New `study' on vet homeopathy

[Asolepius]

I have just received a news release from the Faculty of Homeopathy, reporting on a study published in its house journal. I haven't seen the full text of the paper yet, but it is reported to be a survey of owner satisfaction with treatment, involving 767 animals over 6 months. It is claimed to be a pilot for a future controlled study. The citation is:

Mathie RT, Hansen L, Elliott MF, Hoare J. Outcomes from homeopathic prescribing in general practice: a prospective, research-targeted, pilot study. Homeopathy 2007; 96: 27-34

PM be if you want the news release (scanned). If anyone can get the full text of the paper that would be great.

[The Don]

There was some reference in the Bristol Evening Post this Saturday about some kind of new study. It was in a puff piece about how wonderful vetinary homoeopathy is.

It's not available online unfortunately

[Asolepius]

That will be the planned Bristol Uni study by Mathie et al. There was a news release about it on the uni site but I can't find it now.

[Ocelot]

Surely even if every owner is 100% satisfied then the most that this could prove is that, as a con trick, homeopathy is highly effective. It says little about the effectiveness of the medicine.

[Asolepius]

Belay that order! I just received the full text. I'll post a critique shortly.

[Katana]

Abstract:


Background and aims
Targeted research development in veterinary homeopathy is properly informed by the systematic collection and analysis of relevant clinical data obtained by its practitioners. We organised a pilot data collection study, in which 8 Faculty of Homeopathy veterinarians collected practice-based clinical and outcomes data over a 6-month period.

Methods
A specifically designed Excel spreadsheet enabled recording of consecutive clinical appointments under the following headings: date; identity of patient and owner (anonymised); age, sex and species of patient; medical condition/complaint treated; whether confirmed diagnosis, chronic or acute, new or follow-up case; owner-assessed outcome (7-point Likert scale: −3 to +3) compared with first appointment; homeopathic medicine/s prescribed; other medication/s for the condition/complaint. Spreadsheets were submitted monthly by e-mail to the project organisers for data checking, synthesis and analysis.

Results
Practitioners submitted data regularly and punctually, and most data cells were completed. 767 individual patients were treated (547 dogs, 155 cats, 50 horses, 5 rabbits, 4 guinea-pigs, 2 birds, 2 goats, 1 cow, and 1 tortoise). Outcome from two or more homeopathic appointments per patient condition was obtained in 539 cases (79.8% showing improvement, 6.1% deterioration, 11.7% no change; outcome not recorded in 2.4% of follow-ups). Strongly positive outcomes (scores of +2 or +3) were achieved in: arthritis and epilepsy in dogs and, in smaller numbers, in atopic dermatitis, gingivitis and hyperthyroidism in cats.

Conclusions
Systematic recording of data by veterinarians in clinical practice is feasible and capable of informing future research in veterinary homeopathy. A refined version of the spreadsheet can be used in larger-scale research-targeted veterinary data collection.

[NeilC]

I hear a recent study proves giving heroin to addicts cures them of their addiction. Apparantly 100% of them were satisfied by the treatment.

[The Don]

That will be the planned Bristol Uni study by Mathie et al. There was a news release about it on the uni site but I can't find it now.
That's the fella. Barely a week goes by without the Evening Post promoting some form of woo. Must be something to do with the fact thay so many ley lines cross under the Evening Post building :rolleyes:



http://www.bris.ac.uk/news/2006/1033.html

[Zep]

All TOO predictable. Writing up a satisfaction survey of like-minded kooks as "science".

Hand up who wants pizza! Great! EVERYONE does! Unanimous decision!

"Background and aims
Targeted research development in domestic comestibles is properly informed by the systematic collection and analysis of relevant clinical data obtained by its practitioners. We organised a pilot data collection study, in which 8 siblings and related family members collected practice-based clinical and outcomes data over a short period..."*

See how easy it is? :rolleyes:

*Alterations highlighted.

[Cuddles]

owner-assessed outcome

So they're asking totally unqualified people if they think their tortoise looks happier than it did before. At least they're consistent with their standard of research.

[crackers]

Abstract:

"Conclusions
Systematic recording of data by veterinarians in clinical practice is feasible and capable of informing future research in veterinary homeopathy. A refined version of the spreadsheet can be used in larger-scale research-targeted veterinary data collection. "

So, the study concluded that veterinarians are able to record data systematically? And that recording data can be helpful in future research? Wow, those are some revolutionary conclusions.

[Loss Leader]

It is claimed to be a pilot for a future controlled study.

They're always pilot studies suggesting properly controlled research. Yet somehow those controlled studies never seem to get done.

[ponderingturtle]

They're always pilot studies suggesting properly controlled research. Yet somehow those controlled studies never seem to get done.

Well never published anyway.

[geoman]

"Conclusions
Systematic recording of data by veterinarians in clinical practice is feasible and capable of informing future research in veterinary homeopathy. A refined version of the spreadsheet can be used in larger-scale research-targeted veterinary data collection. "

So, the study concluded that veterinarians are able to record data systematically? And that recording data can be helpful in future research? Wow, those are some revolutionary conclusions.


Hold yer horses, it only says feasible, not that they actually can record the data. But on the whole it´s hard to disagree with the conclusions.

Oh no - this just in from the Bristol Evening Post - Real Scientist, with PhD and Everything, Agrees with Pro-Homeopathy Study Conclusions!!!!!

ETA: oh, and I like the "specially designed" spreadsheet bit. I tried to prove the existence of Bigfoot by entering data into Excel, but I just used the spreadsheet I use to track my expenses, and it ended up I owed him 2 pound 75p.

[malbui]

ETA: oh, and I like the "specially designed" spreadsheet bit. I tried to prove the existence of Bigfoot by entering data into Excel, but I just used the spreadsheet I use to track my expenses, and it ended up I owed him 2 pound 75p.
Thanks for making me snort with laughter after a crappy day! :D

[ysabella]

I notice they didn't have any ferrets in the study. I think ferrets are more skeptical than most pets. :ferret:

[RichardR]

"Conclusions
Systematic recording of data by veterinarians in clinical practice is feasible and capable of informing future research in veterinary homeopathy. A refined version of the spreadsheet can be used in larger-scale research-targeted veterinary data collection. "

So, the study concluded that veterinarians are able to record data systematically? And that recording data can be helpful in future research? Wow, those are some revolutionary conclusions.Yes but they used a spreadsheet. Don't forget that.

[KingMerv00]

Results
Practitioners submitted data regularly and punctually, and most data cells were completed. 767 individual patients were treated (547 dogs, 155 cats, 50 horses, 5 rabbits, 4 guinea-pigs, 2 birds, 2 goats, 1 cow, and 1 tortoise). Outcome from two or more homeopathic appointments per patient condition was obtained in 539 cases (79.8% showing improvement, 6.1% deterioration, 11.7% no change; outcome not recorded in 2.4% of follow-ups). Strongly positive outcomes (scores of +2 or +3) were achieved in: arthritis and epilepsy in dogs and, in smaller numbers, in atopic dermatitis, gingivitis and hyperthyroidism in cats.


No sign of a control group. One group should have been treated with placebo and the other should have been treated with uh....placebo.

[Katana]

No sign of a control group. One group should have been treated with placebo and the other should have been treated with uh....placebo.

Issues with this particular study aside, not being a vet, I wonder how the placebo effect would work on animals. In the realm of veterinary research, is the double-blind RCT the gold standard? Does it need to be? One wouldn't think that animals would have expectations about the benefit of a particular treatment, so are they subject to a placebo effect?

I ask this because I have pondered it before when I hear about alternative medicine helping animals with a variety of ailments. For something that many of us would attribute to a placebo effect if seen in humans, I feel more inclined to suspect a true benefit if seen in animals even without a placebo group because they don't know that they're supposed to be helped.

Thoughts?

[Asolepius]

Issues with this particular study aside, not being a vet, I wonder how the placebo effect would work on animals. In the realm of veterinary research, is the double-blind RCT the gold standard? Does it need to be? One wouldn't think that animals would have expectations about the benefit of a particular treatment, so are they subject to a placebo effect?

I ask this because I have pondered it before when I hear about alternative medicine helping animals with a variety of ailments. For something that many of us would attribute to a placebo effect if seen in humans, I feel more inclined to suspect a true benefit if seen in animals even without a placebo group because they don't know that they're supposed to be helped.

Thoughts?This is an ancient canard (but they didn't study any ducks). Note that here they didn't do any objective tests of any kind. They didn't even examine the patients, they just asked the owners. It's the owners who are subject to placebo effect. Look at this thread (http://forums.randi.org/showthread.php?t=73136). Despite the obvious fact that the owner was killing her dog with homeopathy, she was delighted with it. This happens with human disease as well. Some people would rather go to an early grave than admit that their woo healer is a fraud.

[Mojo]

Issues with this particular study aside, not being a vet, I wonder how the placebo effect would work on animals. In the realm of veterinary research, is the double-blind RCT the gold standard? Does it need to be? One wouldn't think that animals would have expectations about the benefit of a particular treatment, so are they subject to a placebo effect?

I ask this because I have pondered it before when I hear about alternative medicine helping animals with a variety of ailments. For something that many of us would attribute to a placebo effect if seen in humans, I feel more inclined to suspect a true benefit if seen in animals even without a placebo group because they don't know that they're supposed to be helped.

Thoughts?Here's Dr. Toby Murcott (http://women.timesonline.co.uk/tol/life_and_style/women/families/article663675.ece), the Times's resident placebo enthusiast, on the subject (scroll down to the bottom of the page for his comment). One argument is that animals cannot experience a placebo effect so any positive result is proof of homoeopathy. But pets’ owners can experience a placebo, and this can contribute to, say, a dog’s recuperation. A Norwegian trial published last May supports this.His argument on sCAM generally seems to be that it works via the placebo effect.

And he never seems to give enough detail on the trials and studies he cites to make them easy to look up.

[Geckko]

Just one immediate question jumps to my mind.

How can this be anything but a biased sample? I would not allow a vet to treat any animal of mine using homeopathy and I guess that would go for nearly everyone else here. This leaves the sample open to a bias in favour of those people previously disposed to the effifacy Homeoptathy.

Is this issue addressed in any way in the paper?

[Katana]

Mojo and Asolepius, you make good points.

My statements were meant to apply to veterinary research beyond this study, but I should have read the abstract that I myself posted (ahem - my bad) more closely before questioning the need for a control group.

I was imagining that they would use some objective measure of improvement in which case I didn't think that a control group would be necessary.

But, yes, a study would certainly need a blinded design with a control group if it relies on the owners' assessments - to mitigate any placebo effect, if you will, on them.

[Katana]

Just one immediate question jumps to my mind.

How can this be anything but a biased sample? I would not allow a vet to treat any animal of mine using homeopathy and I guess that would go for nearly everyone else here. This leaves the sample open to a bias in favour of those people previously disposed to the effifacy Homeoptathy.

Is this issue addressed in any way in the paper?
Trying to redeem myself...

From the "Discussion" section:
The overall rate of positive outcome in 79.8% of follow-up patients is somewhat higher than that reported in clinical outcome studies of homeopathy in humans.7, 8 and 12, A score of +2 or +3 was recorded in 61.6% of follow-ups overall, with a slightly lower figure obtained for dogs (60.0%) and a rather higher one for cats (66.7%). It is notable that 87% of these (see Table 2) were in patients overall whose condition was reported as chronic (ie longer than one month duration). This supports the often-expressed view that homeopathy can have a positive impact in patients with long-term health problems. The high frequency of patients with problems categorised as ‘Dermatology’ is consistent with this viewpoint, though this was not the category with the highest percentage of +2 or +3 scores: especially high positive scores were most apparent in patients presenting with uro-genital complaints (although numbers of patients are much lower).

It is important to note that data analysis was not carried out on an intention-to-treat basis. Consistent with the prospective design of the study, the outcome statistics refer only to patients who were re-assessed at follow-up. Any controlled research that is informed by such outcome findings would properly involve intention-to-treat analysis, but the purpose here was to seek trends in homeopathic prescribing and outcomes, and thus to begin to inform future research. A control group was inappropriate to a study of this type.

{snip}

The outcome score used was a generic 7-point Likert scale. Although not strictly validated for the purpose adopted here, such scales have been validated in other research settings13 and have been used in medical homeopathy outcome audits in the past.12 and 14 The scale was chosen for its simplicity and convenience: in a study aiming to provide trends of outcome information for any condition or symptoms, identifying patients with scores ±2 or ±3 was sufficient. For targeted research in named medical conditions and species, however, it is much more important to have validated outcome scales. Such research would also rightly attribute a clear baseline reference assessment against which to gauge any health changes that may be due to homeopathy. Our scale assessed only changes from a baseline recalled by the owner. Controlled research would also normally ascribe specific time-points for follow-up assessment; in a non-controlled data collection study such as this, patients are assessed opportunistically when they return to the veterinarian. This inevitably means that the follow-ups intervals—even for a single named medical condition and species—are highly variable.

Relying on client recall is one of the several potential sources of outcome bias in studies of this kind. Additional sources of bias (probably positive in nature) include: (a) the ‘vet-with-owner’ dialogue and ‘desire to please’ bias in identifying the outcome score; (b) the fact that veterinarians may have selected, unwittingly, some of their most promising cases for homeopathy instead of conventional treatment; (c) owners attending a homeopathic veterinarian may have more confidence in the therapy and empathy with its practitioners. Empathy has been shown to have a positive association with outcome (enablement) from homeopathic treatment in humans.15 It is conceivable that inter-personal relationships of this kind may play a therapeutic role in the homeopathic treatment of companion animals. Targeted research would usefully address issues such as this. Only one of the veterinarians in the current study felt that outcomes scores had a positive bias.

From "Participating Veterinarians' Views":
Sample of specific comments/suggestions by practitioners:

The study has made me aware that I use homeopathy more frequently than I had realised. It has also gave me an incentive to use it more, so thank you for that!


Useful to be able to quote success rates to new clients.


Follow up information was difficult to track if it fell outside of the consultation—I get a lot of information via letter, e-mail, phone, etc and it is missed in the recording.


Multiple conditions for the same animal became confusing as with homeopathy you are not treating a condition as such, so you had to keep thinking conventionally and breaking things down.


The horse yard I visit was a nightmare (92 horses) and I only recorded single horse constitutionals where in fact a lot of group (herd) remedy work is carried out.


I did at times persuade the owner to go for a lower score than their first choice. There is always the risk of owners wanting to be positive and to please the practitioner—and being desperate for a good result.

Here's a link to the paper (http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6WXX-4MTJGVV-7&_user=489293&_coverDate=01%2F31%2F2007&_rdoc=7&_fmt=full&_orig=browse&_srch=doc-info(%23toc%237170%232007%23999039998%23641812%23F LA%23display%23Volume)&_cdi=7170&_sort=d&_docanchor=&_ct=20&_acct=C000022698&_version=1&_urlVersion=0&_userid=489293&md5=93a057d44974665f366db16e6d869448). It may not be free to the public, though, but I figured I'd give it a try.

[Amapola]

Mojo and Asolepius, you make good points.

My statements were meant to apply to veterinary research beyond this study, but I should have read the abstract that I myself posted (ahem - my bad) more closely before questioning the need for a control group.

I was imagining that they would use some objective measure of improvement in which case I didn't think that a control group would be necessary.

But, yes, a study would certainly need a blinded design with a control group if it relies on the owners' assessments - to mitigate any placebo effect, if you will, on them.

Katana, I had a friend who owned a drug testing company for animals and as I recall she always used a control group, no matter what. She tested drugs to determine the dosage that is put on the bottle so vets and their clients would know correct dosage. She would even test the drug to see how much would actually kill an animal, or if it would cause other harmful effects. (Owners don't always follow the directions on the bottle or from the vet and sometimes conclude that "more is better", giving the animal more than what is recommended.)

I never heard of her testing any homeopathic remedies, but then such testing is expensive. The company that makes the product would be the one who would hire her to do the testing.

As a livestock breeder for pretty much my entire life I would never use a homeopathic remedy to treat an animal. Unfortunately I have noticed an increase in the number of vets who offer "alternative" animal treatments. It's almost as though some of them can not stay in business unless they offer it. People seem to be demanding it. If people would just think, and demand the same level of testing that they demand for other drugs used in livestock, that would be fine with me. Homeopathic drugs would then no doubt go the way of "bleeding" as a treatment.

If anyone is interested in seeing an example of a *REAL* study: LIQUAMYCIN® LA-200® (http://www.fda.gov/cvm/FOI/668.htm)

[Dogdoctor]

Issues with this particular study aside, not being a vet, I wonder how the placebo effect would work on animals. In the realm of veterinary research, is the double-blind RCT the gold standard? Does it need to be? One wouldn't think that animals would have expectations about the benefit of a particular treatment, so are they subject to a placebo effect?

I ask this because I have pondered it before when I hear about alternative medicine helping animals with a variety of ailments. For something that many of us would attribute to a placebo effect if seen in humans, I feel more inclined to suspect a true benefit if seen in animals even without a placebo group because they don't know that they're supposed to be helped.

Thoughts?

There are various types of effects that can be confused with placebo. For one thing the owners attitude can affect how well the animal does. Also researcher bias or if the owner is biased and reporting how well the animal does it can produce effects similar to placebo. You need to do double blinded controlled studies with animals also.

[KingMerv00]

This is an ancient canard (but they didn't study any ducks). Note that here they didn't do any objective tests of any kind. They didn't even examine the patients, they just asked the owners. It's the owners who are subject to placebo effect. Look at this thread (http://forums.randi.org/showthread.php?t=73136). Despite the obvious fact that the owner was killing her dog with homeopathy, she was delighted with it. This happens with human disease as well. Some people would rather go to an early grave than admit that their woo healer is a fraud.

Thank you for clarifying what I meant.

Don't we also need a group of pets/owners that got no treatment at all to establish a baseline satisfaction level?

[Dogdoctor]

An example of a placebo type of effect. I saw a cat that had cancer which was untreatable and the owners were interested in making the cat more comfortable. They gave it a Bach flower essence treatment and the cat went into a coma and died. The owners translated that as the product made the cat sleep peacefully till it died. Reality is the product may not have had any effect and the cat went into a coma from the cancer. Owners will report any signs as positive if they believe a product works.

[Katana]

There are various types of effects that can be confused with placebo. For one thing the owners attitude can affect how well the animal does. Also researcher bias or if the owner is biased and reporting how well the animal does it can produce effects similar to placebo. You need to do double blinded controlled studies with animals also.

I know, DD. I realized my mistake.

See my post: http://forums.randi.org/showpost.php?p=2321120&postcount=23

[Dogdoctor]

I know, DD. I realized my mistake.

See my post: http://forums.randi.org/showpost.php?p=2321120&postcount=23

I find the whole idea of placebo type effects in animals interesting. It's not quite the same as humans but very similar.

[George152]

Issues with this particular study aside, not being a vet, I wonder how the placebo effect would work on animals. In the realm of veterinary research, is the double-blind RCT the gold standard? Does it need to be? One wouldn't think that animals would have expectations about the benefit of a particular treatment, so are they subject to a placebo effect?

I ask this because I have pondered it before when I hear about alternative medicine helping animals with a variety of ailments. For something that many of us would attribute to a placebo effect if seen in humans, I feel more inclined to suspect a true benefit if seen in animals even without a placebo group because they don't know that they're supposed to be helped.

Thoughts?

Well I suppose having worked with livestock for many years I can make some claim of knowing how animals operate.
There was a TV program where one of these 'alternate' practicioners was working with a horse that had apparently foundered.
He led the horse off and it really looked in pain and moved badly.
Then he stood the horse and didd his 'magic' whereupon the horse moved off freely.
All he did was the old gypsy trick on standing the horse on its off step.
And of course after his 'treatment' the horse moved off on its on step.
There are nerves in many animals that when manipulated will make the animal do all sorts of strange things..
Homeopathetic is a scam