Perfect imperfections: the difficult balance between perceiving individual stories and large statistics simultaneously

Dr Rakesh Modi is a GP and Primary Care Unit researcher on the landmark NIHR-funded SAFER trial investigating screening to detect an undiagnosed heart condition responsible for one in ten strokes. Here he writes on the difficult balance between perceiving individual stories and large statistics simultaneously, and how both the pandemic and the SAFER trial highlight this issue.

Americans are killed by lawnmowers at a roughly stable rate of 70 per year. Year after year.

These are things we hear, things we read, and things we accept. But how would any particular person end up dying by lawnmower? Did they trip because they put on the wrong shoes? Forgot to replace the duct taped-wire? What unfortunate decisions did they make that day of days? Perhaps the question is, what decisions must had been made for them?

Sidestepping a history of determinism as deftly as an uncle at a wedding party of six, we accept a level of inevitability when we see this fairly stable statistic – it was meant to be. We can accept being in a plane and seeing waves down below looking like a stationary screwed up piece of shiny blue paper, but when surfing, we can always hope the next wave is the right one. Likewise, we can grasp the statistics of the failures of health of large populations but lose our grip over the unfortunate circumstances that befall us or our loved ones.

We GPs – running never-ending patients up and down the gauntlet many times a day, and knowing that Statistics will demand its sacrifices regularly in the form of mistakes – still struggle with the individual stories of those failures. The person who has been and could have been. That elusive thought that never came to you (and really, never should have). This is about wearing an individualistic physician’s fascinator, and trying to find the head space for a massive public health bonnet. Not an easy balance.

Why am I weighing you down with this? I had a recent debate turned soapbox on a social media site for GPs, about the failure rate of Covid-19 tests. Many of my colleagues were advocating telling a patient to self-isolate because the patient had a reduction in their sense of smell, despite a negative Covid test. We do not want to be responsible for the spread of Covid. We will have failed. I feel that disquiet. But, I found myself articulating a different position: public health rules are intended for large populations. We are not aiming for immediate eradication, perfection, satisfaction: we are aiming for a taming – a flattening – of the curve. There will be cases that we miss but overall, we aim to reduce the R number to less than 1. Large scale systems that aim for perfection are not a viable utopia when it comes to population health.

Am I just a callous academic – hardened by the numbers I crunch underfoot? I argued that we GPs frequently employ this mind-set, that all rules have their indelible exceptions: the missed cancer diagnosis because the patient didn’t fit the criteria for referral, the appendicitis that didn’t hurt on the right-hand side. Furiously working to rub out those exceptions, expending energy and resources, would erase some good that was meant for other patients, whilst thinning out the fabric of the health system. Covid has just been greater, hurt us more, forced us to make greater compromises – and made our failures seem greater. We ask the population to accept the whims of fortune for the greater good, but if we as clinicians can’t internalise that concept, how can we expect this of the public at large?

Thinking about screening as part of my research in the SAFER trial, makes me realise that GPs and the public tread this path daily. We are investigating whether testing all people aged 70 and over for atrial fibrillation – an irregular heart beat that can cause strokes – is worthwhile. Just like breast screening, cervical screening, any screening, this test will illuminate many people with the diagnosis, but miss some others. When we explore what health professionals and participants think about it, we might see that this is an accepted clause in the contract of getting a screening test. Que sera.

But highlighting and explaining our implicit thoughts about screening, and the ethical frameworks within which screening programmes are developed and delivered, is important. We might help clinicians and the people we serve to understand how our stochastic behaviours – the apparently random things we do every day – add up to create the predictable normal curve: how good and bad fortune are invariably entangled.

So, sometimes when we ‘fail’ and we feel like the grass is greener on the other side, maybe we just need to accept it, get that lawnmower out, and take our chances.

Reading

For those that are interested and unencumbered, some reading that has stimulated me on this topic can also be summarised at different levels:

One-to-one:  ‘Where Does it Hurt’ by Max Pemberton.

Population: ‘Screening: Evidence and practice’ by Angela Raffle,  Anne Mackie and Muir Gray

Universal: ‘Micromegas’ and ‘Zadig’ by Voltaire

For those encumbered, consider the classic paper about trying to achieve perfection in tests by Neuhauser and Lewicki in  NEJM 1975: ‘What do we gain from the sixth stool guaiac?’

Find out more

Written by Dr Rakesh Modi, GP in St Neot’s and Wellcome Trust PhD Fellow, Primary Care Unit, University of Cambridge.

@rak_modi