Computer model helps diagnose fever in children

A computer model can outperform clinicians at diagnosing fever in children, a study suggests.

Australian researchers created a computerised diagnostic model to distinguish between serious bacterial infections and minor febrile illness in children.

They commented that although febrile illnesses are common in children, it is often difficult to tell apart bacterial infections, such as pneumonia, urinary tract infection and meningitis, from self-limiting fever caused by viruses.

The researchers followed 15,781 healthy children over five years old and tracked admissions to an emergency department with febrile illness - a body temperature of 38°C or more - over a two-year period.

Clinicians evaluated the patients, and the diagnosis was then checked by standard tests and follow up. Any signs and symptoms noted were added to the computerised model. The clinicians' diagnosis was then compared against the computer model's prediction.

The results showed bacterial infections were present in 7% of child fever cases.

Only 70-80% of children with bacterial infections were prescribed antibiotics on initial consultation with a clinician. 20% of those with an identified bacterial infection were probably over-treated, the researchers said.

The performance of the diagnostic model for each infection was acceptable or better than clinical evaluation.

The authors concluded: ‘By combining routinely collected clinical information into a statistical model, we have demonstrated that a clinical diagnostic model may improve the care of children presenting with fever who have suspected serious bacterial illness.'

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins

Register

Already registered?

Sign in

Before commenting please read our rules for commenting on articles.

If you see a comment you find offensive, you can flag it as inappropriate. In the top right-hand corner of an individual comment, you will see 'flag as inappropriate'. Clicking this prompts us to review the comment. For further information see our rules for commenting on articles.

comments powered by Disqus