REDUCTIO AD DUAS: CAUSAL VARIABLES AND THE FAILURE OF MODERN MEDICINE

A good doctor is worth his weight in gold. In fact, the price of good doctors is going up because they are getting increasingly rare – and therefore more precious. If you indeed have a good doctor, let him know that he (or she) is appreciated. You should appreciate what you have because the medical profession hasn’t exactly been setting the world on fire recently. Come to think of it, this may not be the best idiom to use because doctors have actually been setting the world on fire recently.

The medical profession as a whole, with preciously few exceptions, engaged in a very serious breach of faith against their patients and the public when they willingly participated in forcing an untested vaccine upon the western populations recently. They recommended a vaccine they knew nothing about as a cure for a virus they knew nothing about – without even carrying out a basic risk-benefit analysis. Doctors are big on risk ratios, risk indexes and risk factors when it comes to ‘preventive medicine’ – but not when it comes to vaccines which may or may not insinuate themselves into the human genome. Things could be better in doctor land these days.

There has been a crisis of credibility and a crisis of professional behavior unfolding in the medical profession for a long time – like a slow motion train wreck. People started to become aware of it when Professor John Ioannidis wrote a seminal essay on published medical research in 2005 where he argued that a majority of all published research in medical journals is false due to sloppy methodology and other causes. He furthermore suggested that up to 90% of all published medical research is flawed and basically useless. Others have reached similar conclusions.

Read up on this here:

https://en.wikipedia.org/wiki/Why_Most_Published_Research_Findings_Are_False

https://link.springer.com/article/10.1007/s00192-017-3389-1

The response of the medical profession to these devastating results was to change nothing and to continue publishing sloppy and fake research. In the nearly two decades since the essay was published, things have not improved. Instead they have most likely gotten worse. During that time sloppiness in research has been matched with dishonesty as doctors are increasingly working as drug pushers for the western drug cartels – regardless of whether the drugs they prescribe are safe or not.

In addition to the vaccines, doctors push drugs which are almost certainly useless at best and harmful at worst. Statins, for example, are the biggest class of drugs prescribed in the West despite the fact that research is unclear on whether they prevent heart disease or cause it. They prescribe the latest fashion weight loss drug Ozempic even though it is becoming clear that it is outright dangerous. They prescribe hormones and antidepressants which interfere with or damage the emotional system of the brain without considering the consequences – which are often violent. An enormous number of drugs are pushed despite being clearly dangerous, mostly or completely useless, and run through fake drug trials. Cheap drugs which actually work are enthusiastically discredited to protect the drug cartel’s interests. This is standard behavior in the medical profession – and as always we are left to ponder whether stupidity or malice is to blame.

Medicine can roughly be split up into two fields. Firstly there is the professional field which includes highly trained experts such as surgeons. These people are the engineers and artists of the medical profession, constantly developing and improving their methods and equipment. They can reattach limbs, perform minimally intrusive heart surgeries and remove tumors from the brain. Their accomplishments have been magnificent. Secondly it is the ‘causal’ or ‘preventive’ field, where causes of disease and the effectiveness of treatment are investigated through research. The causal field is the basis of disease prevention and partly of drug treatments. ‘Causal’ medicine rests upon all the fake research Professor Ioannidis and many others have written about. The entire causal field and the doctors which practice causal medicine all float on top of a giant steaming pile of research excrement where they can pick any result they want and reach any consensus they want.

This means that a doctor can be a fantastic professional who is extremely competent at his job, while at the same time, he can turn into a blithering idiot when he starts talking about causal research and treatment based on it. This is an unfortunate duality for the profession.

How did it come to this?

Cause and effect

One of the very big issues in science – and indeed in all other aspects of life – is to identify true causes of events, diseases, and everything else. The main task of any researcher is to isolate and identify true causal variables. This is the main purpose of the field of scientific methodology, particularly research design. Unfortunately, modern academia seems to have stopped teaching methodology to a large degree, replacing it with statistics. Statistics is not methodology – it is just a tool used in the field. This simple fact doesn’t seem to be well known these days, including among statisticians and, of course, among doctors.

The problem is greater than just lack of knowledge and training in methodology. A very large part of our ‘expert’ communities doesn’t even seem to be aware of the need to isolate causal variables at all. In some cases, it is even purposefully ignored or discouraged. Identifying causal variables can, after all, be ‘problematic’ for the Consensus and the scientist. Who in their right mind wants to isolate the true causes of gun violence, the Ukraine war, the cancer epidemic, the diabetes epidemic, the ADHD epidemic, the heart attack epidemic, recent excessive unexplained deaths, the decline of the West, or Boeing’s ‘quality control’ crisis? Well, only people who are outside the relevant ‘expert’ community and can cope with persecution and job loss.

Isolating causal variables is more difficult than you might think. If we are looking at a dependent (effect) variable, say ‘frequency of heart disease’ and want to find its causes, we will be faced with a complicated situation. Possible causal variables (independent variables) will need to be identified and tested (under controlled conditions if possible) if we want to isolate the true causes. The problems do not end there because possible causal variables have certain inherent issues:

There are often more than one, and sometimes many. If you do not identify them all and include them in your analysis, the missing one(s) may radically alter the predictive value of the ones you are using. A causal variable outside your analysis can influence your analysis drastically, and even make non-causal variable look causal. Statistically driven methodology tends to regard a statistical model as representative of reality if it has high predictive power. That assumption is false.

Causal variables often form ‘chains.’ A good example is the neurotransmitter serotonin and assorted mental problems. A serotonin reuptake inhibitor is a treatment for certain mental issues. Serotonin is increased and the issues supposedly go away. Now, is the cause being treated? Not at all. Serotonin deficiency doesn’t happen by itself. The neurons don’t just decrease serotonin production for the hell of it. Something else causes the drop. What causes the drop may even be caused by something else. Causal variables often line up in a chain starting with the ‘real cause’ – or the primary causal variable. It then affects a secondary causal variable, which may affect a tertiary causal variable – and so on. Medical research focuses on the immediate causal variable – i.e. the one closest to the effect. That variable is then treated and search for the real cause stops.

Variables often interact – which means that a variable is only causal as long as some other variable has a certain value. An example of that could be ‘stress only causes anxiety if the person has certain personality characteristics.’ If the characteristics are different, stress will have other effects or fewer effects/symptoms. Causal variables often form an interacting ‘cluster’ which can be difficult to untangle. The most common strategy to deal with this is to ignore the interactions and assume everybody and everything is the same and to apply general treatment.

Reductio ad duas

The three issues mentioned above are far from the only issues needing to be addressed in research design – but they are difficult enough to deal with on their own. Addressing them requires serious methodological skills and resources. These skills are rare and sometimes the resources aren’t there. In medicine those skills are very rare, and I don’t think I’m exaggerating when I say that the medical profession is notoriously bad at research and methodology. This doesn’t apply to all doctors – thank God. I apologize to the very few of them for this generalization.

Modern medicine has developed a strategy to deal with all this complexity. I have given that strategy a pompous Latin name; ‘reductio ad duas’ – or reduction toward two. This refers to the tendency in medicine to only work with two variables at the same time – a single independent variable and a dependent variable. The independent variable is a presumed causal variable which may or may not be primary, and the dependent variable is the effect – usually some disease or symptom of a disease.

To illustrate what reductio ad duas means and what kind of consequences it can have, I will use an example:

There is a link between smoking and heart disease (CVD). The link is visible in a strong correlation between the two. Let’s say it’s r=.60. This would mean that smoking explains 36% of the variance in the heart disease variable. The common way researchers deal with this is to transform this 36% into some kind of risk ratio which can be used like a formula. You can perhaps even plug a different number of cigarettes per day into the formula and see the risk factor/ratio change. You may even be able to ‘calculate’ by how much an additional cigarette per day shortens your life.

This is also done when multivariate analyses are used – i.e. when there are more possible causal variables than smoking included. If smoking explains 36% of the variance in a multivariate model, it’s treated the same with a risk index developed based on that percentage.

This is where things get interesting. Let me explain why:

1.       A risk index/ratio is a ‘consumer product.’ It is used to convince patients to stop smoking and doctors to press them into stopping smoking.

2.       It becomes a goal in itself, an entity on its own, independent of other possible causal variables which may even interact with it. This means that many doctors stop thinking about other possible causal variables and almost solely promote this ratio to patients.

3.       Even if a multivariate method was used, the 36% does not represent the ‘causal part’ of smoking in heart disease. It is more likely to represent the absolute maximum it can be.

4.       Any causal variable not included in the analysis will almost certainly (upon inclusion) decrease the ‘share’ of smoking in heart disease.

5.       Other causal variables inside the analysis are almost certainly overestimated too because of variables missing from the analysis.

6.       The probability that all relevant causal variables have been included in any analysis is zero.

7.       The probability that the 36% is too high is 100%. The probability that the risk factor/ratio is wrong is 100%. The probability that it is overestimated, perhaps massively, is extremely high.

8.       The overestimation of the effects of smoking may be massive. Instead of 36%, it may actually be 20%, or 10% or 5%. No one knows.

9.       Despite this, doctors will promote the ratio, pressing patients into stopping smoking. They will do this at the expense of other possible causal variables, such as stress, ultra-processed foods, seed oils, couchpotatoitis, etc.

10.   Some of these other variables may be as important as smoking or even far more important.

11.   Because the doctor has fixated on the two variables (smoking>heart disease) and their ratio, he will neglect causal factors which may be seriously critical and damaging to the heart. Research will also ignore these variables and leave them out of statistical models.

12.   The research into the causes of heart disease, and its treatment, has thus been reductio ad duas – into a single risk ratio representing only two variables whose specified relationship is without any doubt overestimated.

13.   This heavily influences medical research in such a way that researchers focus on one causal variable because they are looking for a ratio to promote to patients and to other doctors.

14.   The constant search for sexy risk ratios thus drives doctors into sloppy research which becomes fragmented and useless because it is seeks out single causal variables rather than complex causal patterns.

This does not mean you should smoke like a chimney – you shouldn’t. But you should consider other factors too, including stress and what you eat. If your doctor does not mention these factors as potentially as important as smoking, then he is misinforming you. If he doesn’t express at least partial ignorance of causal factors to you, he is misinforming you. He is pretending to understand something he doesn’t – in most cases anyway.

Transforming a share of variance (such as the 36%) into a risk factor/ratio should never be done. The ratio will almost always be false, and almost always overestimated. Medical research in general, and the practice of ‘causal/preventive’ medicine, is heavily based on these ratios and the reductio ad duas of causal variables. This is what is partly responsible for fragmenting and destroying medical research – along with massive incompetence when it comes to scientific methodology which in many cases includes the inability to differentiate between cause and effect.

The issue of diminishing competence

The simplification of medical research is not only because of the obsession with ratios. Medicine is a very big field and a very complicated one. The complexity level of most subjects and research ventures is very high. In other words: medicine is difficult but it has been reconfigured as easy. The complexity level and the level of difficulty is such that only few people would be able to conduct proper research and create proper theories and models – even if methodology was properly taught in medical schools.

Medicine has been growing as a field and the number of doctors has increased a lot. As fields grow, the average IQ and the average skill level drops. A smaller and smaller proportion will be able to do proper research and theoretical work. At the same time research output increases along with the increase of people in the field. Consequently, larger and larger proportion of research will be carried out by people not suited for that kind of work. The result is an increasing proportion of bad research and bad theory. This is a bigger problem the more complicated the field is because more and more simplification will be required.

We see this very well in psychology too. Psychology has grown massively in recent decades as craziness increases in the West. Subsequently the average IQ of psychologists has been ‘trending downward’ and is now probably close to the average IQ of bus drivers. The result is the ‘replication crisis’ in psychology caused, among other things, by utter incompetence of most psychology researchers.

We also see this in universities in general. As more and more people receive university education, the lower the average IQ and competence of the students will be. Now undergraduate students have almost trended down to the mean IQ of the general population. These students will soon be doing medical and psychological research.

https://heartlanddailynews.com/2024/01/college-students-average-iq-has-fallen-17-points-since-1939/

As stated earlier the response to this complexity and decreasing competence has been the simplification of medical research, including the reductio ad duas method. This effect is very obvious if you browse through published research. There is an unbelievable number of research papers which only focus on one causal variable along with a single effect. All these papers are more or less useless because they show inflated or non-existent effects.

In the last decades, or indeed the last century, a health crisis has been rising in the West. Every single aspect of both physical and mental health has worsened – and a significant part of people are not even able to have children anymore. Let’s say that we wanted to create a framework to explain why this is happening – some kind of theory or explanatory mechanism we could test. As a result we could suggest how our society should be changed to stop these trends.

Let’s say we started by reviewing the current medical research base looking for clues. It is likely that would not yield any useful conclusions. A research base where 90% or more of research works with single causes, provides inflated effects, and has sloppy methodology, will be useless. This means that in order to truly understand and solve the problems we have, we would have to discard it totally and start from scratch - because the purpose of the research up to now has not been to solve health problems at all. The purpose has been to sell drugs and to become ‘famous’ so people will be recognized when they go to medical conferences.

Is there a solution?

It’s easy to criticize but solutions are hard. However, the first step is always to understand the problem and its causes. It is clear that the research problem in medicine cannot be fixed without massive structural changes in the field. Those changes are not likely to take place anytime soon because they would turn the medical profession upside-down. Still, I’m going to list a few of them.

·         Methodology of science must be a part of every medical program in every university. The scope should not only be research, but also treatment of patients and advice to them.

·         Medical education should be changed to emphasize and to accept uncertainty in causes and treatment. Doctors need to be forced away from simplicity - and certainty based on simplicity. Uncertainty is far better for patients, and for the doctor’s development as a professional.

·         Hospitals, insurance providers, and other parties should be severely limited regarding the information they can demand from doctors. This would cut down on doctors’ paperwork and make them more available for patients. This would consequently cut down on the demand for doctors in general, resulting in less competence dilution in the field.

·         The use of risk ratios and risk indexes should be banned – both in research and patient consultation.

·         Single-cause research should as a rule be rejected by medical publications, as well as meta-analyses of single-cause research or which derive single cause results from multi-cause research.

·         Drug companies should be banned from financing research by practicing doctors. This will decrease the incentive for faking results and decrease the volume of bad research.

·         Practicing doctors should be completely banned from participating in drug trials. They should only be allowed to replicate and criticize them. That would turn them from accomplices into agents of quality control.

·         Doctors should be banned from taking payments from drug companies and from going to conferences financed by them.

In general, a clear split must be engineered and enforced between doctors and drug companies to remove the monetary incentive for falsified research and dishonest advice to patients. Any relationship with a drug company should result in a dismissal. You either work for patients or a drug company, not both. Setting doctors outright up as adversaries to drug companies would even be better than the current situation.

All this sounds extreme but it must happen if the situation is to be improved. Of course it won’t happen – we all know that. Corruption is profitable and admitting that a large part of the profession is incompetent damages the credibility of all.

I would very much welcome suggestions on what else could and should be done.