Analysis of the April 2003 Draft Report of the New Zealand Ministerial Advisory Committee on Complementary and Alternative Health (MACCAH)


June 27, 2003

Section 2: Policy Issues Surrounding CAM
2.3 Research, Evidence and Efficacy
[Comments in red by Stephen Barrett, M.D.]

Existing research and evidence on CAM

It is estimated that over 4000 research trials have been carried out on CAM to date (Peters and Gillam 2001). However, the quality of the research has been variable and there is a perception that the evidence base for CAM is weak (see pages 16­17). [Perception? By whom? There is more than a “perception.” There is very little evidence supporting “CAM” methods and much evidence indicvating that most are not effective. Evaluations should not be based entirely on clinical trials. Plausibility of the underlying theories should also be taken into account. Methods that are implausible or untestable can still be evaluated.] In addition, CAM research is not always easy to access and may be difficult for lay people to interpret. It is important that consumers, practitioners and policy-makers have access to up-to-date, research-based evidence on CAM in order to make informed decisions. [Yes and no. Many methods that don’t work can be evaluated without additional research.] In particular, they need evidence on safety, efficacy and cost-effectiveness. The Ministry of Health’s proposed CAM database will provide summaries of existing evidence on the safety and efficacy of CAM treatments (see page 13). It will also offer links to evidence-based CAM websites. [I am not aware of any “CAM” Web sites that offer eviudence-based information or that tell consumers what does not work.]

Evidence on the safety of CAM is particularly important. [I disagree. “Safety information” is usually gathered after people are injured. Moreover, safety and effectiveness are inter-related because ineffective methods are unsafe when used instead of effective ones. It would be much better for consumers to be advised about what doesn’t work so that fewer people rely on methods that are ineffective. What consumers need is information on how the possible benefits compare with the probable risks. This is available for most mainstream treatments and difficult to get for most “CAM” methods.] Apart from the legislation described in Section 1, there is currently no specific requirement for self-regulated CAM practitioners to provide evidence for the safety of their practice. However, if the proposals for a trans-Tasman agency to regulate therapeutic products become law, importers and manufacturers of CAM medicines will have to provide evidence for the safety of their products. [Why shouldn’t they be required to provide evidence of effectiveness? Why should the general public be guinea pigs for unproven “CAM” methods? What doesn’t MACCAH express concern here about lack of effectivess and lack of plausibility?]

Very little research has been carried out on the cost-effectiveness of CAM, either in New Zealand or overseas (Peters and Gillam 2001). [True — and it never will be, because there isn’t enough money in the work to test every crank method that is proposed.]

Questions
15
16
17
18
Do you consider that consumers have satisfactory access to existing research on CAM?
What steps should be taken to improve access to existing research on CAM?
Do you believe that more research is required on CAM?

In your opinion, what aspects of CAM require further research? For example:

— safety
— efficacy
— cost-effectiveness
— other (please specify).

[Research priorities should be applied equally whether or something is labeled “CAM.” Since research funds are limited, they should be used to pay for the research that has the greatest chance of providing useful information.]

Appropriate research methods

There are difficulties in using ‘conventional’ quantitative methodologies, such as the randomised controlled trial (RCT), to evaluate some CAM modalities (Nahin and Strauss 2001). However, many argue that it is possible to apply the key principles of evidence-based medicine to CAM (Vickers 1999). For instance, RCTs can be designed to take important aspects of CAM, such as the patient­therapist interaction, into account (Vincent and Furnham 1997).

Other quantitative and qualitative methodologies may be more appropriate than RCTs, depending on the modality to be evaluated and the research question. The Foundation for Integrated Medicine has published a table suggesting appropriate research designs for investigating key CAM research questions (Foundation for Integrated Medicine 1997: section 2, p. 11). The methods include case-control studies, action research and qualitative studies. [Qualitative studies have very little if any value.]

Question
19
What comments do you have, if any, on the kind of research methods that are appropriate for investigating CAM?

[Research design should not depend on whether or not a method is labeled “CAM.”]

The concept of ‘levels of evidence’

Biomedicine often uses a ‘levels of evidence’ model to grade evidence from treatment evaluation studies according to the research methods used (see Section 1.3, pages 16­17). Systematic reviews and RCTs are considered to offer the strongest evidence for the effectiveness of medical treatments. It might be appropriate to develop a similar model for research methods used to evaluate CAM treatments. [Some methods can be studied without RCTs. But the overall approach should be the same for all methods.]

Questions
20
21
Do you think the concept of ‘levels of evidence’ is useful in CAM research?
What ‘levels of evidence’ do you believe are appropriate? CAM research in New Zealand?

It may be appropriate to establish some kind of research group, or at least commission individual research projects, to evaluate CAM in the New Zealand context. There may be a role here for the Health Research Council, the government-funded agency responsible for purchasing and co-ordinating health research in New Zealand.

Arguments against setting up such a group or commissioning New Zealand-based projects include the costs involved and the shortage of expertise in this field. An alternative option would be to build up links with, or commission work from, overseas research institutes with specialist CAM knowledge.

Given the limited capacity and funding available to support CAM research in New Zealand, it might be appropriate to examine areas such as consumer satisfaction here but look overseas for research on effectiveness and safety.

[Testimonial evidence might be a factor that could influence research priority. But measuring consumer satisfaction for purposes of judging effectiveness is a waste of time and money.]

Questions
22
23
24
Should New Zealand be carrying out its own CAM research?
What areas should New Zealand CAM research focus on? For example: ­ individual CAM modalities (eg, acupuncture, herbal medicine)
— particular disorders or health issues (eg, CAM for depression)
— certain population groups (eg, CAM for people with disabilities)
— other (please specify).
Who should carry out CAM research in New Zealand? For example:
— government agencies
— universities
— private research institutions
— CAM professional associations
— the CAM industry
— other (please specify)

.[Research priorities should be applied equally whether or something is labeled “CAM.” Since research funds are limited, they should be used to pay for the research that has the greatest chance of providing useful information. Government funds should not be used to test implausible methods (such as homeopathy, reiki, iridology, and reflexology) because the probability of a useful outcome is zero.]

Evaluating existing research

If a dedicated CAM research programme is not possible given New Zealand’s limited capacity, it might be appropriate to devote resources to evaluating the existing international research on CAM. An established research institute might be commissioned to carry out such an evaluation and advise on which findings to implement. The New Zealand Health Technology Assessment clearing house (NZHTA) is an example of a New Zealand-based organisation that carries out this type of work. The NZHTA reviews and evaluates research on all aspects of health care, and has already carried out some work on CAM. NZHTA is based at the Christchurch School of Medicine.

Evaluated research summaries on the proposed CAM database may also eventually play a role in informing health policy in relation to CAM.

Questions
25
26
Who should be responsible for evaluating existing research on CAM?
Do you have any other comments about research, evidence and efficacy?

[I am not aware of any “CAM” methods where research has been done and not evaluated. The public needs information on what not to use, based on the plausibility or implausibility of the practice. Whether government agencies can supply such information is unclear. Worldwide, few government agencies have provided information on what doesn’t work. Many agencies and organizations state that various methods are unsubstantiated, but few recommend abandoning methods that are unsubstantiated and lack a plusible rationale.]