Please use the form below to submit comments. Also provide an e-mail address and name. Your e-mail address and/or name will be used only to communicate with you about this or future comments you may submit. I am particularly keen to receive references to published material that contradicts the assertions and arguments I have made.
By submitting the above comment, I grant to Ross Alan Hangartner the right to incorporate the comment in full or in part, literally, paraphrased, or conceptually, as he sees fit, into State of Pain or other writings that he may create in the future. However, I don't grant permission to include my name or e-mail address, or to use them in any other way than to contact me for follow-up. I understand that by submitting the comment I acquire no right of any kind in State of Pain or other writings of Ross Alan Hangartner.
Last updated: Fri, Jun 30, 2017
I happened across the following passage in a chapter titled "Assessment of Patients with Chronic Pain" in the Handbook of Pain Assessment. ("Pain behavior" means any behavior of the patient that might be related to pain.)
The challenge to an examiner is to interpret the pain behaviors of the patient....[Although] pain behaviors are sometimes determined entirely by an abnormal biological process in a patient's body, they are typically also influenced by various psychological factors and by the social environment of the patient....A useful way to conceptualize this challenge [interpreting pain behaviors] is to think of a regression equation....1
A regression equation is an algebraic equation that is used to analyze the results of an experiment or to form a mathematical model from the results. If you did an experiment to test conservation of momentum, your regression equation might simply be p1 = p2, the conservation of momentum equation that we saw earlier. After you conduct your experiment, the computer will calculate how well your data “fits” the model (the equation). Your regression equation could include additional factors, for example, p1 = p2 * l, where “l” stands for the light intensity in the next room. In this case, the computer should tell you that knowing the light intensity in the next room really doesn't help to explain momentum, so you wouldn't add “l” to your model.
The equation for pain behaviors suggested by the authors of the above quote is this:PB = f(Xa1, Xa2, Xa3,...Xan, [biomedical factors] Xb1, Xb2, Xb3,...Xbn, [nervous system sensitization] Xc1, Xc2, Xc3,...Xcn, [psychological factors] Xd1, Xd2, Xd3,...Xdn) [systems or contextual variables]
The equation says that pain behaviors are possibly caused by many biomedical factors, many factors related to nervous system sensitization, many psychological factors, and many systems or contextual variables. (Systems or contextual variables
in this context may be a euphemism for an unresolved claim for disability benefits.) Which factors should be included in the model may or may not be known, they may or may not be measurable, and the interrelationships among the factors are not known. For many of the factors it is not known whether they would increase or decrease pain behaviors.
This equation is meant to convey the difficulty that a doctor faces in assessing a patient who complains of pain. A very similar equation could apply to models of the pain experience. The particular equation above is probably sufficiently complex as to be unprovable in any practical way. Although this depends a good deal on the (unknown) relationships among the variables, the number of subjects (or observations) required can reasonably be expected to increase exponentially with the number of factors being considered. In other words, when you double the number of factors, you square the number of observations needed, and when you treble the factors, you cube the number of observations needed.
Some models, if accurate, can be proven. Other models, even if correct, can not. This issue of complexity is one of the reasons that much of the experimental evidence about pain is about simple situations and simple pains, simply because a simple model can be proven more easily. There's nothing inherently wrong with research on simple, controlled situations. There are, however, important consequences. Two chief ones are; 1) that you can't safely generalize the results to more complex situations and 2) that you don't know how other factors interact with the factors you studied. There is much reason to believe, moreover, that pain is a complex phenomenon, and therefore much of what is proven about pain, although true in a limited context, is less true in a broader, more realistic context.
If you will now, turn your mind away from issues of provability and return your focus to the mundane problem of the doctor assessing a pain patient. What are the factors that the doctor supposes drive pain behaviors? Given that the equation itself is unknown, why does the doctor feel it is necessary, appropriate, or ethical to fill in the unkown factors? What errors will be made, and who will be affected by them? To learn more, continue reading.