Current Issue

Referral Appropriateness and Structural Barriers in Basic Emergency Obstetric and Newborn Care Implementation: Evidence from Zamboanga City Medical Center, Zamboanga Peninsula, Philippines

Abstract

Background. The Basic Emergency Obstetric and Newborn Care (BEMONC) program equips lower-level facilities to manage uncomplicated deliveries and stabilize obstetric emergencies, reducing unnecessary transfers to tertiary hospitals. Despite sustained training coverage in the Philippines, tertiary facilities continue to absorb high obstetric referral volumes. This study determined the prevalence, clinical profile, institutional sources, and structural determinants of non-indicated and potentially avoidable obstetric referrals received by a tertiary government hospital in Southwestern Mindanao from February 2024 to April 2025. Methods. A retrospective cross-sectional chart review examined 300 consecutive obstetric referral records from the Department of Obstetrics and Gynecology Referral Census. Non-indicated referrals were defined as uncomplicated normal spontaneous vaginal deliveries without a documented tertiary-level indication. A broader potentially avoidable category captured uncomplicated vaginal deliveries transferred for weak, non-medical, or capacity-related reasons. Referring facilities were cross-referenced against the institutional BEMONC training registry. Descriptive statistics, exploratory odds ratios, and sensitivity analyses were performed. Results. Of 300 referrals, 27 (9.0%) were strictly non-indicated and 35 (11.7%) were potentially avoidable. Cesarean section accounted for 40.7% of outcomes. Generic escalation (42.7%), absence of an obstetrician (18.0%), lack of incubator/neonatal support (17.0%), and absence of an anesthesiologist (15.7%) accounted for 93.4% of referral reasons. Primary care facilities had higher odds of avoidable referral than hospital-level facilities (OR 7.26, 95% CI 2.91–18.10). Training volume showed no significant correlation with referral appropriateness (Spearman's rho = 0.02, p = .94), and facility training status was not significantly associated with avoidable referral after adjusting for facility level (OR 2.47, 95% CI 0.51–11.89). Conclusion and Implications. Strictly non-indicated referrals were uncommon. Avoidable transfers reflected structural resource limitations rather than poor clinical judgment. These findings suggest BEMONC training alone is unlikely to reduce avoidable obstetric referrals unless matched by personnel, equipment, and referral governance. Program evaluation should incorporate facility-level referral performance alongside training coverage. Keywords: BEMONC, maternal health, obstetric referral, referral appropriateness, structural barriers