I regularly scan the new publications in alternative medicine hoping that I find some good quality research. And sometimes I do! In such happy moments, I write a post and make sure that I stress the high standard of a paper.
Sadly, such events are rare. Usually, my searches locate a multitude of deplorably poor papers. Most of the time, I ignore them. Sometime, I do write about exemplarily bad science, and often I report about articles that are not just bad but dangerous as well. The following paper falls into this category, I fear.
The aim of this systematic review was to assess the efficacy and safety of herbal medicines for the induction of labor (IOL). The researchers considered experimental and non-experimental studies that compared relevant pregnancy outcomes between users and non-user of herbal medicines for IOL.
A total of 1421 papers were identified and 10 studies, including 5 RCTs met the authors’ inclusion criteria. Papers not published in English were not considered. Three trials were conducted in Iran, two in the USA and one each in South Africa, Israel, Thailand, Australia and Italy.
The quality of the included trial, even of the 5 RCTs, was poor. The results suggest, according to the authors of this paper, that users of herbal medicine – raspberry leaf and castor oil – for IOL were significantly more likely to give birth within 24 hours than non-users. No significant difference in the incidence of caesarean section, assisted vaginal delivery, haemorrhage, meconium-stained liquor and admission to nursery was found between users and non-users of herbal medicines for IOL.
The authors concluded that the findings suggest that herbal medicines for IOL are effective, but there is inconclusive evidence of safety due to lack of good quality data. Thus, the use of herbal medicines for IOL should be avoided until safety issues are clarified. More studies are recommended to establish the safety of herbal medicines.
As I stated above, I am not convinced that this review is any good. It included all sorts of study designs and dismissed papers that were not in English. Surely this approach can only generate a distorted or partial picture. The risks of herbal remedies for mother and baby are not well investigated. In view of the fact that even the 5 RCTs were of poor quality, the first sentence of this conclusion seems most inappropriate.
On the basis of the evidence presented, I feel compelled to urge pregnant women NOT to consent to accept herbal remedies for IOL.
And on the basis of the fact that far too many papers on alternative medicine that emerge every day are not just poor quality but also dangerously mislead the public, I urge publishers, editors, peer-reviewers and researchers to pause and remember that they all have a responsibility. This nonsense has been going on for long enough; it is high time to stop it.
7 Responses to Herbal medicines for the induction of labor? No, definitely not!
Leave a Reply to Edzard Cancel reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
I was struck by your opening comment about good quality alternative medicine research.
Too often, I think CAM skeptics are accused of simply being anti-CAM. No more, no less. And I think that mistaken belief hurts our cause.
I’m sure fellow skeptics would agree that we would love to see good quality research on CAM. Because then—as so many others, including you, have pointed out—it would cease being CAM and simply become a good mainstream treatment.
I mean, who wouldn’t like to be cured with the woo-woo we often see offered by homeos and naturos? It’s easy and non-invasive. We just want evidence first, that’s all.
I am also interested in CAM and hence I read the full article after reading your blog. However, I wonder if you really read the whole article or you just scanned through after reading the abstract. Anyway, if you had done this review, how would you report the results? The authors simply reported what they found and acknowledged the flaws of the papers they included and other limitations. They further rightly advised against use of herbs in pregnancy. I checked the supplementary files and they have included sub-analyses by type of herbs etc. which addresses one of your concerns above. I agree with you about not including English studies but this is a limitation that sometimes cannot be avoided. I am currently conducting a review and this is one of the realities I am facing. To be honest, your advice that “I feel compelled to urge pregnant women NOT to consent to accept herbal remedies for IOL” is not different to what the authors recommended.
the authors included all sorts of study designs and dismissed papers that were not in English. There is no good reason for that in my view.
Spot on Jay!! I also think the Prof rushed to write this blog before he had thoroughly engaged with the discussion of the paper. I have read your 2002 review Edzard on a similar topic and I am sorry it is no better if we are to judge it based on the same standards you are applying to this review. The authors clearly and convincingly justified the use of the “sorts of study designs” and you are wrong to say they dismissed non-English papers. In fact some non-English papers with abstracts in English were included and they made it very clear in the paper. Of course no paper is perfect, but I think your blog is misleading and baseless.
Conflict of interest: I know one of the authors, but I have never discussed with him anything about this paper.
if it’s just ‘some’ papers. the review cannot be systematic!
What matters is the inclusion criteria. They did specify beforehand that they would only include studies in English or those in other languages, but with a detailed abstract in English. “Systematic” simply means they followed defined steps in the conduct of the review that if you repeat you would get similar results. It has nothing to do with the exclusion criteria.
I don’t think this is correct.
a review that pre-defines clearly, for instance, ‘we include all the studies that we have on our desks’ is not systematic;
as you said, they included SOME non-English papers.