Bach Flower Remedies are the brain child of Dr Edward Bach who, as an ex-homeopath, invented his very own highly diluted remedies. Like homeopathic medicines, they are devoid of active molecules and are claimed to work via some non-defined ‘energy’. Consequently, the evidence for these treatments is squarely negative: my systematic review analysed the data of all 7 RCTs of human patients or volunteers that were available in 2010. All but one were placebo-controlled. All placebo-controlled trials failed to demonstrate efficacy. I concluded that the most reliable clinical trials do not show any differences between flower remedies and placebos.

But now, a new investigation has become available. The aim of this study was to evaluate the effect of Bach flower Rescue Remedy on the control of risk factors for cardiovascular disease in rats.

A randomized longitudinal experimental study was conducted on 18 Wistar rats which were randomly divided into three groups of six animals each and orogastrically dosed with either 200μl of water (group A, control), or 100μl of water and 100μl of Bach flower remedy (group B), or 200μl of Bach flower remedy (group C) every 2 days, for 20 days. All animals were fed standard rat chow and water ad libitum.

Urine volume, body weight, feces weight, and food intake were measured every 2 days. On day 20, tests of glycemia, hyperuricemia, triglycerides, high-density lipoprotein (HDL) cholesterol, and total cholesterol were performed, and the anatomy and histopathology of the heart, liver and kidneys were evaluated. Data were analyzed using Tukey’s test at a significance level of 5%.

No significant differences were found in food intake, feces weight, urine volume and uric acid levels between groups. Group C had a significantly lower body weight gain than group A and lower glycemia compared with groups A and B. Groups B and C had significantly higher HDL-cholesterol and lower triglycerides than controls. Animals had mild hepatic steatosis, but no cardiac or renal damage was observed in the three groups.

From these results, the authors conclude that Bach flower Rescue Remedy was effective in controlling glycemia, triglycerides, and HDL-cholesterol and may serve as a strategy for reducing risk factors for cardiovascular disease in rats. This study provides some preliminary “proof of concept” data that Bach Rescue Remedy may exert some biological effects.

If ever there was a bizarre study, it must be this one:

  • As far as I know, nobody has ever claimed that Rescue Remedy modified cardiovascular risk factors.
  • It seems debatable whether the observed changes are all positive as far as the cardiovascular risk is concerned.
  • It seems odd that a remedy that does not contain active molecules is associated with some sort of dose-effect response.
  • The modification of cardiovascular risk factors in rats might be of little relevance for humans.
  • A strategy for reducing cardiovascular risk factors in rats seems a strange idea.
  • Even the authors cannot offer a mechanism of action [other than pure magic].

Does this study tell us anything of value? The authors are keen to point out that it provides a preliminary proof of concept for Rescue Remedy having biological effects. Somehow, I doubt that this conclusion will convince many of my readers.

10 Responses to Bach Flower Remedies: this might be the most bizarre study I have seen for a long time

  • Not paying for an article from an alt med magazine so going purely on the abstract. First red flag is cohort size, not much in the way of firm conclusions can be gained from six animals. Second, no actual figures have been released so no idea can be gained on the clinical significance of the result. Third, and a major problem, is that there is only one time point for the biochemical markers, there is no figure for the natural variation of these estimations. If we didn’t already know that the water in group A is no different from B and C the best that can be concluded is that there is a small chance they might be on to something but it isn’t worth doing a larger study, not that it was worth doing this one.

  • Doesn’t Rescue Remedy use alcohol as its solvent? It seems strange to use water as a placebo for that, unless water is used as the solvent in the product.

  • The answer is: Brandy. Flower Essences are generally ‘preserved’ in brandy, and that is the flavor of all of them I’ve tasted in the past. Generally, the ‘mother tincture’ where flowers were floated on pure water in a glass bowl in sunshine is preserved in brandy, then a drop or so of that is put in small vials filled with brandy. Thus, the only suitable control for such an experiment is a similar percentage of brandy without the ‘vibrations’ of the ‘remedies.’ Small amounts of alcohol have biological effects on humans and probably more on rats, who aren’t used to sitting down with a cocktail in the evenings.

    Bach’s site mentions that the ingredients include 5x dilutions of the flower essences and ‘27% Grape Based Brandy’ as the ‘inactive ingredient’ and ‘preservative.’ ( That is 54 proof, decent enough to have on the rocks. Nanoliters are small, and I’m not sure how they chose this measurement (or if it’s accurate?). 200 nanoliters is 0.0002 ml. Typically 1 drop of water is 0.050 ml, so good luck with measuring the difference between 100 and 200 nanoliters. 200 nanoliters is .004 of a drop of water and 100 is .002 drop. What a bizarre choice of dosing. I wouldn’t be surprised if they made a mistake in reporting the measurements (if I’ve made a mistake in my interpretation please correct me).

    • There are different methods – some use water and some alcohol – but I understand that Nelsons Original Bach products use brandy in the first stage, but grape alcohol to dilute it. The final products are classed as foods and are nothing more than this grape alcohol and, as such, are not allowed to make any medicinal or health claims.

  • 200μl is 0.2ml ie about 4 drops. 4 drops is the usual dosage for Rescue Remedy which any ex acupuncturist should know.
    So they got the usual dosage right.

    Maybe the researchers carried out 9 other trials and only reported this one?

    • Thanks for the correction, I was being lazy and read it as nanoliters not microliters.

      It looks like human blood alcohol levels from one drink are about .25 ml/kg ( Rats are often 1/2 kg (500 gm) which means they would need .125 ml alcohol to get to .25 ml/kg. At 25% alcohol (close to the 27% on the Bach site), 4 drops is .05 ml alcohol. It’s still a real amount, likely to have some influence on platelets and other blood factors. Regardless, water isn’t a suitable placebo control.

  • These rats were three month old. They seem to still be growing. That would be consistent with (a) the data in the paper and (B) some growth charts that I pulled off the internet for wistar rats. At the end of the study control group A had gained 83.83g (on average) while group C had gained 48.20g. This was, according to the authors, statistically significant. isn’t it worrying anyone that Bach remedy appeared to be inhibiting the growth of these rats?

    I read the entirety of the the results section and then looked desperately for the rest. sadly there wasn’t any more than 13 lines of the results and one table. This is all to describe the entire study which claims to be longitudinal. However, no there is no baseline data (before treatment) and only day 20 data is published, even though data is collected at two day intervals. for example group A appears to weight more than the other groups but there is no indication of whether group A started out heavier. The authors refer to histological data but I can’t find any of that in the paper.
    The methodology goes into some detail on the statistical analyses. “statistical analysis with repeated measures ANOVA was used with a post hoc Tukey test” It is, however, not at all clear whether all pairs of data are compared by this test and only significant data is published. There is no clear indication of the variation in values within each group. Mean and standard deviation would seem to be a minimal requirement here.
    I agree with the comment above regarding numbers. There appears to be no power testing to determine the numbers of animals required for the study and, given that there is “no external funding” for this project it seems unlikely that any power calculations have been peer reviewed prior to the study.
    I agree with the above comment regarding an appropriate control.

  • Sorry. I’m back having thought about this some more. This niggled at me until I went and looked at the results again. Let us assume for a moment that this paper is methodologically and statistically sound. I think that the paper shows that Bach Remedy is changing the “metabolism” (I can’t be more precise because this is not my field of expertise) in Wistar rats. I was interested to know what the normal levels of glucose were in Wistar rats. I found three sources putting the normal healthy range of glucose at between 100 and 130mg/dL. In this paper the blood glucose levels were assessed in all three groups after 20 days of treatment. The control group (water only) had a glucose level of 118mg/dL, right in the middle of the normal range for the sources I found. The bach remedy treated rats had glucose levels of 29.62mg/dL. Effectively that is 1/3 of the normal levels of glucose at best.

    Perhaps someone knows better than me. What is the actual, normal range of blood glucose for a Wistar rat?

    Based on this paper if I were a rat I wouldn’t take Bach Remedy.

  • “18 Wistar rats which were randomly divided into three groups of six animals each ”
    high-density lipoprotein (HDL) cholesterol
    total cholesterol
    Also, anatomy and histopathology of the heart, liver and kidneys were evaluated (for what?)

    So … FIVE variables with SIX rats per group, plus the histology. That puts the sample size far below what you need to have reliable statistics.
    Testing small groups for inordinately LARGE numbers of variables, then sifting out the correlations and making claims.

    ” When enough hypotheses are tested, it is virtually certain that some falsely appear statistically significant, since almost every data set with any degree of randomness is likely to contain some spurious correlations. If they are not cautious, researchers using data mining techniques can be easily misled by these apparently significant results.”

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Subscribe via email

Enter your email address to receive notifications of new blog posts by email.

Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted but you must tick the box: “Save my name, email, and website in this browser for the next time I comment.”

The most recent comments from all posts can be seen here.