Jan 01 2010
Causation is not so simple to determine as one would think. A mantra at SBM is ‘association is not causation’ and much of the belief in the efficacy of a variety of quack nostrums occurs because improvement occurs after use of a nostrum, therefore improvement occurs because of use of a nostrum. It is why vaccines as a cause of autism are so compelling to some. Vaccines are given at the same time autism starts to manifest. It would require more intellectual power than I have not to conclude, wrongly, that vaccines caused the autism. Concluding causation from sequential events is how the human mind works, and reality, as we know and ignore, constantly conspires to fool us into making false causal connections. In Infectious Diseases I see the error almost daily. The patient had a fever, patient was given antibiotics, fever went away. Therefore the antibiotics treated an infection. Well, maybe, maybe not. One of my mantras is ‘antibiotics are not antipyretics’ and you must be very careful before concluding that the fever went away because of the penacephalone.
I blog and podcast in large part to educate myself. It is amazing how much I learn in the process of preparing for one of the entries in my multimedia medical empire. Areas of knowledge that I had no idea existed can be revealed for exploration. I do not have formal training in epidemiology, although part of my job is hospital epidemiology. Like much of medicine, I acquired what I do know from on-the-job training and lots of reading. So imagine my surprise and delight to discover Hill’s Criteria of Causation, thanks to its application to chiropractic subluxation.
In 1965, Austin Bradford Hill published “The Environment and Disease: Association or Causation?” Dr. Hill, an occupational physician, sought to answer the questions (another discussion of the paper here):
How in the first place do we detect these relationships between sickness, injury and conditions of work? How do we determine what are physical, chemical and psychological hazards of occupation, and in particular those that are rare and not easily recognized?…In other words we see that the event B is associated with the environmental feature A, that, to take a specific example, some form of respiratory illness is associated with a dust in the environment. In what circumstances can we pass from this observed association to a verdict of causation? Upon what basis should be proceed to do so?
He then proceeds to list criteria (he refers to them as viewpoints) that help in determining causation. I like frameworks for thinking about processes. They provide a starting point for considering problems. Becoming a doctor is, in part, internalizing frameworks. Early in training you carry lists and papers that remind you how to evaluate acidosis, or the physiology of heart failure or suspected meningitis. After time and repetition these lists are internalized and you can evaluate the problems without resorting to lists. As an intern one of the papers I carried around was The New England Journal of Medicine article on the physiology of the Swan-Ganz catheter, and I would refer to it with each patient who had a Swan. One day I did not need to refer to the paper. I had internalized the information, and tossed the paper into the trash. Frameworks are not the be all and end all, but do serve as foundation upon which to build ideas. Part of being a specialist is to recognize when the framework doesn’t apply. However, I never had a formal framework for thinking about what constitutes causality in medicine. It is worth reading in the original if no other reason as an appreciation of a time when the medical literature was not dry as dust and devoid of humor and style. Current medical journal writing is often an excellent replacement for Ambien, even when you are fascinated by the topic. To call it the medical ‘literature’ is to refer to the phone book as literature.
The viewpoints to consider in determining if association is due to causation:
1) Strength. How strong is the association between the cause and the effect? Hill uses the example of chimney sweeps, who died of scrotal cancer at rates 400 times the normal population. It killed Bert, or so I was lead to believe. He points out that a strong association like scrotal cancer and chimney sweeps is good evidence in favor of causality from an environmental exposure. He also points to that small effects in a population can still be considered strong associations and uses the death rates from Snows evaluation of the 1855 cholera outbreak, where the death rates from the contaminated water were 17/10,000 vs. 5/10,000 in the general population. Not a huge increase in mortality, but a strong association none the less.
The strength of ‘alternative’ therapies usually hovers around background noise, usually at the level of personal experience. If acupuncture or homeopathy were 400 times superior to placebo, there would no discussion of its validity. Many medical therapies are not 400 times as effective as placebo, but the strength of the association between cause and effect is well above background noise.
2) Consistency. Almost every study should support the association for there to be causation. He uses the example of cigarettes as a cause of lung cancer, “The Advisory Committee to the Surgeon-General of the United States Public Health Service found the association of smoking with cancer of the lung in 29 retrospective and 7 prospective inquiries.” From the vantage point of 2009, where the carcinogenic effects of cigarettes are well established, this example is amusing, but instructive. Over time, as studies progressed, there was a consistent association between smoking and cancer.
He also warns about the importance of a good control group, the inclusion of which often spells the death of ‘alternative’ therapy efficacy, as the recent studies in acupuncture have demonstrated:
Patients admitted to hospital for operation for peptic ulcer are questioned about recent domestic anxieties or crises that may have precipitated the acute illness. As controls, patients admitted for operation for a simple hernia are similarly quizzed. But, as Heady points out, the two groups may not be in pari materia. If your wife ran off with the lodger last week you still have to take your perforated ulcer to hospital without delay. But with a hernia you might prefer to stay at home for a while‚ to mourn (or celebrate) the event. No number of exact repetitions would remove or necessarily reveal that fallacy.
As readers of this blog are probably aware, there is one consistent result in ‘alternative’ therapies: increasing the quality of the study decreases the efficacy until the best studies show no effect.
3) Specificity. Since diseases can have multiple etiologies and therapies can have multiple effects, this is a weaker criteria. However, given the knowledge of physiology and biochemistry since 1965, we have more sophisticated techniques for measuring and determining specificity. From the perspective of opportunistic infections, with no knowledge of viral pathophysiology, HIV is hardly a specific cause of disease. GIven the ability to measure HIV viral load and an understanding of the consequences of HIV depletion of CD4 cells, HIV has high specificity for causing AIDS.
“In short, if specificity exists we may be able to draw conclusions without hesitation; if it is not apparent, we are not thereby necessarily left sitting irresolutely on the fence.”
As Dr Hall has discussed, many ‘alternative’ medical paradigms completely lack specificity and are the one true cause or treatment of all diseases, be it subluxation, a liver fluke, or colonic toxin build up. Fools all; infections are the one true cause of all disease.
4) Temporality. The order should be exposure, disease, treatment, resolution. Cause should proceed effect.
Does a particular diet lead to disease or do the early stages of the disease lead to those particular dietetic habits? Does a particular occupation or occupational environment promote infection by the tubercle bacillus or are the men and women who select that kind of work more liable to contract tuberculosis whatever the environment‚ or, indeed, have they already contracted it? This temporal problem may not arise often, but it certainly needs to be remembered
5) Biological gradient. Also known as dose-response. A little exposure should result in a little effect, a large exposure should cause a large effect. Certainly well known to anyone who drinks alcohol; I suppose all homeopaths must be teetotallers.
The comparison would be weakened, though not necessarily destroyed, if it depended upon, say, a much heavier death rate in light smokers and a lower rate in heavier smokers. We should then need to envisage some much more complex relationship to satisfy the cause and effect hypothesis. The clear dose-response curve admits of a simple explanation and obviously puts the case in a clearer light.
Most ‘alternative’ therapies are binary and have no gradient of effect. I suppose that a chiropractor could say your spine is partly unsubluxed as a result of half a spinal manipulation or an acupuncturist saying, your chi is partly unblocked as I used too few needles. I suppose. I assume a reader will comment on the validity of this observation.
6) Plausibility. The effect must have biologic plausibility. I would take it a slightly differently: not only should it be biologically plausible, but should not violate well known laws of the universe. Hill points out “but this is a feature I am convinced we cannot demand. What is biologically plausible depends upon the biological knowledge of the day.”
Yet there is a difference between what is not yet known but possible — for example Helicobacter as a cause of gastric ulcers, which is odd but not impossible — and what almost certainly will never known because it cannot exist without a radical rewrite of all of science: meridians or water memory, which are odd and impossible. I know that there are more things in heaven and earth than are dreamt of in my philosophy. But you have to prove it to me.
In short, the association we observe may be one new to science or medicine and we must not dismiss it too light-heartedly as just too odd. As Sherlock Holmes advised Dr. Watson, “when you have eliminated the impossible, whatever remains, however improbable, must be the truth.” A nice quote, but not necessarily the case. Sometimes what remains is, however improbable, still nonsense. Which leads to:
7) Coherence. “On the other hand the cause-and-effect interpretation of our data should not seriously conflict with the generally known facts of the natural history and biology of the disease.”
I have discussed in prior entries that those who call themselves ‘holistic’ rarely are and a good physician understands a disease from the microscopic to the entire world. I know cholera, for example, from the level of the effect of the toxin on cellular receptors to the world wide changes in potable water that lead to the spread of disease and much in between. There is a coherence of understanding of the disease.
In a wider field John Snow‚ epidemiological observations on the conveyance of cholera by water from the Broad Street Pump would have been put almost beyond dispute if Robert Koch had been then around to isolate the vibrio from the baby nappies, the well itself and the gentleman in delicate health from Brighton. Yet the fact that Koch‚ work was to be awaited another thirty years did not really weaken the epidemiological case though it made it more difficult to establish against the criticisms of the day.
Ignoring most anatomy or physiology or other biologic understanding, most ‘alternative’ therapies have no coherence when placed in the context of the known universe. Homeopathy is, above all, totally incoherent.
8) Experiment. Always nice. Written in 1965, before the massive increase in biomedical research funding, experiments were not as vital in understanding diseases and treatments as they are today.
Unfortunately for most ‘alternative’ medicine, experiments rarely support their theory or efficacy. Not that it ever matters to the practitioners. I think about how my practice has changed over the last 25 years: adding, subtracting and modifying what I do as the data comes in. Consider the 44,000 articles in Pubmed in infectious disease that are published last year. I wonder how much chiropractic (233 articles on 2009) or acupuncture (1000 articles in 2009) or naturopathy (19 articles in 2009) or homeopathy (162 articles in 2009) practice changed as a result of published studies. It cannot be all that hard to keep up and, so, change accordingly.
9) Analogy. If one virus, for example, can cause a disease, then it is reasonable to suggest that a second virus could be responsible for a similar disease. Analogy is not the same as metaphor: both are the preferred methods of understanding ‘alternative’ therapies, but with little comparison to objective reality. He clearly states these are guidelines, and not to be followed blindly.
What I do not believe‚ and this has been suggested, that we can usefully lay down some hard-and-fast rules of evidence that must be obeyed before we can accept cause and effect. None of my nine viewpoints can bring indisputable evidence for or against the cause-and-effect hypothesis and none can be required as a sine qua non. What they can do, with greater or less strength, is to help us to make up our minds on the fundamental question, is there any other way of explaining the set of facts before us, is there any other answer equally, or more, likely than cause and effect?
The importance of considering all the data, the preponderance of information, in deciding cause and effect.
Hill is also not enthusiastic about statistics, the dreaded p-value:
No formal tests of significance can answer those questions. Such tests can, and should, remind us of the effects that the play of chance can create, and they will instruct us in the likely magnitude of those effects. Beyond that they contribute nothing to the, proof‚ of our hypothesis … Between the two world wars there was a strong case for emphasizing to the clinician and other research workers the importance of not overlooking the effects of the play of chance upon their data. Perhaps too often generalities were based upon two men and a laboratory dog while the treatment of choice was deducted from a difference between two bedfuls of patients and might easily have no true meaning. It was therefore a useful corrective for statisticians to stress, and to teach the needs for, tests of significance merely to serve as guides to caution before drawing a conclusion, before inflating the particular to the general… Yet there are innumerable situations in which (tests of significance) are totally unnecessary‚ because the difference is grotesquely obvious, because it is negligible, or because, whether it be formally significant or not, it is too small to be of any practical importance… What is worse the glitter of the t table diverts attention from the inadequacies of the fare.
Statistically significant nonsense is still nonsense. The article puts into perspective the ongoing problem of the meta-analysis. I always say that the meta-analysis is good for a general understanding of an intervention but rarely provides definitive answers. When meta-analyses are compared to subsequent randomized controlled trials, the meta-analyses got it wrong 35% of the time. As a result, I think meta-analyses are great if they support your prior beliefs and can be safely ignored if they contradict them. That is the problem with meta-analyses, they are good at the mathematics/statistics of multiple studies but fail to take into consideration the other viewpoints as enumerated by Dr. Hill. Far be it from me to suggest that the Cochrane reviews may be wanting as they are often considered the be all end all of analysis, but their reviews in the few areas I know a little about always leave me unsatisfied.
Dr Hill ends with a discussion on the importance of then using the information of when association merges into causation and to consider at what point we need to act on the information. Stopping a nausea medication because it may cause birth defects has a different impact than stopping the burning of fuels in the home as cause of lung disease. In the end we have to act on our information, even if it is always incomplete.
All scientific work is incomplete, whether it be observational or experimental. All scientific work is liable to be upset or modified by advancing knowledge. That does not confer upon us a freedom to ignore the knowledge we already have, or to postpone the action that it appears to demand at a given time.
Who knows, asked Robert Browning, but the world may end tonight? True, but on available evidence most of us make ready to commute on 8:30 the next day.
8:30. It was a more leisurely era.
The explicit application of Hill’s criteria is uncommon, at least if ‘Hill’s criteria’ is used as search criteria on Pubmed. There are areas on medicine that are not clearcut as to causality. Do some therapies work? And if so, in what populations? Is X the cause of disease Y?
Recently, Hill’s Criteria were applied to chiropractic subluxation, and subluxation was found wanting. In Chiropractic theory, spinal subluxation is considered to be the cause of nearly all disease. As readers of this blog know, subluxation was made up by DD Palmer and
The Association of Chiropractic Colleges paradigm statement (ACC Paradigm) suggested that, chiropractic is concerned with the preservation and restoration of health, and focuses particular attention on the subluxation. It also defined a subluxation as “a complex of functional and/or structural and/or pathological articular changes that compromise neural integrity and may influence organ system function and general health.
Subluxation is the basis of the practice of chiropractic. It there is no subluxation, the raison d’etre for chiropractic disappears with a pop or perhaps a crack.
So when Timothy A Mirtz et. al searched the literature looking for support for the most fundamental support for the practice of chiropractic and came up with bupkis, well, imagine if the literature search for germs as a cause of infection or atherosclerosis as a cause of heart attack and found nothing. Zip. Nil. Nada. Diddley squat.
What would happen to Infectious Diseases or Cardiology? I would hope they would disappear as specialty. If there is no basis for the practice, it should be abandoned. Right?
What is the support for subluxation, using Hills ever so helpful viewpoints? I will summarize with the table from article
As the conclusion states,
There is a significant lack of evidence in the literature to fulfill Hill’s criteria of causation as regards chiropractic subluxation. No supportive evidence is found for the chiropractic subluxation being associated with any disease process or of creating suboptimal health conditions requiring intervention. Regardless of popular appeal this leaves the subluxation construct in the realm of unsupported speculation. This lack of supportive evidence suggests the subluxation construct has no valid clinical applicability.
So why continue with chiropractic, at least based on the treatment of subluxation? Got me. And pay for it with tax dollars or insurance premiums? If chiropractic is based entirely on nothing substantial, then nothing is what should reasonably be paid.
One wonders about other alternative therapies: homeopathy, acupuncture, various energy therapies etc. Even if a meta-analysis demonstrated marginal statistical benefit, when Hills viewpoints are considered, I doubt any would hold up.
63 Responses to “Causation and Hill’s Criteria”