When we learn or teach new, complicated topics, we can almost never do so by imparting the entire complexity of the topic. We need to simply, or reduce things down to make them easier to communicate and understand.
When learning about how the human body functions, we often do so by reducing the body down to individual systems and localise functions to anatomy. E.g. the cardiovascular system is responsible for delivering blood, the respiratory system oxygenates blood, the nervous systems communicates messages from the brain to the tissue and vise versa, the musculoskeletal system produces motion etc.
We may also use a lot of similes to explain individual components, e.g. the muscle being like an engine, the heart being like a pump, the nerves being like electric wires, cartilage being a rubber tyre, articulating to surfaces within one another and wearing down slowly with use.
Reducing the complexity of something to make it easier to learn or teach is a brilliant way of imparting knowledge. The problem is that we can’t accurately explain the human body by reducing its function down to simple statements and similes.
Joint cartilage does help articulate two surfaces together like a car tyre, but its actual function is completely different to that of a car, with perceived ‘wear and tear’ not being predictive of dysfunction or symptoms (Ref, Ref, Ref). Nerves don’t just carry messages to and from the brain, but can also modulate these messages based upon changes in the brain, immune factors or hormones (amongst other things) (Ref).
Jacques Vaucanson’s Defecating Duck from 1738. Representing the concept of animals as machines. Originally made to show animals as machines, incapable of independent thought, but also illustrates the absurdity of reducing our understand of anything biological to just mechanics.
If a joint can function well with significant amount of cartilage worn away, is it accurate to say it works like a tyre? If a nerve can modulate the message it sends, is it accurate to refer to it as just a cable transferring information? As my friend Daniel Arbilla put it, “Analogies and metaphors can only take us so far before nuance is needed.” (You can – and should – check out his own work here).
When we understand the body in all its complexity, you start to realise it can’t be accurately reduced to a simple explanation. But why is this important?
New information, old beliefs.
The problem we face is that when presented new information, we can interpret it based upon all our previous information and beliefs.
We don’t actually think that cartilage operates like a tyre, but when explained is these terms we are more likely to believe it’s role is similar. So when presented with information that tells us that loosing weight improves symptoms of knee osteoarthritis, the logical interpretation is that it was the decrease in weight resulting in reduced physical load on the joint, reducing the wearing away and symptoms. And to be fair, there is evidence that for every kilogram of weight lost, there is a 4-fold decrease in knee forces(Ref).
However, if we’re not careful we might not read further and find that both exercise and diet is more effective than just weight loss alone (Ref). That exercise alone, without a change in weight results in improvements in pain (Ref). And that the relationship between weight loss, pain and function is not linear (Ref, Ref, Ref). This all suggesting that joint forces may not be the only factor at play.
If you read more widely, we find that there is also a relationship between obesity and hand/wrist osteoarthritis at similar proportions to reported rates as the knee (Ref, Ref). This challenges the idea that obesity is primarily about weight on a joint. As one study put it, “obese people do not walk on their hands”(Ref). We also see that there are groups of people who suffer from changes in nervous system function called central sensitisation, which typically results in hyperalgesia and allodynia, worsening symptoms regardless of joint load (Ref, Ref).
This is the same situation we’re in with many other entities within the human body.
Tendons were theorised to be like a rubber band and pain was as a result of tissue degeneration. However, local changes at the tendon fail to solely explain why tendons become painful, with no proportional relationship and many asymptomatic individuals showing ‘degenerative’ signs on imaging (Ref). We also know that tendons can both positively and negatively adapt to load, but we no current idea of the current mechanisms why it chooses to do one or the other (Ref). So does a rubber band analogy fit? Is there a better way to explain it?
We also see the idea that pain is a measure of tissue damage (Ref, Ref, Ref), pronation as a risk factor or cause of injury (Ref, Ref, Ref), the idea of the foot functioning like a spring (Ref) as just as a few more examples of simplifications that just don’t hold true.
But why do we need to make a fuss about this? Can’t we just ‘know’ that it isn’t an accurate representation?
Our need for replacement theories.
Well that isn’t really how our brains like to work.
In educational psychology studies we’ve found that science students engaging in a process of formulating and disconfirming theories were more likely to believe a hypothesis they personally disconfirmed when they had no other theory to replace it (Ref, Ref).
This is a huge problem as it tells us that, even when we’ve been trained in a scientific method, we’d rather believe something that makes sense but isn’t true than be faced with no theory at all. When we combine this with the fact the human body as I demonstrated above, isn’t able to be accurately reduced into theories and models, in addition to the myriad of biases and logical fallacies we are known to suffer, it is easy to see how even the most well intentioned practitioner can end up believing and applying theories that have no support or have even been debunked.
And we see this when we review practitioners management plans. One systematic review of Physiotherapists (which I am by no means saying is the only profession where this is a problem, just what this one study focused on), found that on average 63% of patients received recommended, evidenced based care, 27% received non-recommended care and 45% received management that had no recommendations at all (Ref).
In some common conditions like plantar fasciitis the difference is much more stark, with patients being prescribed care that had no recommendation 98% of the time, while only 29% received recommended, evidenced based care.
Table from Zadro et al, 2019
While the exact cause of this hasn’t been identified and is likely to be very complex and multifactorial (as will be the solution), one problem we can start addressing right now is how we think about how we think about the human body.
So what now?
There are a number of different ways we could approach this problem. In a discussion about models of Pain and movement, Todd Hargrove argues that we can have models and theories, but like in other disciplines such as physics, we have to figure out in what situation or context that model or simplification holds true and also when it doesn’t. I would highly recommend you read his article here.
Personally I have found the solution for me is relatively simple, but really difficult to put into practice; simply to follow the evidence.
We often end up in situations where we want to simplify things down because we feel we need to understand something in all its complexity. I would argue the biggest problem is the belief that we need to understand something completely and with 100% certainty to identify and manage it. I argue instead we should be happy to sit with the uncertainty of not knowing, than have the false certainty of believing a theory with no support (or worse, one that has been thoroughly debunked).
I argued that this is something we can be bad at with paradigms of foot function in my previous writing here. I also highly recommend reading the article Tolerating Uncertainty – The Next Medical Revolution by Simpkin and Schwartzstein.
Take patellofemoral pain for example. We’ve yet to identify what exactly the cause of the syndrome is or what our therapies are doing to assist. Yet this doesn’t stop us from identifying appropriate management strategies such as; exercise based therapies, load management strategies and a small amount of passive therapies that can assist in the short term, such as taping and foot orthoses (Ref).
Or how we’ve identified in one study that the orthoses modification that reduced the most strain on the plantar fascia was forefoot valgus (lateral) padding (Ref), going against the teaching of foot orthoses being an anti-pronatory device requiring varus (medial) posting. I don’t know exactly why this is the case, or have an exact understanding how this integrates with a model of foot function and pathology. However, it doesn’t stop me from applying this (and getting a good level of success if you want a little anecdotal evidence).
This seems a far cry from what we’re often taught, or what patients expect. That we should have a crystal clear picture of exactly what is happening, why it is happening and how we’re going to return it to ‘normal’. It almost feels unscientific.
However, at least for this moment, avoiding over-simplifications, tolerating the uncertainty of not knowing and following the evidence where we have it, even if it goes against our current models and ideas, maybe the most scientific thing we can do right now.