There’s nothing like a good circular argument, what our esteemed Latin ancestors called circulus in demonstrando. You would think that scientific papers that focus on theories of why patients are disabled must be well grounded in objective measurement. After all, this is a topic that can determine whether or not some poor, injured soul gets to feed his family. However, for the pain neuroscience education crowd, circular reasoning has become a staple of the research house of cards on which the movement is built. A new paper in aboriginal peoples with low-back pain is a great example. Let me explain.
The Circular Argument Is a Trap That Has Been Plaguing Pain Researchers for Decades
Pain is tough to quantify, but in the past decade or so, we have actually found a few ways to measure it with advanced technology. However, historically, since injury and pain were difficult to measure, a classic pain research trap is to tie beliefs about disability to more physical disability and then conclude that the disability is caused by the beliefs. To show why this is a problem, let’s say our ability to diagnose what’s wrong with the complex systems in your car was limited. Let’s also say that you drive a “lemon” that sometimes starts in the morning and sometimes doesn’t start. You begin to develop the belief that you have, in fact, purchased a lemon. However, we could also say that since we have no way to tell if your car is, in fact, a lemon, that your beliefs about your car being a lemon have caused it to be a lemon.
If we suspend our thought experiment for a moment, we can examine the car to see if it’s a lemon or has significant issues in mechanical or electronic structure or function. If we did, we would see that the car is broken; hence, our logic connecting your beliefs to the car being a lemon as the reason it wouldn’t start are circular and seriously flawed. In conclusion, what breaks up the circular argument is the degree to which we are able to examine the car objectively.
Pain Studies That Don’t Include Complex and Objective Diagnostics Are Just like Our Lemon Study
While it’s easy to see the ridiculousness in the above study of a car that is believed to be a lemon, which, therefore, causes it to be a lemon, sometimes in pain studies, the best reviewers miss the obvious circular reasoning. Why? Because the causes of chronic pain are hard to diagnose and because most reviewers are frustrated by these patients themselves; hence, there is an inherent bias in believing that there is nothing wrong with these patients. However, our ability to diagnose pain has gotten better and better. In other words, our ability to objectively determine if a patient’s body is “a lemon” is improving.
Let’s take a simple study where we examine the beliefs of patients with chronic low-back pain by giving them questionnaires. We observe that the patients with the most pain and disability have the most significant beliefs about their own disability. For example, they score higher on questionnaires measuring thoughts and negative beliefs concerning disability. Hence, we conclude that it is the patients’ beliefs that cause their disability. As a corollary, we also conclude that we can change those beliefs through counseling, and, therefore, we can change their disability.
In our above study we, of course, are missing one thing: the ability to objectively tell if the patients who are the most disabled really are the most injured and, therefore, they have every right to have the most thoughts about their own disability and a pessimistic view about recovery. This is just like our car study, when we suddenly examined the car to determine if what we thought was a lemon really was mechanically a lemon.
So how could we do this study properly? It would be quite complex. First, we would have to perform extensive imaging of the low back, including using newer techniques to image the brain for pain responses (fMRI) and/or atrophy. Next, we would require double diagnostic blocks with blinded short- and long-acting anesthetic of all of the major structures of the low back to rule in or out that any one or more areas were causing pain. We would need movement-based low-back imaging looking at quantitative movement analysis. We would need multiple physicians to examine the patient and determine whether he or she was credible with regard to having a real chronic pain condition or was considered not credible. You get the idea. Doing a study like this right would be a very expensive affair, which is why 99% of these studies take the cheap and unethical way out—they skip all of this testing.
The New Aboriginal Pain Study
Now that you have an idea about how to construct a high-quality chronic-pain study looking at how beliefs can alter pain, let’s take a look at a study that was just published that was lauded by the PNE, chronic pain denier, set. Remember, this is a group of physical therapists who are trying hard to deny patients in chronic pain appropriate treatment by trying to convince them that their pain problem is really just hyperactive nerves and should be ignored.
The new study was performed with native aboriginal people in Australia. First, it’s very clear in the introduction of the paper that the authors have begun with a conclusion that beliefs about pain and structural changes in the spine define disability due to chronic low-back pain. They make absolutely no attempt to hide this bias:
“By influencing or perpetuating negative beliefs, there is the potential for healthcare practitioners to have a negative impact on their patients disorder. Negative perceptions about CLBP reflect widespread sociocultural attitudes in Western societies and are hypothesised to underlie the 20th century ‘epidemic’ of CLBP disability.”
Also interesting is that they claim that aboriginal peoples are “a population previously identified as protected against the disabling effects of low back pain due to cultural beliefs.” Reading the introduction confirms the origin for this statement is research that has shown that while aboriginal peoples do have low-back pain, they don’t seem as disabled by the phenomenon. This makes common sense as they live in a society where there are few options to treat the back pain and not working through the pain is not an option. We also read that other studies have actually shown the opposite, that aboriginal peoples are disabled by pain.
The researchers interviewed the subjects about their pain and beliefs and recorded and then transcribed their conversations. The researchers found that those subjects who described the most pain and disability also had the most descriptions about how some medical professional had told them about their low-back damage. That sounds about right, as the patients with the most disability would have been expected to see the most medical practitioners, would have been imaged the most, and would, therefore, recall more about what they had been told was wrong with their back. However, the researchers interpreted all of this to mean that the disability had been caused by the medical professionals who spoke with the patients.
The researchers also found that patients who had a more negative outlook on the future had the most pain. The researchers interpreted this as, the false perception of pain and damage (incorrect thoughts planted by evil healthcare practitioners) were influencing the patient’s outlook on life.
At no time did the researchers actually attempt to run all of the tests discussed above to determine if these patients did have physical problems that made them legitimately disabled. They didn’t try to find out what was wrong at all. What’s bizarre is that the researchers never even looked at the medical records of these patients, which would have been readily available in a country with a socialized medical care system.
How Did This Really Bad, Internally Conflicted Study Get Published?
How is it possible that a study that rests its conclusions on the idea that the subject’s spines aren’t really damaged never tried to determine if the patient’s spines were, in fact, damaged? How did this pass the most basic peer review? My guess would be that it was reviewed by other PNE pain deniers who fast-tracked it through the publication process.
As you can see above, the study is the worst case of circulus in demonstrando I’ve reviewed in a long time. Drawing conclusions about causation based on shoddy evidence. It’s also an example of reverse causation, or the concept that correlation doesn’t mean causation. For the latter, just because the level of disability and the belief about disability are correlated, doesn’t mean that one caused the other. Or in our car example, just because people with cars that are lemons believe more often that their car is, in fact, a lemon, doesn’t mean that believing your car is a lemon causes it to be a lemon!
The upshot? The PNE denier crowd loved this paper and for good reason as it gives a veneer of credibility to their claims that we healthcare providers disable patients. While I personally believe that there is some kernel of truth to the idea and that certain patients are more susceptible to this type of influence, the PNE chronic pain deniers try to take the argument way beyond its actual clinical importance.