Hot on the heels of thinking about thresholds for action and inaction comes a really interesting paper looking at the risk of serious infection in children with blood or CSF cultures taken in the South East of England (in 1m – 15y olds).
Before going on – what proportion of cultures do you think were positive?
The study took a prospective approach to the collection of positive samples from all the included microbiology labs, and then came up with what % of cultures, and estimates of what % of admissions, gave a positive result.
Now – if half of children are admitted with ‘suspected infection’, if half of those have significant comorbidities (like cancer), and only 2% of all admissions have a microbiologically defined severe infection – are we using ‘suspected infection’ as a catchall to make our admissions sound sensible? Or are we over-admitting, and not using low-risk rules of observation as outpatients? Or should we read this as showing that we’re pretty good – just by taking blood cultures you quadruple the proportion with MDI (to ~8%).
How do you interpret these results?