Midwife-performed checklist and ultrasound to identify obstetric conditions at labour triage in Uganda: A quasi-experimental study.

Not randomised: women were compared by time period, so changes in staff, patients, or care over time could explain the results. Midwives who used the checklist or ultrasound also helped document outcomes after birth, which could make suspected complications more likely to be recorded. Some outcome diagnoses, such as preterm birth and oligohydramnios, may be inaccurate, and prior antenatal diagnoses were not systematically captured. Women needing immediate intervention were excluded, so the findings do not apply to the most urgent labour presentations. Enrolment and missing outcome data differed across phases, and Phase 3 had lower enrolment. This could introduce selection bias. Participants differed across phases in education, fuel source, obstetric history and antenatal visits. Adjusted analyses did not remove the sensitivity and positive predictive value patterns. Only some midwives were trained, and workload, staff rotation, machine defects and power or battery problems may affect real-world use. The study was done in a single district hospital in Uganda. It was powered for the composite outcome, not individual conditions or birth outcomes.

The study was supported by the East Africa Preterm Birth Initiative, funded by the Bill & Melinda Gates Foundation. The authors report no financial conflicts of interest.

Do not conclude that midwife-performed ultrasound or the checklist improves maternal or newborn outcomes. The study measured detection at labour triage, not birth outcomes, and the ultrasound was added to standardised documentation and a checklist in a non-randomised phased study, so the effect of any single component cannot be separated.