Intended for healthcare professionals

Views And Reviews No Holds Barred

Margaret McCartney: Alarm overload makes a difficult job harder

BMJ 2017; 358 doi: (Published 31 July 2017) Cite this as: BMJ 2017;358:j3593
  1. Margaret McCartney, general practitioner
  1. Glasgow
  1. margaret{at}

“Just ask . . . could it be sepsis?” The Sepsis Trust has issued posters instructing parents to ask this, as “it’s a simple question but could save a life.” It’s also produced a list of what to look out for. Any child who is breathing quickly, has a fit or is unconscious, looks mottled, blue, or pale, has a non-blanching rash, is very lethargic or difficult to wake, or is abnormally cold to touch “might have sepsis”; and anyone whose child under 5 is not feeding, is vomiting repeatedly, or has not passed urine in 12 hours is urged to seek advice.1

General practice is not hospital, and it has a different population. Detecting already very sick children in general practice is usually easy, as they’re obviously ill. It’s also very rare. Children with infections are often unwell, but not seriously: they are managing physiologically. Detecting the unwell child who’s going to become very ill without secondary care intervention is very hard.

All children with certain common conditions—such an upper respiratory infection or an open wound—are at risk of sepsis, but they’re unlikely to contract it. In the early stages of illness it’s impossible to be sure that this child in front of me won’t develop sepsis in the coming hours or days.

Children can be terrifying, because they can get ill quickly. They can also get well quickly: parents of children who perk up in the consulting room often apologise for bringing in their now smiling child as an emergency (I’m invariably pleased, although I usually wonder whether I’m being falsely reassured).

Before anyone accuses GPs of being lax, consider the number of alerts already present that distract from, rather than assist in, management

The commonness of minor illness and the potential for rapid change mean that tools used for risk assessment in secondary care can’t easily be used safely in primary care.

In recent years a new tool was added to many GPs’ electronic note systems. GPs keyed in the temperature recording, and it triggered a cascade of further interventions to investigate and manage sepsis.2 I know many GPs who switched it off or stopped recording temperatures in the way that made it light up.

Before anyone accuses GPs of being lax, consider the number of alerts already present that distract from, rather than assist in, management. If every doctor read every alert generated during every consultation and drug prescription the NHS would be immobilised, to the detriment of patients waiting for assessment and worsening while doing so.

There is another problem. Sepsis is not the only life threatening emergency. Tools to encourage the consideration and exclusion of just one diagnosis, when many are possible and equally important, don’t work the way primary care should. Is it right to place an onus on parents—and, if so, shouldn’t they also inquire whether their unwell child has asthma, meningitis, a brain tumour, or significant arrhythmia?

Given the unconscionable strain on the NHS, I fear that false positive risk assessments will mean delays for people who really do need inpatient care.



View Abstract