Have you ever heard the expression, ‘If it seems too good to be true, it probably is.’
With respect to musculoskeletal medicine, the diagnostic paradigm has a stranglehold on the health system, but I’m going to put forward that it’s a paradigm that sometimes does not fit. Is it time we started to re-think our approach? In some contexts, does the diagnostic paradigm work against what we want to achieve?
Is the relative simplicity of the diagnostic process meant to reconcile something so exponentially complicated like musculoskeletal pain conditions? Can something so complex be boiled down by such simple ‘labels’?
Setting the scene
Like me, you may deal with patients attending with poor narratives, poor thinking and labels for the reason of their pain that they hold dear. It’s not as simple as blaming the label, of course, but one of the main reasons that hold patients back from a state of ‘action’ in dealing with a painful condition is the lack of an appropriate label or diagnosis. We’ve all seen it,
‘I’ve been to 5 specialists, 10 physios, a chiropractor, a naturopath, a horse-whisperer… no-one can tell me what’s wrong, everyone has given me a different diagnosis.’
This patient has been used and abused by the diagnostic paradigm, and is firmly stuck on the treadmill. Once again, it’s not always as simple as blaming the previous clinicians for being poor diagnosticians or that the patient has been just been given the incorrect diagnosis over time. Don’t get me wrong here, either, the skills of the diagnostic process (and here, I don’t mean our clinical reasoning process, because they aren’t actually the same thing) are important – integral even. And in my team, I can say we take pride in being excellent diagnosticians who generally deal with being ‘the last resort’ or a ‘tertiary referral centre’ amazingly. But in my team, we see predominantly Workcover cases; many of these are complex to begin with, made even more complex by the diagnostic process. It doesn’t seem to help these patients, in fact, it seems to hinder them. Searching for a diagnosis keeps them stuck. Another label is actually not the thing they need.
Having – or not having – a diagnosis is not the main issue preventing these complex patients from getting into an ‘action’ stage, it’s that the diagnostic paradigm has produced an unrealistic expectation. The system has told them that a label is exactly what they need to get better, and they come to trust this expectation. In some cases trying to fit a diagnosis on a painful condition is, as we all know, like trying to dress up a heaving wild boar. Sooner or later that pig is going to realise that its new clothes don’t really fit well and he/she would prefer to go and roll in the s*&!.
What are we doing when we diagnose?
“Diagnosis may be defined as the determination of the cause or nature of an illness by evaluation of the signs, symptoms and supportive tests in an individual patient. Diagnostic criteria are a set of signs, symptoms, and tests for use in routine clinical care to guide the care of individual patients.” Ref
Evaluating signs and symptoms using supportive tests to guide our care. That is at its simplest definition; it is a logical process and one that works fantastically well in other disciplines of medicine.
But are we doing something else when we diagnose? In a complex world with no absolutes, I think the use of diagnoses shouldn’t be so readily bandied by practitioners. Although some progressive specialists and practitioners actually can sit in the ‘grey’, the unfortunate world we live in dictates that the patient expects an absolute. They are conditioned to believe if they don’t have a diagnosis, then they can’t improve.
Complicated is an understatement. If you think of a specific internal medicine specialty, the boundaries are fairly well established. There’s a general normative range. There are – at least a handful of – definitive tests.
In musculoskeletal medicine and pain conditions resulting from musculoskeletal sources there is the inherent complexity of the human being that is experiencing the pain. Then there are the multitude of tissues, sources, movement dysfunctions, control issues.. the list goes on. Pain drivers are wide and varied, this is a long complicated process to get across to a patient. The reductionist approach – providing a simple cause and effect model for a patient – does yourself, and definitely the patient, a disservice. It enables the ‘pill for every ill’ message to continue pushing forward, which in turn creates more doctor shopping and increasingly extreme treatment approaches.
Better education is needed throughout the system as a whole, we can all agree on that. That means this needs to start at the very beginning, with the diagnostic process; our education reflecting what is important about the process – the steps to improvements rather than the label itself.
From the very simple – the bog standard knee meniscal tear – to the complex presentations, diagnosis is complicated.
I think we are all aware of the clinical shortcomings of physical tests for many presenting conditions, Thessaly’s, McMurray’s and Apeley’s all receiving about 50% accuracy when compared to MRI (the gold standard) (Ref). The same can be said about a more complex joint, the shoulder – the pathoanatomical approach is just horribly poorly performing.
In more obscure presentations, like fibromyalgia, the diagnosis is more like a crap-shoot (Ref). In cases like this, in particular, is another ‘label’ what this patient needs? Without this, won’t they still get best standard care?
The traditional approach
The pathoanatomical approach is the favoured approach by many. If you think this approach is completely appropriate in 100% of your patients, then do me a favour and just stop reading, like now…
It is useful, and I actually think the pendulum has swung very far away from this approach, to the detriment of many patients. I’d like to see the pendulum settle back and this approach is considered more useful. We all know pathoanatomy is found in asymptomatic patients, but something that irks me about using this argument to explain pain, is it includes a logical flaw based on information that you currently don’t have. You can’t prove that the pathoanatomy is NOT causing a patient’s pain, so saying it doesn’t play a role is sometimes a very inaccurate statement. The absence of proof is not the proof of absence.
We have exploratory work to create more relevant diagnostic systems. A great read into creating a movement system – or pathokinesiologic – diagnostic model by Ludewig and colleagues (here), is a step in the right direction. There are some tools and approaches designed to assist diagnosis (again, not to get confused with the clinical reasoning process). The pain and movement model can be used to educate patients but also provide a more multi-faceted approach to diagnosis. Neuro Orthopaedic Institute (I’ve always thought it such a misnomer??) approaches, as above, push the pendulum very far one way; this approach is very useful and is a far cry from the woefully inappropriate pathoanatomical approach but is it also too far gone to enable access to surgeons, GPs or those with a much more ‘biomedical’ background?
As for diagnostic classification systems, well, we have literally thousands of these relating to specific bodily regions. Based on region, duration of symptoms and irritability they are useful for communicating with other health professionals and third parties (like insurance companies), for research purposes and clinical practice guidelines.
It is all designed to categorise, label and make some pretty complex stuff fit into nice neat little boxes. Here’s the theme; it’s not that simple.
What do we do about it?
I think a few main approaches need to be explored:
1. Being careful with exactly when and where a diagnosis is required. In absence of changing the diagnostic-curative paradigm and the system as a whole, we should consider placing a diagnosis on some people in the same light as other nocebos.
2. Avoid chasing the diagnosis for a patient, for the above reasons and it will be another thing that holds them back from real behaviour change.
3. A review of those who are in a position to diagnose and those who aren’t. The use of simply pathoanatomical labelling should be thought of as a risk to patients, and those that don’t have up-to-date training need to fulfil a requirement to do so.
This is incredibly important in compensable settings, where the only professionals legally allowed to give the most meaningful diagnosis are the ones with the least training in musculoskeletal diagnosis (GPs).
Please don’t mistake this post for a rail against the process of diagnosis, simply a discussion around its utility in the clinical environment and the musculoskeletal context as a whole. Because in a scenario where physiotherapists are the primary diagnosticians in compensable environments, I honestly don’t think the end result would be much more different. Those likely to ascribe poor meaning to a diagnosis (even if correct) would present with the same rumination, lack of action and negative beliefs.
Should it be given less value? I think so. Should it change form? In some cases, yes. In these cases, I think it is detrimental and a diagnosis should be labelled in the same fashion as any other nocebo. Just as our language around certain elements of practice has changed to avoid nocebo, our approach to traditional diagnoses should also change.
A re-think is needed in my opinion, a re-working of when and where a diagnosis is required. Buy-in from the system as a whole to ensure a simple ’cause and effect’ model is not always appropriate would be great. I think the most important thing to get right, is the knowledge and practice of when not to give a diagnosis and, if every stakeholder supports this decision, the patient can get on with the important thing – getting better.