Explaining Pain using a multilevel model of pain

This blog follows on form some of the ideas in my previous two blogs, which highlight the problems of mind-body dualism and the usefulness of a biopsychosocial model which incorporates ontological depth. I have included below an A4-sized copy of a multilevel model of pain which I’m currently using with some patients within the Pain Service to explore some of the complex interactions which can be influenced to improve somebody’s outcome. It’s not suitable for everyone, but it has sparked off some interesting discussions with patients who wanted to understand more about pain. In particular, it seems to be helpful for people who have followed a purely biomedical model, and ended up struggling with a reductionist way of thinking about pain.

A purely biomedical model focuses on identifying an impairment (a fault with a structure or its function) which is leading to disease. The patient presents to a healthcare practitioner with symptoms, the practitioner takes a history, examines to elicit signs, and arranges any necessary investigations. This information is all collated to form a diagnosis, which then guides the treatment, which is focused upon fixing the fault. This is a model which most of us grew up with, and were probably trained in. Our patients are of course also used to it, and often expect us to be working with them to identify the broken part (the “Find It and Fix It” approach). I work in a secondary care Pain Clinic, so I meet a number of patients who have persisted with this way of thinking, but it has not led them to a successful outcome. Sometimes they are well aware of this themselves, and may have reached an understanding that they are not really expecting a cure. However, sometimes the “Find it and Fix It” approach is still their main aim. I may therefore spend a few minutes exploring these interactions using the multilevel model, and this seems to have helped a number of people to broaden their view of pain management to incorporate a number of other strategies.

The model makes reference to a number of domains which we know are relevant to the operation, development and prognosis of a pain problem. It is structured in a broadly hierarchical way, with smaller scale domains at the bottom, and larger scale domains at the top. This doesn’t mean that any one domain is more important than any other: they are all operating at once, and one domain may for a time be more relevant for an individual than another domain.

The model is consistent with a biopsychosocial model, with Louis Gifford’s Mature Organism Model, with the World Health Organisation International Classification of Functioning, Disability and Health (ICF), with a General Systems approach, and with Critical Realism, which is a “metatheory” which I used as a basis for my PhD. So, I don’t think there’s anything new in this way of organising ideas, but it does allow me to discuss quite a few issues with patients, all of which fit onto a sheet of A4 paper.

Critical Realism is one of a number of “integrated theories” which argue that we shouldn’t just be focusing on one domain of reality in order to make sense of what’s happening in the world. It suggests that reality is structured in a hierarchical manner, for example that psychological events are built on a foundation of biological events, which are at a smaller scale built on chemistry and below this, physics. This isn’t in itself a new suggestion, but it links to other interesting ideas. Firstly, it argues that biological events “underdetermine” psychological events: in other words, that our thoughts and feelings depend upon neurobiology but are not completely determined by them. This is an idea which is often called “emergence”: in plain English, this means that “the whole is greater than the sum of the parts”. There is a practical relevance to emergence: a burst of adrenaline (a biological event) may lead to a change in our thinking and emotions which may lead to a change in our communication with others and our behaviour. However, we interact with others and our environment at a different level of reality from the chemical level: somebody experiencing heightened adrenaline may end up running out of a supermarket, or scoring a goal if they happen to be playing football. All of these events are not just determined by neurobiology, but neurobiology is foundational to these events. What actually happens is strongly dependent upon context.

A further consequence of this way of thinking is that activity at one level can influence events at another level, with mechanisms acting which may not be immediately obvious. Clinically, I use this diagram to discuss with patients how, for example, a “significant other” not taking their pain seriously may affect the patients’ thoughts, which may in turn affect their autonomic nervous system, muscle tension, emotions, etc. It is easy to draw arrows on the table which link up these areas: each of these arrows represents a mechanism, a way in which events at one level of reality can influence events at another level. I did this once with a group of patients, and they were able to draw many arrows on the diagram, which immediately represents a complex system with interactions which patients recognised as “vicious circles”.

Once our patients are clear about the nature of these complex interactions, it is easy to understand why we would not expect a strong correlation (measured on a population level) between abnormalities in the tissues, and pain itself. However, partial correlations are still important within a systems approach: we shouldn’t overlook weak correlations if they are part of a broader, interactive model which is, taken together, a more explanatory one. It may be that they lead us to explore whether a combination of other factors (such as genetic factors, diurnal variations in cortisol levels, nerve root irritation, catastrophizing, poor sleep and reduced activity) which when combined together make a better model. This would require researchers from different specialties to collaborate to understand the emergent properties of complex systems that end up with suffering and disability.

multilevelModel