Where a child is treated for sepsis is linked to survival

Design

Retrospective cohort using national Medicaid claims; regional standardised mortality

Population

36,552 paediatric sepsis episodes, US publicly insured children, 2021-2023

Primary outcome

Death within 30 days of sepsis onset

Effect

Community-acquired mortality 3.9%; children's-hospital regions aOR 0.76 (0.61-0.94); rehospitalisation 16.7%

A cohort study of all US publicly insured children used Medicaid data to examine 36,552 sepsis episodes in 2021-2023, with death within 30 days as the outcome and regional standardised mortality ratios to compare areas.

Community-acquired sepsis carried a 3.9% 30-day mortality, with age, complex chronic conditions and organ dysfunction predicting death. Regional standardised mortality varied widely (ratios from 0 to 3.82), with more high- and low-mortality outlier regions than chance would predict. Regions served by a children's hospital had lower mortality (adjusted odds ratio 0.76), as did regions with higher sepsis incidence. Nearly one in six episodes led to rehospitalisation within 30 days.

The signal is about systems, not individual care: outcomes depend partly on where a child is treated, supporting regionalised paediatric acute care and strong referral links to specialist centres. It is US administrative data, so the associations are ecological and may not transfer directly elsewhere.

• Community-acquired paediatric sepsis had a 3.9% 30-day mortality.
• Regional standardised mortality varied widely, from 0 to 3.82.
• Regions with a children's hospital had lower mortality (odds ratio 0.76).
• Nearly one in six sepsis episodes led to rehospitalisation within 30 days.
• Strong referral links to specialist centres may matter for survival.

Why it matters

It reframes some sepsis deaths as a question of access and system design, not only bedside care.

Don't overread it

This is US administrative data with region-level associations; it cannot attribute outcomes to any single aspect of care or transfer precisely to other health systems.

The statistics, in plain English

A standardised mortality ratio above 1 means more deaths than expected for that region's case mix; the wide spread and excess outliers suggest real system-level differences, not just chance.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.