Submission ID 129659
| Issue/Objective | Background: Maternal mortality in Ethiopia remains high, while most of these deaths are preventable. Early detection of deterioration and prompt response are essential to reduce these preventable deaths. The Maternal Early Warning System (MEWS) is a reliable clinical tool for this purpose. However, its effectiveness is underexplored and its bedside use is inconsistent. Objective: To evaluate the effectiveness of Maternal Early Warning System model for predicting and reducing severe maternal outcomes. Method: A quasi-experimental study design was conducted among 1138 obstetric inpatients from May to September 2025 in North Shewa Zone, Ethiopia. The intervention group was monitored using the MEWS chart, which included vital signs, oxygen saturation, urine output, consciousness, and pain, and for the postpartum women; vaginal bleeding, uterine contraction, and perineal tear, were categorized as Green, Yellow, or Red. The control group received the standard monitoring. A multivariate GEE model with Poisson regression was used. Result: The mean duration from admission to the first trigger was significantly reduced by 4.7 hours (5.61 vs. 10.27 hours), trigger to physician evaluation by 22.6 minutes (49.3 vs. 71.9 minutes), and trigger to clinical intervention by 11.3 minutes (14.6 vs. 25.9 minutes) in the intervention group. Women also had one fewer ultrasound scans (1.32 vs. 2.30) and a shorter hospital stay by about 0.5 days (4.83 vs. 5.29 days). Women monitored with the MEWS chart had a 20% lower risk of developing severe maternal outcomes (aRR = 0.80, 95% CI: 0.68-0.93). Additionally, MEWS-monitored women were 9% more likely to be triggered for timely clinical response (aRR = 1.09, 95% CI: 1.05-1.12). Conclusion and recommendation: The MEWS monitoring improved early detection of maternal deterioration, accelerated clinical response, reduced unnecessary investigations, shortened hospital stays, and lowered severe maternal outcomes. It is recommended for routine use in hospitalized women, with further research needed to assess its effectiveness across different risk groups and settings. Keywords: MEWS, Maternal monitoring tool, Adverse obstetric outcome, Ethiopia |
|---|---|
| Methodology/Approach | A parallel, quasi-experimental study design was implemented from May to September 2025 in North Shewa Zone, Amhara Region, Ethiopia. The total population of the zone is estimated at 2,429,108, of which 1,203,366 are females (16). There are eleven public hospitals in the zone: 3 General hospitals, 7 primary hospitals, and 1 comprehensive specialized hospital. The study was conducted in four purposively selected hospitals. Debre Berhan University Hakim Gizaw Hospital and Debre Berhan Comprehensive Specialized Hospital (CSH) were designated as intervention sites, while Enat Hospital and Mehalmeda Hospital served as control sites. These hospitals were chosen because they are comparable in terms of service provision, the presence of obstetricians, availability of intensive care units, and their status as government facilities. Intervention The intervention group was monitored using a statistically developed and validated Maternal Early Warning System (MEWS) chart (Supplementary file 1), which replaced the standard vital signs sheet for all enrolled participants at the intervention sites. The MEWS chart, previously validated and described in detail elsewhere (10,17), is a simple, observation-based tool designed to facilitate the early detection of maternal clinical deterioration. It categorizes parameters into three color-coded zones: Green (normal; no concern), Yellow (moderate abnormality; requires closer observation), and Red (severe derangement; demands immediate attention). The chart incorporates twelve key maternal clinical parameters (Table 1). Monitoring follows an escalation protocol. If all parameters remained within the normal range (green), routine monitoring continued as per standard practice. A single yellow alert prompted repeat observations within 30 minutes while maintaining standard monitoring. A patient was considered triggered for further assessment if they received one red score or two yellow scores, in which case repeat observations were carried out within 30 minutes, followed by confirmation of findings through history and examination. Monitoring frequency was increased, and corrective measures such as administration of intravenous fluids, oxygen at 10 L/min if required, antihypertensives, review of charts, or appropriate maternal positioning (e.g., left tilt for pregnant women) were initiated. If the patient's condition stabilized, routine monitoring was resumed; however, persistent or worsening red alerts required immediate review by a senior obstetrician within 60 minutes, which could lead to emergency intervention, urgent referral, or transfer to the intensive care unit (ICU). In cases where three or more yellow alerts or at least two red alerts were identified, immediate obstetrician review was mandatory, reassessment was conducted within 15 minutes, and continuous monitoring was started. If the situation remained unresolved, escalation of care to an anesthesiologist, critical care, and pain medicine specialist was required. This structured response system was designed to ensure timely recognition and rapid escalation of care, ultimately aiming to prevent progression to severe maternal morbidity or mortality. Table 1: Components of maternal early warning scores (MEWS) criteria for prediction of Severe Maternal Morbidity, Ethiopia, 2025. Physiologic parameters Normal values Yellow alert Red alert Systolic blood pressure 100 - 150 mmHg 90 - 100 OR 150 - 160 <90 OR ≥160 Diastolic blood pressure 50 - 89 mmHg 90 - 110 ≥110 Temperature 36 - 37.9 ◦C 35 - 36 ◦C ≤35 ◦C OR ≥38 Heart rate 50 - 99 BMP 40 - 50 OR 100 - 130 <40 OR >130 Respiratory rate 10 - 20 BMP 21 - 30 <10 OR >30 Oxygen saturation 95 - 100% 90 - 95% <90% Urine outputs ≥30 ml per hour - <35 ml/hour for 2 hours for catheterized women Consciousness level Alert Responds to voice To pain or unresponsive Pain score None to mild pain [0 - 3 pain score] Intermittent pain at rest, moderate pain with movement [4 - 7 pain score] Persistent pain at rest, severe pain with movement [8 - 10 pain score] For Postpartum or Post CS Women Vaginal bleeding No bleeding Mild to moderate, i.e., spotting/light bleeding Active, i.e., soaking a pad in <1 hour Uterine contraction Well-contracted uterus Soft but responsive to massage Boggy, poorly/ non-contracted uterus Perineal tear No tear 1st and 2nd degree tear 3rd and 4th degree tear Control group The control hospitals were continuing with their existing/standard clinical monitoring practices. This involved monitoring and recording temperature, pulse, blood pressure, and respiratory rate on vital sign sheets. |
| Results | Effect of MEWS intervention on clinical monitoring The independent samples t-test was performed to analyze the effect of the MEWS chart intervention on key clinical monitorings. The mean duration from admission to the first trigger was significantly shorter in the intervention group (5.61 ± 2.81 hours) compared to the control group (10.27 ± 3.49 hours, p < 0.001). Similarly, the mean time from trigger to physician evaluation was markedly lower among women in the MEWS group (49.30 ± 13.28 minutes) than those receiving standard care (71.87 ± 14.78 minutes p < 0.001). The time from trigger to clinical intervention was also significantly shorter in the intervention group (14.58 ± 7.86 minutes) compared to the control group (25.87 ± 9.23 minutes, p < 0.001). In contrast, the number of ultrasound scans was significantly higher in the control group (2.30 ± 1.92) than in the intervention group (1.32 ± 1.46, p < 0.001)). Additionally, the mean length of hospital stay was notably reduced among patients in the intervention group (4.83 ± 7.51 days) compared to those in the control group (5.29 ± 8.80 days, p < 0.001)). Effect of MEWS monitoring on Maternal outcomes Women monitored using the MEWS chart had a 20% lower risk of developing severe maternal outcomes compared to those who received standard care (aRR = 0.80, 95% CI: 0.68-0.93). Furthermore, women in the MEWS-monitored group were 9% more likely to be triggered for timely clinical response than those in the control group (aRR = 1.09, 95% CI: 1.05-1.12). |
| Discussion/Conclusion | The implementation of the MEWS chart in obstetric care has several important clinical implications. First, it facilitates early detection of maternal deterioration, as evidenced by the reduced time from admission to the first trigger and from trigger to physician evaluation. This allows healthcare providers to intervene promptly, preventing progression to severe maternal outcomes. Second, the MEWS system improves the timeliness and efficiency of clinical response, as shown by shorter intervals to clinical intervention and physician assessment. This ensures that abnormal physiological changes are addressed quickly, which can reduce morbidity and potentially prevent life-threatening complications. Third, early recognition and intervention appear to reduce the need for additional investigations, such as ultrasound scans, suggesting more targeted and efficient clinical management. Similarly, the shorter hospital stays in the intervention group indicate that prompt management may accelerate maternal recovery, improve patient flow, and optimize resource utilization in healthcare settings. Finally, the findings suggest that MEWS can enhance overall patient safety, clinical decision-making, and quality of care in obstetric settings. Incorporating MEWS into routine maternal monitoring may help standardize assessments, improve communication among healthcare teams, ensure timely escalation of care, and ultimately reduce the risk of severe maternal complications, particularly in resource-limited or high-volume hospitals. |
| Presenters and Affiliations | Mesfin Tadese Debre Berhan University Saba Desta Tessema Debre Berhan University Dr. Solomon Hailemeskel Debre Berhan University Chaltu Takele Debre Berhan University Dr. Getaneh Dejen Debre Berhan University Dr. Lidya Asalefew Mekonnen Debre Berhan University Prof. Getnet Mitikie Kassie International Institute for Primary Health Care - Ethiopia |
