qualitative studies, and the use of real world observational data to create an “evidence house.”5

    With finite resources, the case for CAM in the NHS will be judged on economic grounds. But the growth in evidence on clinical effectiveness for some complementary and alternative treatments is not matched by evidence of cost effectiveness. This is the main conclusion one can draw from Canter and colleagues' short report in this issue (p 880).6 Looking for randomised studies of complementary or alternative therapies done in the United Kingdom, the authors could locate only five papers for review, four of which reported trials of spinal manipulation. Though the review does not formally assess study quality, it reports that manipulation may be cost effective. In its narrow focus, however, the report fails to address the complexities of cost effectiveness studies in complementary and alternative medicine.7

    By contrast, the multi-method inquiry by Smallwood published last week spawned a broad, if not sprawling report.8 Smallwood was commissioned by the Prince of Wales to investigate whether CAM could save the NHS money in the treatment of chronic conditions. His findings are based on a literature review of studies from the United Kingdom of the big five CAM traditions (acupuncture, homoeopathy, chiropractic, osteopathy, and herbal medicine),9 costed case studies of the provision of CAM in primary care, and interviews with favourably disposed stakeholders.

    The report is not clear about the method of the unsystematic literature review and contains no explicit appraisal of study quality nor synthesis of data on cost. Owing to a paucity of data, Smallwood does not reach any definitive conclusions about the cost effectiveness of CAM but does identify potential savings. For example, a week's supply of St John's wort, with effectiveness equivalent to paroxetine,10 costs only 82p, compared with £1.62 for paroxetine. His case studies suggest that complementary and alternative programmes can lead to savings in direct costs, but these savings will be greatly diminished or abolished when set against the overall costs of providing these services. The provision of specific complementary and alternative interventions by members of existing primary healthcare teams might offer scope for cost savings in such settings.

    The report concludes that complementary and alternative therapies should be targeted at the “effectiveness gaps” of conventional health care,11 particularly in managing chronic pain and mental disorders, and in palliative care. We think this is a useful concept but were perplexed by Smallwood including asthma, for which conventional treatment is generally effective and safe.

    Despite its limitations and the likelihood of bias in its conclusions, we believe that the Smallwood report fulfills a useful political function. It should promote more investment in research on the cost effectiveness of complementary and alternative treatments. Nevertheless, the report's principal recommendation—that NICE (the National Institute for Health and Clinical Excellence) carries out a full assessment of the cost effectiveness of these therapies—is ill advised.

    A more sensible recommendation to NICE would be that, in developing the scope of new guidelines on chronic conditions, the institute pays greater attention to reviewing complementary therapies. Therapists with particular expertise in complementary and alternative treatments for each specific condition should be invited to join guideline development groups. These groups can wrestle with the philosophical and methodological dilemmas over what study designs should be included in the evidence base of the guidelines. Uncertain evidence of effectiveness does not preclude a positive recommendation in a guideline, and original modelling of cost effectiveness can be part of guideline development.12

    Lastly, those making decisions about integrated medicine in the NHS should consider each complementary or alternative therapy on its merits, using a broad range of appropriate scientific evidence including data on cost effectiveness. Such decision making, if done transparently, may change the public perception of scientific medicine for the better.

    Papers p 880

    Embedded ImageAdditional references w1-w3 are on bmj.com


    • Competing interests GF was paid by Weleda to chair a guideline development group on the use of mistletoe as a cancer treatment and by NICE to chair two guideline development groups.