Views And Reviews No Holds Barred

Margaret McCartney: Asthma and the catwalk

BMJ 2016; 353 doi: (Published 31 May 2016) Cite this as: BMJ 2016;353:i2975
  1. Margaret McCartney,
  2. general practitioner
  1. Glasgow
  1. margaret{at}

Are GPs guilty of issuing asthma inhalers to children as “fashion accessories,” as charged recently in the Archives of Disease in Childhood—subjecting young people to the adverse effects of corticosteroids with the same whim we use to furnish our wardrobes with this season’s catwalk collections?1

Or are we guilty of being party to the deaths of young people with asthma in the United Kingdom, of whom 93% were being inadequately managed, as suggested in a Royal College of Physicians audit that found 1200 such deaths?2

One conclusion is that GPs are forever condemned to shack up with the Devil on a punctured life raft. Another is that perhaps this balance of underdiagnosis, overdiagnosis, and treatment typifies general practice: we live in uncertainty, dealing in probabilities, where the most accurate diagnoses are made in the fullness of time and with a “retrospectroscope.”

Clearly, some patients are likely to gain much benefit from having an asthma diagnosis with vigorous treatment and follow-up. For other patients, probabilities, based on history and examination, hint strongly at a long term, recurrent, wheezy condition that will respond better to treatment than to none. But this is often likely to be a condition in flux, and the National Institute for Health and Care Excellence’s proposal to use exhaled nitric oxide testing to make more accurate diagnoses sounds useful, until you read of its wide range in sensitivity.3

We’re using a map for the wrong road. Diagnoses can be useful but also encourage confirmation bias and can expire over time, and they’re emphatically binary when it may be more accurate and honest to talk of “probable” or “likely” asthma.

The coding systems that GPs use were not designed for nuance but for data gathering, collection, research, and audit on a grand scale. Nomenclature that tries to contain uncertainty may be confusing and troubling, as we’ve seen before with “mild cognitive impairment” (later life forgetfulness, benign senescent forgetfulness, but also a harbinger of dementia?)4—it pushes the problem around and doesn’t deal with the central dilemma of uncertainty.

Rather than seeking a definitive diagnosis, it’s safer to be honest. This may be a chronic condition, but it often won’t be: we can test a treatment, but patients should return for reappraisal. This takes time.

In a healthcare system obsessed with equating quality with speed and the belief that more is always better, challenge is needed from within. Certainly, let’s make diagnostic accuracy better, when and if we can. But let’s also accept that, when the stakes of under-treatment are high, it’s not fashion that most doctors are concerned about, but childhood death.



View Abstract