<![CDATA[Introduction: Maternal deaths in low- and middle-income settings are driven largely by delays in recognising and treating obstetric emergencies at the primary-care frontline. Real-time specialist tele-mentoring may compress these delays, yet trial evidence on hard maternal and neonatal endpoints remains scarce in Indonesia. We evaluated whether a specialist tele-mentoring triage system reduces severe maternal morbidity and mortality in high-risk pregnancies. Methods: In a closed-cohort stepped-wedge cluster randomized trial, ten primary-to-referral facility clusters across two Indonesian metropolitan regions sequentially crossed from conventional referral to a 24/7 tele-mentoring triage system (audiovisual specialist consultation, portable cardiotocography/vital-sign telemetry, and a Maternal Early Warning Score [MEWS] algorithm) over six two-month periods. The primary outcome was a composite of maternal death and WHO-defined severe maternal morbidity. Intention-to-treat analysis used generalized linear mixed models adjusting for secular trend and cluster random effect. Results: Among 1,858 high-risk women, the composite outcome fell from 7.97% (95% CI 6.39–9.88) during control person-time to 4.41% (95% CI 3.27–5.93) during tele-mentoring (adjusted OR 0.51, 95% CI 0.36–0.73, p<0.001; absolute risk reduction 3.55%; number-needed-to-treat 28). Door-to-treatment time shortened by 85.6 minutes (Cohen's d 1.77, p<0.001). MEWS discriminated the composite outcome well (AUC 0.84, 95% CI 0.80–0.88). NICU admission (OR 0.66, NNT 19, p=0.003) and 5-minute Apgar <7 (OR 0.58, p=0.002) also improved. Conclusion: A specialist tele-mentoring triage system roughly halved the odds of severe maternal morbidity and mortality in high-risk pregnancies and improved neonatal outcomes. Scaling specialist-guided digital triage could meaningfully strengthen obstetric emergency referral in Indonesia.]]>
