Humans are to be distrusted. Or so we are told. With our emotions. Vulnerability to error. Confirmation biases. This has lead to a not so subtle erosion of belief in the human subject. Whilst the general subject cannot be touched so is viewed with distain by empiricists. This leaves me asking does empirical evidence back this up. Should humans be distrusted? Always? Should we live alone on objectivity and measurement?
It is possible this is part of the legacy of scientism and logical positivism. So what are they? ! Logical positivism is a philosophy which states that something is only cognitively meaningful if it can be observed (verification) by the senses (empirically). Implicitly this values objects as these are able to be verified easily. Whilst stripping meaning from questions like “is this good/right?” or other ethical questions as they remain unverifiable. This was very popular in the 1930’s and 40’s with intellectuals such as Bertrand Russell, Alfred Whitehead, Hans Hahn and Rudolf Carnap. This means we have to see, hear, touch, smell, taste it. This fits a biomedical model perfectly. But this starts to break down if we believe that psychological, social, spiritual are in fact real or important. And irreducible to biology, molecules, atoms (contra to physicalism). It also fits the physical sciences well too. In fact under this guise and at its extreme Science becomes the arbitrator of reality and meaning. This is one reason why the term scientism has emerged. It is made carnate when we exclusively worship objectivity. We exhort that only matter matters. We forget both the presence and richness of subjectivity. And their intertwined nature.
Aside these philosophical discussions. Beyond the realisation that subjectivity is omnipresent. Is there data to suggest subjective measures are horrifically misleading?? Well patient satisfaction is a good place to start. It is a subjective marker and there does seem to be evidence to show that in fact satisfaction is linked to more discretionary services (such as imaging), has no association with quality of technical care (whether guidelines were followed) and higher mortality (although this was not linear and only seemed to affect the very highest scores).
So does this strike a fatal blow to subjective measures. Not so fast. As may seem common sense. It depends. Just like I wouldn’t expect asking someones favourite colour to be useful to measure a clinical outcome because it is subjective. It turns out that patient satisfaction is more powerfully driven by whether expectations are met. This makes intuitive sense. How can we be satisfied about our ‘medical’ treatment unless we know that treatment in the context of medicine? What we can be satisfied with are things we know well. How we are treated as human being. How well things were communicated. How seriously we were taken. These are things we are much better prepared for. So it makes sense when asked for satisfaction scores that people score vicariously from things they know than what they do not.
So lets throw patient satisfaction out the window? Woah there! If we understand what it is they are actually rating it remains useful. After all we are not in medicine to sustain life but to give quality of life (obviously the 2 overlap sometimes!). Patient satisfaction does seem to be correlated to compliance. Whilst Patient Satisfaction with Decision (PSwD) seems to predict compliance with future therapy/input. These all seem worth considering. Although compliance isn’t necessarily exactly what I’m chasing, more alliance.
The danger with dismissing patient satisfaction is that our world view (or ontology if you want to sound clever) defines successful outcomes from outside the person and without their permission. This objective interpretation comes from clinicians, organisations, statistics and government policies. Atul Gawande walks through a third way in his book ‘Being Mortal‘ (which I would highly recommend!). Here he rejects over-medicalising and objective health whilst also realising the flaws with pure subjective satisfaction. Instead he travails the chasm between them. Interpretation in its holist sense must not be an information drop of treatment options and statistics before leaving the patient decide. The role of the clinician is to journey between the two and provide an individual interpretation. The majority of patients will not be able to journey in this terrain. Truth is we find it difficult as clinicians. Despite our years of experience, service and learning there remains uncertainty in the navigation. We can be tempted to stick to the lands we know as these are easier to navigate and familiar. But this is no guarantee that it’s where they want to go.
Moving on from patient satisfaction there does seem to be good evidence that (particularly if you have a subjective question) a subjective measure is likely to be as good and maybe better than an objective one. For example subjective markers on acute and chronic physical stress changes are more responsive, powerful and sensitive than objective markers (Saw et al 2016), decreased perceived recovery was associated with injury risk (van der Does et al 2016), whilst perceived recovery-stress balance was associated with football injury (Laux et al 2015). There examples of self report being useful across a variety health settings. We cannot robotically ignore people and use our chosen measures by proxy. Neither should we drop our expertise, insight and knowledge which may help us navigate. Subjectivity can be and is reliable despite calls for cold hard facts (whatever these are). But we must not ignore that expectation may influence.
Be more human. Be less robot.
Thanks for reading this far.