Pathways to well-being: Untangling the causal relationships among biopsychosocial variables, 2020, Karanamuni et al

Paywall, https://www.sciencedirect.com/science/article/abs/pii/S0277953620300654
Sci hub, https://sci-hub.se/10.1016/j.socscimed.2020.112846

 

Untangling is the last thing they want to be doing, the whole thing relies on smokescreens, changing definitions and concepts, and general confusion inpenetrable to mere mortals. Without that, it's nothing. One of his colleagues had better stop this chap untangling before it's too late.

 

The key to untangling the causal relationships is to study all of the variables prospectively in the same study.

The key to making a mess of all the factors is to study them separately and then write endless hypothetical papers about non-specific associations published in social science journals.

Illnesses and diseases are perpetuated by highly specific feedback loops, not linear and abstracted "P->B" relationships...

The article by Karunamuni et al. reads like an undergraduate essay - mention everything that could be vaguely relevant and hope something sticks (though to be fair, many medical doctors write manuscripts overstating similar non-specific findings).

In the cited reviews examining relationships between objective health outcomes and subjective well being, the negative outcomes were all on likely behaviourally mediated outcomes, namely obesity, high blood pressure, lower fitness, smoking, heart disease and so on. Yet Karunamuni states:

I mean, maybe, but no one has ever found higher than 'suggestive' quality evidence for those claims. (the evidence is not of sufficient quality to state causative relationships)

 

Continuing,

Wait, what?

Dowd et al. Is "Social determinants and BCG efficacy: a call for a socio-biological approach to TB prevention."
Li et al. is "Editorial: Why Vaccines to HIV, HCV, and Malaria Have So Far Failed—Challenges to Developing Vaccines Against Immunoregulating Pathogens"
Phillips 2011 is another essay in a low impact sociology journal, providing no direct evidence.

Dowd et al only cites Phillips 2011 as evidence of psychosocial stressors affecting immune responses to vaccines...


Phillips in turn cites Pederson 2009 "Psychological stress and antibody response to influenza vaccination: a meta-analysis."
https://www.ncbi.nlm.nih.gov/pubmed/19486657

But they seem to be mistaken that non-normalised antibody titres post vaccination mean anything about the likelihood of a successful immune response. It is only a bad sign if no titres are detected at all. The immune response can only be determined with a subsequent infectious challenge, evidence of which is not provided in this review and results of the review are also potentially contaminated in an uncontrolled manner prior exposure to influenza or influenza vaccines.

An example of a prospective study, a succession of 3 HepB vaccines for medical students found no effect of anxiety on antibody titres at followup, nor rate of seroconversion (ignoring the misleading cherrypicking of results in the abstract). https://www.ncbi.nlm.nih.gov/pubmed/1553399

On the contrary should be acknowledged that those are not specific, nor empirically tested models of disease/illness. Instead these 'models' rely on the usual junk methodology. Namely: lets pick a bunch of nonspecific factors and draw arrows between them based on our preconceived ideas about bio-psycho-social relationships and call it a model.

 

This remains unchanged. This pre-science natural philosophy approach of putting your brain and intuition to a problem without taking account objective reality never works and only produces rhetoric that things "appear to", "can be", "may be", "potentially contribute", "can explain". This is the process that created the celestial spheres, which certainly appeared to explain things even though they didn't. It remains plagued by an obsessive belief that psychology must, absolutely HAS TO, be the most critical factor. As long as you begin with a conclusion in mind you will always fail at un-anchoring from those beliefs.

Re-framing is not further explaining. And re-branding for the sake of re-branding is not significant. This model remains one of the most vague, non-specific and non-predictive constructs among all of science, unchanged by another poor attempt at justifying this massive escalation of commitment to yet another set of brands of old pre-science beliefs, whether they be BPS, FND or any of the identical acronyms deployed to create confusion.

Clinical psychologists and psychiatrists seriously need to get some training from economists. Economics has already gone through this phase of using simplistic models that assume all other things being equal without actually making sure they are. As a field, economics has progressed enormously faster at acknowledging the flaws in their thinking. Mostly because economic failures are very calculable and there are enormous pressures against repeat failures. Psychosomatic ideologies, on the other hand, are always promoted no matter how thoroughly they fail, unable to learn anything out of the failures.

Ugh... case in point:

None of which can qualify the direction of causality, which is merely assumed to be as described here. Even though the other way around makes 100x more sense and actually explains things, up to and including the actual evidence base. This whole thing is infantile logic, wanting something to be true at all costs.