The RCN, CAM and the menopause – Part one – The herbs don’t work?

I’ve recently stumbled upon the Royal College of Nursing’s “Complementary approaches to menopausal symptoms” and I’m not overly impressed.  So I’ve decided to review it in two posts.  This post looks at some of its introductory sections and then focuses on what it has to say about herbs and the menopause. 

Part two will look at its inadequate treatment of the remaining complementary therapies (acupuncture, aromatherapy, reflexology and homeopathy)

Don’t get me wrong, the document has its strengths: a good discussion of the placebo effect and information on possible side-effects from herbs, to name two.

However, although it presents itself as a guide to help, “nurses working in the field of menopausal health” to meet their, “responsibility to educate women about alternative choices in a balanced but evidence-based way” it, I think, lets them down on a number of important points. 

It is suffused with some common logical fallacies; notably appeals to common practise and tradition.  There also appears to be some preference for positive evidence over more recent negative evidence – I have no opinion on whether this is by accident or design. 

As I have said, the document contains quite a lot of good sense.  For example, in its prominent discussion of the placebo effect, it makes these points about CAM therapies: 

“When reviewing the evidence for therapies for menopausal symptoms, it can be difficult to find studies of complementary therapies compared with placebo.  Such lack of evidence may be partly due to lack of investment in research, but also lies in a belief amongst some practitioners that in studying such therapies it is difficult to have a control group which is blind to the therapy as complementary therapies often consist of many components and may include therapeutic processes which are unique.  It is also important to recognise that menopausal symptoms are usually, by their nature, self-limiting and will generally improve over time.”  [p.6] [Emphasis mine]

Reminders about the poor quality of much of the CAM literature and the denial by some of the practitioners of proper research methods are useful; as is the observation that the document deals with what are usually self-limiting symptoms.  However, at no point does the document challenge the assertion of CAM apologists that the double-blind RCT methodology cannot be applied to their particular therapy.  It is obvious that generally it can. 

The section on “Accountability” is interesting as well.   It cites the relevant code of conduct.

“The Nursing and Midwifery Council Code of Professional Conduct requires that nurses and midwives must be convinced of the safety and relevance of any therapy and be able to justify its use (NMC, 2002).” [Emphasis mine]

Whilst this doesn’t seem to require any particular standard of evidence for recommended therapies, it does at least require that nurses and midwives should be able, “to justify its use”.  It also makes additional comments which strengthen this position. 

“When you undertake discussions with women about the menopause, you will find that knowledge of complementary therapies is becoming increasingly essential. If nurses are to remain accountable they should be aware of the benefits, side effects, evidence base and interactions as well as the cost of treatments that they might discuss. You should not recommend products or therapies over others without a good evidence base, and we should all remember that verbal recommendations can carry the same weight as a written prescription.” [Emphasis mine]

After these warnings it goes into herbal treatments.  The quality of this section seems somewhat uneven: good advice and fair comment mixed with some dubious assertions.  In its opening paragraph appeals to popularity and tradition are evident. 

“Many women see the use of herbal remedies as a more natural way of managing their menopausal symptoms than conventional medicines. Indeed, herbs have been used for centuries to relieve an assortment of ailments. There are now a wide variety of products available which can be obtained from many sources such as health food shops, supermarkets, pharmacies, herbalists and even via the internet.”  [Emphasis mine]

Of course, just because something has been “used for centuries” and is widely available doesn’t mean that it’s actually useful in itself.  This statement also seems to accept that herbal treatments can, “relieve an assortment of ailments“.  The document then takes a strange tack. 

“Although many women find herbal remedies useful in reducing menopausal symptoms, there is a lack of data concerning their efficacy and safety, and little overall proof to back up the claims made for these remedies. This appears to be due more to the lack of appropriate scientific studies rather than any absence of effect. More rigorous analyses may yet demonstrate clear efficacy for some of these preparations“. [Emphasis mine]

The assertion that lack of data on efficacy and, “little overall proof to back up the claims made for these remedies” on a, “lack of appropriate scientific studies rather than any absence of effect” belies bias: the document has assumed specific effects in the absence of data. 

Better trails might demonstrate “clear efficacy for some of these preparations”, or they might not.  In fact the evidence of the literature on CAM tends to show diminishing effect with increasing study quality.  As we shall see later, some of these alternative approaches to menopausal symptoms illustrate this nicely.  This bias, an assumption of effects in the absence of evidence blights this document. 

However, another good point in this guidance is that it doesn’t fall into the trap of the appeal to nature.  It points out that, “Like all medicines, herbal treatments may cause side effects.” 

It then proceeds to list herbs that, “are in common usage now and some of the lesser known ones”; further, it points out that, “There are many other herbs not listed that some women may use around the menopause but these have no specific properties for menopausal symptoms.”  This, by implication, seems to be falling into the trap of assuming that those listed have, “specific properties for menopausal symptoms.” 

The list of herbal treatments starts with the well known black cohosh (Actaea racemosa or Cimicifuga racemosa).  I’ve recently had a look at this in detail.  There is at present no Cochrane review on its use for menopausal symptoms (though a protocol has been published this year).  However, there have been trials.  These have been of mixed quality; though a high quality trail was published in December 2006 (to be fair, this was after the RCN published this guidance).  Newton et al. concluded that, “Black cohosh used in isolation, or as part of a multibotanical regimen, shows little potential as an important therapy for relief of vasomotor symptoms.” 

NCAM say this about the herb: 

“This herb has received more scientific attention for its possible effects on menopausal symptoms than have other botanicals. Studies of its effectiveness in reducing hot flashes have had mixed results. A study funded by NCCAM and the National Institute on Aging found that black cohosh, whether used alone or with other botanicals, failed to relieve hot flashes and night sweats in postmenopausal women or those approaching menopause. Other research suggests that black cohosh does not act like estrogen, as once was thought. Black cohosh has had a good safety record over a number of years. Some concerns have been raised about whether it may cause liver problems, but an association has not been proven.” [Emphasis mine]

Again, this view is based on the paper by Newton et al. which was not published when the RCN released this document.  So, what was the RCN’s take on this in late 2006?  They start by pointing out that: 

“The American College of Obstetricians and Gynaecologists (ACOG) supports the use of Black Cohosh for up to six months and in Germany the herb is a recognised and regularly used treatment for the menopause, especially for vasomotor and psychological symptoms.”

The ACOG statement seems well reported on by CAM advocates.  However, these views were ACOG’s position in 2001.  In 2004 an ACOG press release entitled, “ACOG Issues State-of-the-Art Guide to Hormone Therapy” had this to say: 

“SSRIs Relieve Hot Flashes; Herbal Remedies Do Not – The category of anti-depressants known as selective serotonin reuptake inhibitors (SSRIs) can be effective alternatives to HT for the relief of menopausal vasomotor symptoms like hot flashes. As for herbal remedies, the Task Force concludes: “Treatment with wild yam extract, black cohosh, or dietary phytoestrogen supplements derived from the isoflavone red clover has no significant effects on vasomotor symptoms.”” [Emphasis mine]

 “Although black cohosh is a botanical treatment widely used in Europe for menopausal symptoms, “its benefits have been evaluated primarily in small short-term studies using since-invalidated measures,” notes Dr. Schiff. The few randomized, controlled trials on black cohosh showed no significant reduction in hot flashes” [Emphasis mine]

 ACOG’s new recommendations were also covered in an online FAQ.

“Black Cohosh. This plant, also known as snakeroot, “squaw” root and bugbane, has been used for centuries in the treatment of women’s reproductive disorders, although no one knows exactly how – or even if – it works. For the past 40 years, black cohosh has been prescribed in Germany where it is regulated and used by women for hot flashes, depression, and sleep disturbances common during perimenopause.  [Emphasis mine]

Because no large, controlled studies of black cohosh have yet been conducted, no recommended doses have been established, nor have specific claims been allowed regarding the herb’s effectiveness. Black cohosh does not appear to have any effect on bone density or cardiovascular health. Some researchers recommend that you limit its use to six months.” [Emphasis mine]

So it looks like the RCN have made a mistake here.  ACOG’s position at the time they wrote their guide did not support the use of black cohosh.  The comment on the six-month period seems to refer to safety concerns.  This statement does confirm its use in Germany.  This, however, does not confer efficacy.  So, what evidence does the document offer? 

“There is limited supporting data for the use of Black Cohosh, and a lack of high quality trials (Huntley, 2003).  Of the five randomised clinical trials (RCTs) that have been reported using a standardised product, two had poor methodology and the results do not bear clear scrutiny. The other studies were better (Rees, 2006), and one compared Black Cohosh and conjugated equine oestrogen in 62 post-menopausal women who were experiencing at least three hot flushes per day. It found greater improvement in women treated with Black Cohosh than conjugated equine oestrogen groups, although these differences did not reach statistical significance (Wuttke et al., 2003). Another study (Jacobson, 2001) looked at women who had survived breast cancer and were taking tamoxifen. It found no significant difference in the reduction of hot flushes between Black Cohosh and a placebo.” [Emphasis mine]

So we have two trials considered to be of reasonable quality.  One showed improvements that, reportedly, “did not reach statistical significance”.  The other found no statistically significant differences.  This is really not a case of, “limited supporting data for the use of Black Cohosh”: this is a case of no credible supporting data. 

Even before this document was released, authoritative groups were releasing statements which were substantially less supportive than the RCN.  For instance the US NIH published this in a consensus statement in 2005. 

“Black cohosh (Actacea racemosa or Cimicifuga racemosa) is the most studied botanical product. Originally, it was thought that black cohosh had estrogenic properties, but recent work suggests that it does not. In the English language literature, there is little evidence that black cohosh is an effective treatment for hot flashes. However, methodologic issues compromise much of the existing research, and ongoing NIH trials should provide helpful data.”  [ p.19] [Emphasis mine]

There are general lessons here.  The Jacobsen paper reported that, “Both treatment and placebo groups reported declines in number and intensity of hot flashes” showing the importance of methodological quality.  He also concluded that, “Our study illustrates the feasibility and value of standard clinical trial methodology in assessing the efficacy and safety of herbal agents”; which puts claims about the unsuitability of RCTs for the assessment of CAM methodologies into a rational context.

Vitally, whilst the RCN document opines that, “More rigorous analyses may yet demonstrate clear efficacy for some of these preparations” the development of evidence reflected by both NCAM and ACOG shows that the reverse is actually the case.

At least a discussion of risks was included.  Also, the document talks sense about the evidence for Dong Quai (Angelica sinensis), pointing out that, “there was no evidence that it was any better than placebo in relieving hot flushes” (Hirata, 1997) and warning of possible side-effects. 

The discussion of Evening Primrose (Oenothera biennis) oil is reasonable as well.  It points out that, “A small RCT did not show any benefit over placebo in controlling hot flushes (Chenoy, 1994).” However, it also asserts that it appears, “to be useful for mastalgia, although it can take several months to have an effect” without producing any evidence.  This is odd, given that a study available when this document was written concluded that, “Neither evening primrose oil nor fish oil offered clear benefit over control oils in the treatment of mastalgia.” (Blommers, 2002). 

Given that it contains much reasonable guidance, it is disappointing that when disusing ginkgo biloba, the document shows clear bias. 

“It is suggested that Gingko [sic] has a beneficial effect on cognitive impairment and dementia. One randomised trial looking at the effects of Gingko on mood and cognition in post-menopausal women showed better non-verbal memory and sustained attention with Gingko (Hartley, 2003). However, there was no significant difference with menopausal symptoms.  Gingko appears to be relatively safe, with only mild or benign side effects reported. There have been occasional reports of vaginal bleeding.”  [Emphasis mine]

I’ve discussed this in a previous post on the joys of Ginkgo.  This passage is peculiar in a number of ways.  First, when small studies have been cited previously this document has mentioned that they were small.  This was done in the case of Chenov (n=56).  Here the RCN neglect to mention that Hartley was even smaller (n=31).  It was also a one week trial. 

However, the worst part of this is that the same lead author had published larger trials of the same herb before this document was published. 

For instance, in 2004 he published the results of a 12 week trial with 57 participants.  This used a commercially available supplement (Gincosan) containing a combination of Ginkgo biloba (120 mg) and Panax ginseng (200 mg).  The conclusion was: 

“There were no significant effects of Gincosan treatment on ratings of mood, bodily symptoms of somatic anxiety, menopausal symptoms, or sleepiness or on any of the cognitive measures of attention, memory or frontal lobe function. Thus, after chronic administration, Gincosan appeared to have no beneficial effects in post-menopausal women.” (Hartley, 2004) [Empasis mine]

In 2005 results from a trial explicitly designed to follow up the original work was published, co-authored by Hartley.  This ran over six weeks with 87 participants.  It concluded: 

” … The beneficial effects of ginkgo were limited to the test of mental flexibility and to those with poorer performance.” (Elsabagh, 2005) [Emphasis mine]

This, at the very least, contradicts the findings of the trial that the RCN cited.  It also adds to the impression that the findings are statistical noise rather than therapeutic signal. 

The selection of evidence cited by the RCN gives rise to an impression that they cherry-picked the oldest, smallest, shortest study.  Could it be because this happened to be positive for Ginkgo biloba? 

Moving on to Ginseng (Panax ginseng), the guidance seems reasonable. 

“…A 16-week, randomised, placebo controlled trial of 384 post-menopausal women did show significant differences for depression and wellbeing scores, but there were no significant differences for other menopausal symptoms (Wiklund, 1999)…” [Emphasis mine]

This is in line with the 2005 NIH consensus statement, which said that it, “may be helpful with respect to quality-of-life outcomes, such as well-being, mood, and sleep, but it does not seem to affect hot flashes.”

Kava kava (Piper methysticum) provides an interesting example, as “concerns regarding its possible association with liver damage have led to it being either withdrawn or suspended by regulatory bodies in many countries including the UK.”  This is again a good reminder that ‘natural’ does not equal ‘safe’. 

As for the evidence the RCN document quite correctly points out that, “a Cochrane review concluded that kava kava may be an effective option in treatment for anxiety, but that trials looking at its use for menopausal symptoms were inconclusive (Pittler, 2003).” 

The final herbal interventions discussed are the use of phytoestrogens.  Here, again, the trend for better trials to show CAM treatments in a less positive light – the opposite trend to the RCN’s stated expectation – is seen. 

To set the scene, a Cochrane Review on phytoestrogens for vasomotor menopausal symptoms, published in late 2007 concluded: 

“Many women have started using therapies that they perceive as ‘natural’ and safe but they often do not have good information about the potential benefits and risks. This review has evaluated the benefits, risks and acceptability of treatments based on phytoestrogens, a group of plant-derived chemicals that are thought to prevent or treat diseases. Phytoestrogens are found in a wide variety of plants some of which are foods, particularly soy, red clover and alfalfa. Most of the trials in this review were small, of short duration and poor quality. Some trials found a slight reduction in hot flushes and night sweats with phytoestrogen-based treatment but overall there was no indication that phytoestrogens worked any better than no treatment.” [Emphasis mine]

Back in 2006, the RCN opined, “Evidence for the use of isoflavones for relief of menopausal symptoms is encouraging and further research is ongoing.”  When discussing efficacy the document makes the following points: 

“A review of placebo-controlled studies of a standardised 40mg red clover isoflavone demonstrated a reduction in the number of hot flushes experienced compared with placebo (Nachtigall, 2006).” [Emphasis mine]

“Several short duration studies have investigated bone density. A one-year study suggested that, through attenuation of bone loss, isoflavones may have a potentially protective effect on the lumbar spine in women (Atkinson, 2004a). While the effects appear positive, Cassidy (2003a) concludes that the optimal dose is as yet unknown.  More research is underway on effect of isoflavones on post-menopausal bone.”

I’m not in a position to check the details of the review by Nachtigall et al [Edit. 15/12/2008. I’ve checked it and it’s non-evidence: a couple of small (n=15) positive studies against three larger negative studies, including the best so far conducted (n=252):Tice et al., 2003). The paper offers up a meta-analysis with a non-significant difference between the herb and the placebo.]  However, it doesn’t sit easily with the NIH consensus statement, which offered this view in 2005. 

“Red clover leaf (Trifolium pretense) contains phytoestrogens compounds and is believed to work as a weak estrogen.  However, studies suggest that it is not effective in reducing hot flashes. “[p.20]  [Emphasis mine]

This is consistent with NCAM’s  current view that, “… five controlled studies found no consistent or conclusive evidence that red clover leaf extract reduces hot flashes.” 

It’s also odd that a high-quality trail whose details were published in the prestigious Journal of the American Medical Association was not considered in the RCN document (Tice, 2003).  This tested two commercially available supplements containing red clover extract (Promensil and Rimostil).  The authors concluded, “Although the study provides some evidence for a biological effect of Promensil, neither supplement had a clinically significant effect on hot flashes or other menopausal symptoms when compared with placebo.”  [Emphasis mine]

So, what do I make of this document?  It’s a little like the proverbial curate’s egg: good in parts.  However, it seems to overstate the case for the herbs, even taking into account the state of knowledge in 2006. 

Its assumption that lack of evidence was due to lack of investigation, rather than lack of effect, is wrong-headed.  The hope it expresses that, “More rigorous analyses may yet demonstrate clear efficacy for some of these preparations” was ill-founded even when the document was published.  The story of the evidence surrounding black cohosh and phytoestrogens shows, as expected, that the reverse is proving to be the case. 

The misrepresentation of ACOG’s position on black cohosh is particularly troubling.  As is it’s handling of the literature on Ginkgo.  We have also seen several other instances where good quality negative evidence has been missed. 

It would also benefit from being rid of some appeals to common practise and tradition.  

However, on the basis of the RCN’s position that nurses and midwives, “should not recommend products or therapies over others without a good evidence base” it is clear that they shouldn’t be recommending herbal treatments. This document doesn’t provide a “good evidence base” for any of the herbs. I am not sure, in some instances that this document makes it clear. 

I notice that it is due for review next month.  When it is I hope that it will reflect the current state of knowledge: as far as the evidence presently goes, the herbs don’t work – for symptoms of menopause at least.

Read part two, RCN, CAM and the menopause – Part two – Credulous nonsense

Check out my originally un-planned part three, Novogen and the RCN’s placebo pushing algorithm
[BPSDB]

Advertisement