It’s been a while, but after the recent accusations of bias, cherry picking and conflicts of interest from denialists who continue to use the UK Parliamentary Evidence Check to support their claims, this latest example from homeopathy’s ‘academic nemesis’ needed a blog post.
The homeopathic community and its supporters are well used to rising above the continued onslaught of the anti-homeopathy campaign. With limited resources they choose their battles carefully. Imagine their delight then when on Jan 28th 2013 three positive pieces about homeopathy hit the mainstream news :
In the Canadian National Post: Homeopathy Offers Hope
From The Times of London an op ed piece: I’d rather be ‘cured’ by a placebo than rely on science and remain ill
And on CBS’s Doctor Oz show : The Heart of Homeopathy: Like Cures Like (The video is hard to find but it’s there.)
After Doctor Oz’s chatty piece on the usefulness of homeopathy, follow up information was posted on his website along with a paragraph entitled: Could Homeopathy Be Dangerous? The paragraph warned about
“a wide range of adverse effects, including abdominal pain, headaches, nausea, rashes, and pancreas or liver damage.”
“There have also been some reports of serious side effects, including kidney failure, seizures and death.”
Intriguingly the warning followed up with:
“…it’s always very important to consult a homeopathy expert before taking any treatments or giving formulations to children.”
Well that’s good news – advice to consult a homeopathy expert. Not a medical doctor who usually knows little to nothing about homeopathy. Hope yet.
But what about the adverse effects that had been reported?
The paper cited was: Adverse effects of homeopathy: a systematic review of published case reports and case series published in the December 2012 International Journal of Clinical Medicine.
It sounded serious. The article reported that
“..The total number of patients who experienced (adverse effects) of homeopathy amounted to 1159. Overall, (adverse effects) ranged from mild-to-severe and included four fatalities.” And concluded that: “.. Homeopathy has the potential to harm patients and consumers in both direct and indirect ways. Clinicians should be aware of its risks and advise their patients accordingly.”
To a homeopath this is shocking!
In opposition to the current trend of reading the title and conclusion of a study and running to press, the article needed to be tracked back to its source and that’s where any surprise ended.
The authors are Paul Posadzki and Edzard Ernst, former director of the Complementary Medicine department at the Peninsula Medical School in Exeter and self-proclaimed world expert on CAMs. Together they have penned a number of warning papers about the dangers of complementary medicine. Even though Ernst’s position at the school has been eliminated, Posadzki, on faculty, continues in the steps of their erstwhile leader.
The authors have called their study a “Systematic review” which, of course, suggests in depth, rigorous research.
It’s worth taking a close look at their review.
First what they propose:
“The aim of this systematic review was to provide a summary and critical evaluation of the published evidence regarding direct and indirect (adverse effects) associated with homeopathy.”
And then their methodology:
“Five electronic databases were searched to identify all relevant case reports and case series – from their inception and without language restrictions.”
So that’s any cases reported in MEDLINE, EMBASE, AMED, CINHAL and ISI in any language, from anywhere and with no restrictions on time of publication.
And to make sure nothing was missed:
” …our own extensive department files were hand-searched for further articles.”
So a thorough search of all available data.
” To be included, (cases) had to pertain to (adverse effects) associated with the use of any type of homeopathic treatment in human patients. Data from spontaneous reporting systems were included as well. We also included reports where harm was not because of a homeopathic remedy, but was associated with the use of homeopathy as a replacement of conventional treatments.”
And how was the information from the cases chosen?
“Information from the included (cases) were extracted according to predefined criteria and assessed by two independent reviewers.”
What were those predefined criteria and who were the independent reviewers? The authors don’t give this information.
“Causality was estimated based on the description provided by the authors of the primary articles. Any disagreements were settled through discussion.” (My bolding)
Which brings us to the results.
“ The searches generated 378 articles of which 340 had to be excluded. So 38 articles met the eligibility criteria. And of those, 30 pertained to direct (adverse effects) of homeopathic remedies.”
The authors divide the results into two tables – one for adverse effects associated with the use of homeopathy, the other for cases where conventional care was substituted by homeopathy to the perceived detriment of the patient.
“The total number of patients who experienced (adverse effects) of homeopathy amounted to 1159. Overall, (adverse effects) ranged from mild-to-severe and included four fatalities.” (My bolding)
There’s no way to distinguish in those 1159 effects what might have been mild symptoms of an aggravation of course. But that’s still a lot of adverse effects and there’s still the shock of 4 fatalities associated with homeopathic medication.
And so to the tables themselves.
In fact the table for adverse effects associated with homeopathic medicines seems to show reports of 2 fatalities, with another 2 fatalities in the table for substitution of homeopathy for conventional medicine – one of which took a while to find since it was listed under condition rather than outcome.
In the table of adverse effects “directly related to homeopathy” the authors assigned a level of causality as follows:
Certain: 2 cases
Almost certain: 25 cases
Likely: 46 cases
Unclear: 1,070 cases
It seems important to take a look at that series of 1,070 cases of “unclear” causality. All 1,070 cases were reported in a study of childhood poisoning events from a region of Germany. 17,533 cases were investigated, 16,463 were apparently the result of accidental poisoning with allopathic drugs of one kind or another. 1,070 were ascribed to “homeopathic drugs” – the “remedies unspecified”, the causation was “unclear” and the symptoms reported were “mostly mild symptoms (no details provided).”
From a research standpoint there is zero clarifying information. There is no way to know what the study authors defined as a “homeopathic drug” or the circumstances of its use. Given the general confusion about what a homeopathic medicine is, it’s possible these cases were not associated with homeopathy at all and if they were, they were “mostly mild” and possibly homeopathic aggravations.
If this systematic review had not been published in a respected peer reviewed journal it might be suggested that the 1,070 cases had been included to dramatically boost the numbers.
From a homeopathic standpoint and as a scientist – questioning the inclusion of these 1,070 cases seems justified, so that leaves a total of 73 cases of adverse events reported over a span of 33 years which, after discussion, the authors have agreed are apparently directly attributed to homeopathy.
If this number is broken down further using the provided table – of these 73 cases, the authors rate 2 cases as “certain” and 25 cases “almost certain” in terms of causality. After discussion between the authors, the remaining 46 cases have been assigned a “likely” causality.
As just one example of “likely” causation, the adverse effect is listed as “extreme agitation” in a patient using alcohol and amphetamines who also took “Loco X112” whatever that is, apparently “homeopathic slimming drops”. No identification of the contents of the medication is given but we can be sure it was not prescribed according to homeopathic principles. How the “likely” causation was distinguished from the alcohol and amphetamines is not explained.
A review of cases in the “likely” category suggests they should be put aside for now.
Giving the authors, and their systematic review, the benefit of the doubt, it can be concluded that there have been 27 adverse reactions, including two deaths, reported in 33 years. Of course for a homeopath practicing the gentle healing art, that would still be too many – but it’s important to keep some perspective. Allopathic medicine is now reported as one of the leading causes of death in the US, with in excess of 100,000 deaths per annum. Two deaths possibly associated with homeopathic medicine (neither are described by the authors as of certain causality) still compares favorably with the estimated 3.3 million deaths from allopathic medicine over the same period.
A closer look at the fatal cases is interesting. Two are described as “likely” effects of homeopathy. In one case an unspecified homeopathic medicine was given to a patient with
“… a history of neck swelling, cough, haemoptysis, shortness of breath, fever and weight loss.”
The adverse effect is described as
“arsenical keratosis and cancer caused by Arsenic intoxication”
as judged by the author of the case report. Chemotherapy was administered and/but the patient died. No indication of the potency or dosage was given for the homeopathic medicine so it is impossible to know what it was and if the substance was diluted beyond its material level.
In the second “likely” case, death was from acute pancreatitis and necrosis of the head of the pancreas in patient already taking the diabetic drug Glimepride (Amyrl). Both diabetes and the drug Amryl have known associations with acute pancreatitis – of which necrosis of the head of the pancreas is a known complication. Nonetheless, Nux vomica and Rhus toxicodendron were assigned blame with no dosage or potency given.
The third and fourth fatal cases were in the table of 16 cases where homeopathy was apparently substituted for conventional medicine and described as of “certain” causality. One case involved a patient suffering “Pneumococcal pneumonia with purulent pericarditis and coma“. It is described as negligence of care by the authors and “certain” that “unspecified remedies” were responsible. Any homeopath would agree that such a patient should be in intensive care.
The fourth fatal case is listed under conditions as “malnutrition, sepsis and death” due to negligence of care and attributed to “unspecified remedies”. No other information is given about the circumstances of this patient.
Both these cases, however tragic, say nothing whatsoever about the possible dangers of homeopathy but about dangerous decision making – which can, and does, occur with patients in any therapeutic method.
Such is the rigor of this systematic review.
The authors devote half of the discussion section to 16 cases between 1987 and 2011 where homeopathy had replaced conventional medicine apparently to the detriment of the patient.
In the rest of the discussion the authors are at pains to point out that homeopaths might well claim many, if not most, of the adverse effects reported are actually homeopathic aggravations. But in another separate “systematic review of the evidence” for and against the concept of homeopathic aggravations, Ernst et al have concluded that their
“systematic review of 24 placebo controlled trials reporting aggravations ……does not provide clear evidence that homeopathic aggravations exist.”
Not that there is no evidence – but that their ‘systematic review’ didn’t provide any. Semantics are important. Shang et al anyone?
And what kind of homeopathy is this? What is the definition of homeopathic medicines that the authors use in their review? From the published paper this is unclear and where any information is given, seems to encompass a wide range of multiple medicines prescribed for the same patient plus a number of medications with strange names like Loco X112 with unspecified ingredients.
Where the Law of Similars was used is hard to fathom.
It might be interesting to trawl through the original case reports and discover who the prescribers were.
“In the majority of cases, the possible mechanism of action involved allergic reactions or ingestion of toxic substances. Preparations of heavy metals, such as arsenic, cadmium, mercury or iron, which are frequently used in homeopathy can be toxic, if not highly diluted. Other poisons regularly employed in homeopathy include aconitum, kerosene or thallium, which also can lead to serious health problems in sufficiently low dilutions.”
Which once more is to deliberately misunderstand the fundamentals of homeopathic principles. It’s not called the gentle healing art for nothing.
And there’s more intriguing “cake and eat it” from the authors.
“A systematic review of the (adverse effects) of homeopathy, concluded that the incidence of (adverse effects) of homeopathic remedies was greater than that of placebo in controlled clinical trials; adverse effects included headache, tiredness, skin eruptions, dizziness, bowel dysfunctions and allergic reactions.”
Happily cited by Ernst, who can almost single handedly claim credit for the “homeopathy is no better than water” mantra chanted by his skeptic followers. And so the authors returned to their familiar circular conundrum: “Homeopathy, there’s nothing in it” – well except for whatever it is that causes the adverse effects greater than placebo.
Which is it to be Edzard?
In a last word from the authors, the review:
“… does not tell us anything (My bolding) about the incidence of (adverse effects). Considering the widespread use of homeopathy worldwide and the relative paucity of the reported adverse effects, it might be very low. (My bolding) Collectively, these limitations render our review less conclusive than we had hoped.”
The question that really needs to be answered is how did this review, with its known bias of authors, indistinct identification of homeopathic medicines, questionable determination of causality, self-identified paucity of reported adverse effects and general overall glaring lack of rigor, get published in the peer-reviewed INTERNATIONAL JOURNAL of CLINICAL PRACTICE?
In December 2012 the National Centre for Research Methods reported that:
“…. more than half of UK adults appear to believe that homeopathy is as effective or more effective than conventional medicine.”
Seems that the skeptics’ past assertion that attacking homeopaths was ‘like shooting fish in a barrel’, has turned into Ernst et al scraping the very bottom of the barrel. What next Edzard, bringing the Nazis into the debate?
Desperation anyone? Just sayin’.
But on more serious notes :
Does this work foreshadow a shift in direction? Having failed to convince with claims of placebo, is effort being shifted into warnings about danger? What’s next on that agenda – take over by the pharmaceutical industry? See Merck and New Era…. And would it be wise to look more closely at the many other systematic reviews published by Ernst and the Peninsula Medical School’s CAM department?