Economic Fallacy VII: The Net Economic Gain through Pharmaceuticals
(Find more statistics and indicators etc. in our References section!)
It is believed to be “common wisdom” that longer life spans today must have something to do with the ever stronger presence of modern medicine. However, as statistics of some Third World Countries show, the per capita density of medical personnel, esp. studied physicians etc. is not so very neatly correlated with longevity than pure hygiene or nutrition. Plus, even these per-capita statistics in developing or emerging countries may be misleading: academics such as doctors tend to love the good life of a city dweller, a per capita figure does not say if an average villager might ever see a doctor in his whole life! Not everyone is an Albert Schweitzer or a Mother Theresa.
Give people clean water and proper working conditions, but not a TV and a couch and not lots of pills and drugs and they might be healthier than if they have the former but then ruin their health by obesity and substance abuse, under which heading also fall a lot of bona fide pharmaceuticals that are prescribed by consecutive doctors without heeding the side effects of other doctors’ (or even their own) complementary prescriptions. Add to this the problem of double-blind studies and a very mixed picture emerges.
What is a “double blind” study?
“Double blind” does not refer to patients being treated in the dark corridors of an overcrowded dilapidated National Health Service clinic but rather to the way both patients as well as the personnel administering the drug to be tested are oblivious as to which patients get the real thing, i.e. the drug with the pharmaceutical compound to be tested and who gets a mere placebo, a pill that looks exactly the same as the “real” one but consists only of harmless components such as milk sugar etc.
What is the placebo effect?
Listen to Seeking Answers In A Maze Of Health Studies and you get a first glimpse of what might be amiss. Still, the discussion there grants”controlled studies” better performance and reliability than purely”observational” studies. But is that actually true?
A placebo is a medicine that contains no substance known to have a medical effect but a substance that should be absolutely neutral with regard to the medical condition in question. Lo and behold though, even such a supposedly “ineffective” drug often has a healing effect, in other words, if you take 100 people with the exact same condition, e.g. migraine and give half of them a placebo while the others go “untreated”, a considerable share of the people swallowing the placebo pill might report how that pill helped them get better. That is the placebo effect.
Since it can practically always be observed (or one mustn’t take it for granted that it would be absent) it needs to be statistically excluded from the effect of the “real” substance which is why two groups are used, one getting the placebo and one the substance to be tested.
“At least with present trial designs, it is impossible, with certainty, to exclude a potential placebo effect from the so-called control group. The placebo effect, in other words, is uncontrollable. I suggest that the notions of control group and placebo treatment group are replaced by the notions of ‘placebo effect maximising group’ and ‘placebo effect minimising group’, thus stressing the fact that what such trials measure is a relative effect, and an underestimation of the absolute placebo effect. The illusion of absolute placebo effect evaluation must be acknowledged when articles referring to empirical trials on the placebo effect are interpreted.” (Emphasis CrisisMaven – A. Hróbjartsson (Department of Medical Philosophy and Clinical Theory, Faculty of Health, University of Copenhagen, 3, Blegdamsvej DK-2200 N Denmark, DK): The uncontrollable placebo effect; European Journal of Clinical Pharmacology, Volume 50, Number 5 / Juli 1996, p. 345-348)
(Note: this does not necessarily apply to injections or intravenous infusions as these look the same regardless of ingredient thus no need to differentiate in size, form or colour; that’s why we spoke of about 90% and not 100% of studies affected.)
CrisisMaven once knew a young doctor who stood in as a holiday replacement for a doctor in a large farming district. One day an elderly farmer who had a heart condition came to see her and asked: “Doctor, couldn’t you give me that electrocardiograph treatment again – it helped me so wonderfully last time!”. Now you can imagine how much more she could have healed him if she had just applied her “healing hands” to his chest. Or she could have convinced him that rubbing his chest with dung every morning worked miracles – as a suggestion coming from his doctor in white robes it would probably also have worked. That is the placebo effect.
In fact, there should even be two more groups: the study should be conducted with both ill as well as healthy people! So there should really be four subgroups: two ill groups of patients and two healthy groups, each quarter getting either the placebo or the pharmaceutical. But these are technical details, what concerns us more here is the fact that
The placebo effect is (almost) never properly accounted for.
The reason is quite simple: one placebo is better than another! Though no large-scale studies seem to exist, there is evidence that for every condition, e.g. migraine or diabetes there is likely a form, size and colour of pill that works better than another, even if totally devoid of any active ingredient! And here things get complicated and in the end totally impractical, however, this is never reported to you, the general public.
Let’s say that a pharmaceutical manufacturer wants to test a new drug against hay fever. It is to be taken three times a day and if you did, you are meant to suffer a lot less as initial trials are believed to have shown. The manufacturer has chosen the pill to be round and of green colour. What if an independent study of pure placebo effects established that round green pills actually aggravate hay fever? Would that not mask any potential positive effects of the drug? Or if the opposite were the case – if coincidentally that exact size, shape and colour of the pill by itself, devoid of any active ingredient would already have alleviated hay fever to a great extent? Taking it a step further: what if you designed a placebo that was not neutral but actually gave you stomach ache but led patients to believe it was the “real thing”? Would they not take that as an inevitable side-effect glad to put up with as it worked miracles on their hay fever, whereas their belief in the effectiveness of a drug with no side-effects might be a lot smaller?
Why does it matter?
Well, a true double blind study should therefore only be conducted after first establishing the optimum and the worst possible placebo effect, i.e. the exact shape, size and colour (and for best measure: plus rhythm and frequency of medication!) that gave the very best and the absolute worst positive (or maybe negative, aggravating) “healing” effect of any placebo with regard to the condition in question. Then the placebo administered should be precisely of that kind that in the trials showed the best curative effect while the actual drug, for purposes of establishing the net and true curative effect of the active ingredient to be tested and proven should be administered in the worst possible galenic form! Plus you should consider administering the fake placebos through the person with the best supportive demeanor and apparel while the real substance be administered by a butcher-like creature no patient would ever voluntarily trust with their health! And consider changing ambient conditions for the worst when applying the real thing and so on …
Only then would you be sure as to what was the portion of curative effects that was directly attributable to the substance meant to be proven effective.
Only … this would then cease to be a double-blind clinical trial since everyone, patients as well as medical personnel would be aware of who gets what since you could tell the two apart by colour or shape or size etc.! (Patients might still not know which is which but in all probability medical staff might be aware of some studies linking size of placebo effect to outer appearance of the pill which at best would make it a “single blind” trial; in any case the academics would at least know of the potential effects and suspect which drug is which. Now it gets even worse: consider they guess wrong – then you stand the whole idea of ferreting out the net placebo versus net active ingredient effect on its head!)
From here on it gets even nastier: why not do “simple” blind studies but “double” blind studies? The reason is that if the personnel involved knew who gets what it might show in their attitude towards the patient and their medication and thus again trigger a placebo effect or other influence on the patients one way or another. But wait – who would plan such a study in the first place anyway? Those doctors who do not believe in the positive effect of that particular ingredient or people who are generally leaning towards the new drug? You bet! So in many cases the study would be conducted by people whose institution (clinic, university) had a stake in that new product one way or another. It would be a rare event to convince a clinic whose medical college were openly hostile to the new development to conduct the clinical trials meant to prove how superior this new drug was over what they used in daily practice. So there’s placebo effects and similar psychic influences all around and hence no one knows for sure which drug actually has which net effect. Taken together with drug side effects and esp. drug interactions that are hardly ever studied this paints a grim picture of the potential wealth effect brought about by pharmaceuticals to an (already over-) drugged society where the adverse side-effects of legitimate drugs already kill more people (“patients”) than all illegal drugs (“abuse”) combined …
On the contrary it might be totally worthwhile to launch a huge worldwide study into which forms, colours and sizes of pills had which net curative effect on which medical conditions even without any known active ingredient and use these if, as is to be expected, they have fewer side effects than active ingredients in all cases where there is no positive proof that only an active ingredient (drug) can do the same job. After all, most health systems being publicly financed, it’s your money that’s being spent on drugs that may not have a discernible net effect over the (right) placebo or have outright harmful side effects that could in this way be mitigated or avoided. CrisisMaven does not in any way advocate the abolition of pharmaceutical drugs but their application should be based on scientific evidence not on, as a certain eminent railway engineer would call it, “voodoo science“.