ALERT’s mortality signal emerged after adjustment, not in raw rates
Across 16 hospitals and 139,300 neonates, ALERT was linked to lower adjusted odds of fresh stillbirth or death within 24 hours (aOR 0.78, 95% CI 0.65–0.94). Unadjusted comparison and intervention period totals were 3.1% in both, and a full-uptake sensitivity analysis attenuated the estimate to 0.96.
THE HOT TAKE
We observed substantial heterogeneity between hospitals with I² values exceeding 50% for most outcomes.
ALERT observed 134,630 women during five six-month periods at 16 hospitals in Benin, Malawi, Tanzania, and Uganda, yielding data for 139,300 neonates over 1,000 g. Every hospital crossed from comparison care to a locally adapted package of co-design, competency training, quality improvement, and leadership mentoring. The composite co-primary outcome, fresh stillbirth or neonatal death within 24 hours, favored intervention periods after adjustment: odds ratio 0.78 (95% CI 0.65–0.94; P=0.008). Fresh stillbirth alone did not: 0.88 (0.68–1.13; P=0.32).
- 1
16 hospitals randomized
Four high-caseload hospitals per country were randomized within country to a crossover order.
- 2
Five six-month periods
Hospitals began in comparison care and crossed to ALERT at their assigned time.
- 3
134,630 women; 139,300 neonates
The intention-to-treat cohort included liveborn and stillborn infants over 1,000 g born in participating hospitals.
- 4
Hospital effects pooled
Models adjusted for six-month period, estimated each hospital's effect, then pooled those estimates with random effects.
Was this worth carrying into your next shift?
Your reaction stays connected to your NICU360 profile.
Comments
0Join the conversation
Sign in to reply, vote, and keep your participation with your NICU360 profile.