Sage Journals: Discover world-class research

Abstract

Background:

Ethiopia remains among the 5 countries with the highest under-five mortality, yet prospective multicenter data on pediatric intensive care unit (PICU) outcomes are limited.

Objective:

To determine the incidence and predictors of mortality among children admitted to PICUs in Eastern Amhara.

Methods:

A prospective observational study was conducted on 240 children admitted to tertiary-level PICUs between February and April 2025. Kaplan–Meier analysis estimated survival probabilities, and Cox proportional hazards regression identified independent predictors of mortality

Results:

Fifty-eight deaths (24.2%) occurred. Rural residence (AHR: 2.0, 95% CI: 1.2-3.5), the use of inotropic agents (AHR: 2.0, 95% CI: 1.2-3.5) consciousness level (GCS <8; AHR: 2.3, 95% CI: 1.2-4.4), presence of comorbidities (AHR: 1.9, 95% CI: 1.1-3.4) were associated increased mortality.

Conclusion:

PICU mortality was high, especially among rural children, those with severe neurological impairment, comorbidities, or requiring inotropes, emphasizing early identification and close monitoring of high-risk patients.

Keywords

mortality rate predictor pediatric intensive care unit eastern Amhara Ethiopia

Introduction

The Pediatric Intensive Care Unit (PICU) delivers specialized, high acuity care to critically children with actual or impending life threating organ dysfunction. Care provided by a highly skilled multidisciplinary team aiming to stabilize vital function, prevent irreversible organ failure, and improved survival outcomes. Children admitted to the PICU commonly present with severe medical emergencies, traumatic injuries, postoperative complications, or critical infections and may be admitted directly or transferred from emergency department or inpatients ward for advanced monitoring and organ suppor.^1
-3 Despite these lifesaving roles, PICUs are inherently high risk environments, where diagnostic errors, hospital acquired infections, delayed intervention, and complex medication regimens may contribute to adverse outcomes ad increased mortality.⁴

Globally, preventable and treatable conditions remain the leading causes of mortality among children under 5 years of age, particularly in low resource setting.⁵ In high-income countries, the establishment of well-resourced PICUs staffed with trained intensivists, and nurses has resulted in substantial improvement in pediatric survival, with reported mortality reductions ranging from 15% to 60%.⁶ For instance, reported PICU mortality rates are as low as 3.6% in Japan,⁷ and 2.6% in Australia and New Zealand.⁸ In contrast, pediatric emergency and critical care service in many low-and middle countries (LMICs) remain underdeveloped and fragmented.⁹ This disparity is especially pronounced in Sub-Saharan Africa (SSA), where PICU care is constrained by limited infrastructure, and shortage of trained personnel. Inadequate nurse to patient’s ratios, restricted access to essential equipment, and delayed referral systems.^10
-12 As a result, the reported PICU mortality rates were significantly higher in resource-limited settings, ranging from 10.3% to 41.2%.^13
-18

In 2018 alone, an estimated 5.3 million children died worldwide, with nearly 80% of these deaths occurring in SSA and Asia.¹⁹ By 2020, SSA continues to shoulder a disproportionate burden, accounting for approximately 53% global under-five deaths,^20,21 largely driven by severe infections, including pneumonia, diarrhea, and malaria, condition frequently necessitate intensive care when complication arise.²² Ethiopian remains among the 5 countries with the highest under-five mortality rates globally, estimated at 59 deaths per 1000 live births.^5,23 In Ethiopian PICUs and similar regional settings, septic shock, acute medical emergencies, and trauma represent the most common indications for admission.^17,24,25

Previous studies have identified multiple factors associate with PICU mortality, including the younger age, patient’s clinical condition at admission, presence of comorbid conditions, altered level of consciousness, and the need for advanced life support interventions such as mechanical ventilation (MV).^17,18,24
-28 However, the magnitude and direction of these associations vary across studies, reflecting differences in patient populations, health-care infrastructure, study design, and clinical practices.

Despite the high burden of critical illness and mortality, there is a paucity of high quality of prospective and multicenter data on PICU outcomes and their predictors in Ethiopia. Prior studies have largely relied on single center, retrospective chart reviews, which are limited by incomplete documentation, selection bias and restricted generalizability.^24,25,29,30 The lack of robust multicenter evidence hinders the development of evidence-based clinical protocols, efficient resource allocation, and targeted interventions to improve pediatric critical care outcomes. To address these gaps, we conducted a prospective multicenter observational study in tertiary hospital PICUs of Eastern Amhara, enabling standardized data collection across multiple sites, minimizing bias, and enhancing the applicability of findings. By generating context-specific and methodologically rigorous evidence, these studies aims to inform clinical decision-making, support health-system strengthening, and contribute to reducing preventable pediatric mortality in resource-limited settings.

Methods and Material

Study Setting, Design, and Period

This prospective observational study was conducted from February 2025 to April 2025 in tertiary-level hospitals in Eastern Amhara, Ethiopia. The study was carried out in 3 hospitals with established PICU services: Debre Tabor Comprehensive Specialized Hospital (DCSH) and Felege Hiwot Comprehensive Specialized Hospital (FHCSH) and Tibebe Ghion Comprehensive Specialized Hospital (TGCSH). DCSH serves as the main referral hospital for south Gondar and surrounding districts. The hospitals admit over 1160 pediatric patients annually and has pediatric ward with 45 beds, including 6 designated PICU beds. FHCSH is one of the largest referral hospitals in the region, with 118 pediatric beds and an 11-bed PICU. TGCSH is a tertiary teaching hospital with 459 inpatient beds, including 52 pediatric beds and 10 PICU beds.

These hospitals function as the highest level of pediatric care within the regional health system and receive critically ill children through a structured referral pathway from primary hospitals, district hospitals, and emergency departments. Children are admitted to the PICU either directly from emergency units or transferred from pediatric and surgical wards following clinical deterioration requiring advanced monitoring and organ support.

Populations and Eligibility Criteria

The source population for this study consisted of all pediatric patients admitted to tertiary-level hospitals in Eastern Amhara, Ethiopia. The study population included patients aged 14 days to 18 years who were admitted to the PICU during the study period and met the inclusion criteria. Patients with missing or incomplete baseline critical variables within the specified time window, primarily due to clinical urgency or operational constraints, were excluded from the study after attempts to retrieve the data from source document. Additionally, those who died within 2 hours of admission were also excluded, as this timeframe does not allow for optimal PICU care, and their outcomes are more closely associated with the care provided in the emergency department or before admission.

Sample Size Calculation and Sampling Technique

Sample size was determined using an events-per-variable (EPV) approach for Cox regression. With up to 10 predictors and a minimum of 10 event per variable, at least 100 outcome events were required.³¹ Prior Ethiopian and SSA PICU studies report mortality rates of 30% to 50%.^24,32
-34 Assuming a conservative mortality rate of 40%, enrollment 240 patients was expected to yield approximately 96 deaths (EPV ≈ 9.6), which was considered adequate for planned multivariable analysis.

Based on historical admission data, the sample was proportionally allocated across the 3 hospitals, resulting in 73 patients for DCSH, 78 for TGCSH, and 89 for FHCSH. A consecutive sampling approach was employed, whereby all eligible patients admitted to the PICU during the study period were approached sequentially until the hospital-specific sample size was achieved. A flow diagram illustrates pediatric patient recruitment, hospital-level inclusion, reasons for exclusion, and the age distribution of the final study cohort in the PICUs of the 3 hospitals. Two neonates (14-28 days) were admitted to the PICU due to full NICU capacity but were excluded because both died within 2 hours of admission. As a result, all patients included in the final analysis were older than 1 month (Supplemental Figure S1). Each patient was assigned a unique identification and questionnaire code to prevent duplicate inclusion of patient charts in the study.

Study Variables

The primary outcome of this study is time to death. Explanatory variables were socio-demographic characteristics (age, weight, caregiver occupation, etc.), presence of comorbidity conditions, hemoglobin level, admission diagnosis (septic shock, meningitis, and respiratory failure), nutritional status (Z-score), Glasgow Coma Scale (GCS), patient feeding method, interventions; use of MV, urinary catheter, central catheter, use of antibiotics, inotropes, in-PICU complication, and length of MV stay.

Operational Definitions

Co-morbidity: can be defined using the International Classification of Diseases (ICD-11) classification system as the presence of 1 or more additional diseases or conditions that coexist with a primary condition in a patient.³⁵ The Glasgow Coma Scale (GCS): is a clinical assessment tool used to quantify a patient’s level of consciousness and neurological function. Patients who scored <8 were considered unconscious.³⁶ Length of PICU stays: This was the period in hours, days, or months that the patients stayed in PICU from admission to discharge. The presence of organ dysfunction: measured in multi-organ dysfunction syndrome (MODS), refers to a potentially reversible physiologic derangement in 2 or more organ system. In-PICU complications: was defined as any new adverse clinical event occurring ≥24 hours after PICU admission that was not present at baseline and required additional intervention, monitoring, or prolonged PICU stay. Vaccination status: classified based on documentation of completion of the national Expanded Programme on Immunization (EPI) schedule at PICU admission. Children with records confirming full completion were defined as fully vaccinated; those without were defined as not fully vaccinated.

Data Collection Procedures and Quality Assurance

Data were collected prospectively using a structured case record form developed based on comprehensive literature review and prepared in English.^{17,18,24,25,28} Data were pretested on 5% of the sample at a tertiary hospital in the region, and necessary modifications were made. After obtaining written informed consent from caregivers, data were collected by trained medical interns under the supervision of treating physicians.

Baseline data were collected within 2 hours of PICU admission, including comorbidities, vital signs, oxygen saturation, neurological status evaluated using the pediatric Glasgow Coma Scale (GCS), nutritional status determined using anthropometric Z-scores, feeding method, hemoglobin level, admission diagnoses, and the need for invasive procedures such as MV, urinary catheterization, or central venous access. Early stabilization variables, documented within 12 hour, comprised vasopressor use, ventilator support, urine output, fluid balance, and evolving organ dysfunction using predefined criteria. Follow-up data were recorded daily throughout the PICU stay and included daily therapeutic exposures (antibiotics, and other drugs), duration of MV, in-PICU complications such as hospital-acquired infections (HAI; defined per World Health Organization and Centers for Disease Control and Prevention criteria as infections occurring ≥48 hours after admission), and clinical outcomes. The primary outcome, time to death, was measured in days from admission to death or discharge, with censoring at transfer, survival and other reasons.

Standardize orientation and training sessions were conducted to ensure uniform understanding of the study objectives, variable definition, and data collection procedures. Daily supervision and verification of collected data were conducted by the principal investigator. Data were double-entered and validated to ensure accuracy. Multi-center variability was assessed through Kaplan-Meier survival curves, log-rank tests, and hospital-level frailty analysis; however, variability between hospitals was statistically insignificant, and patient-level predictors were analyzed using standard multivariable Cox regression (Supplemental File S1). Investigators accessed medical records solely for research purposes and were not involved in patient care.

Data Processing and Analysis Procedure

Data were coded and cleaned using EpiData v4.6.6 and analyzed in STATA v17. Categorical variables were summarized as frequencies and percentages, and continuous variables as median (IQR). Survival distributions between groups were compared using Kaplan-Meier curves and the log-rank test. Predictors of time to death were analyzed with Cox proportional hazards regression. Variables with P < .2 in bi-variable analysis were entered into multi-variable models, and P < .05 was considered statistically significant. Model assumptions were checked using Schoenfeld and Martingale residuals, and multicollinearity was assessed using variance inflation factors (VIF <1.3).

Ethical Approval and Informed Consent

Ethical approval was obtained from the Institutional Review Board of (an institutional review board in Ethiopia [approval details withheld for peer review]), and permissions were obtained from (the relevant hospital authorities). Informed consent was secured from the legal guardians of all participants after explaining the study objectives and procedures, including the right to withdraw at any time. The study adhered to the Declaration of Helsinki, and all data were anonymized and securely stored to ensure participant confidentiality.

Result

Socio-Demographic Characteristics of Study Participants

In this study, males comprised 50.8% of the study population (n = 122). Mortality was 13.3% among males and 10.8% among females; however, survival distribution did not differ significantly by sex (log-rank P = .4). Similarly, there was no significant difference in survival distribution across age groups (log-rank P = .3). Urban residents accounted for 52.4% of participants (n = 127), and residential location was found to have significant difference in survival distribution (log-rank P = .01). The emergency room was the largest source of admission, represented 45.8% (n = 110), followed by the medical ward (n = 83, 34.6%). Although patients admitted from the emergency room had a higher cumulative incidence of death (n = 25, 10.4%), survival distributions did not differ significantly by source of admission (log-rank P = .7; Table 1).

Table 1.

Sociodemographic-Related Characteristics With Cumulative Incidence of Death and Survival Probability of Study Participants.

Variables		Censored	Dead	Total	Log-rank
Gender	Male	90 (37.5%)	32 (13.3%)	122 (50.8%)	0.4
Gender	Female	92 (38.3%)	26 (10.8%)	118 (49.2%)	0.4
Age in months	1-12	66 (27.5%)	20 (8.3%)	86 (35.8%)	0.3
	13-60	71 (29.6%)	18 (7.5%)	89 (37.1%)
	61-120	22 (9.2%)	11 (4.6%)	33 (13.8%)
	>120	23 (9.5%)	9 (3.8%)	32 (13.3%)
	Median (IQR) in months			18 (5.6-60)	NA
Weight (median ± IQR) in kg				10.8 (7.4-17)	NA
Residence	Urban	107 (44.6%)	20 (8.3%)	127 (52.9%)	0.01^*
Residence	Rural	75 (31.3%)	38 (15.8%)	113 (47.1%)	0.01^*
Nutritional status	Normal	140 (58.3%)	34 (14.2%)	174 (72.5%)	0.03^*
Nutritional status	Malnourished (Z-score <−3)	42 (17.5%)	24 (10%)	66 (27.5%)	0.03^*
Vaccination	Fully vaccinated	121 (50.5%)	38 (15.8%)	159 (66.3%)	0.65
Vaccination	Not fully vaccinated	61 (25.4%)	20 (8.3%)	81 (33.7%)	0.65
Source of admission	Emergency	85 (35.4%)	25 (10.4%)	110 (45.8%)	0.7
	Medical	63 (26.3%)	20 (8.3%)	83 (34.6%)
	Operation theater	24 (10%)	8 (3.3%)	32 (13.3%)
	Recovery	10 (4.2%)	5 (2.1%)	15 (6.3%)
Frequency of admission	First admission	126 (52.5%)	40 (16.7%)	166 (69.2%)	0.5
Frequency of admission	Readmission	56 (23.3%)	18 (7.5%)	74 (30.8%)	0.5
Caregiver education	Not able to write and read	56 (23.3%)	25 (10.4%)	81 (33.7%)	0.4
	Primary	48 (20%)	11 (4.6%)	59 (24.6%)
	Secondary	44 (18.3%)	13 (5.4%)	57 (23.7%)
	Tertiary	34 (14.2%)	9 (3.8%)	43 (18%)
Caregiver occupation	Farmer	35 (14.6%)	13 (5.4%)	48 (20%)	0.8
	Housewife	73 (30.4%)	24 (10%)	97 (40.4%)
	Employee	34 (14.2%)	10 (4.2%)	44 (18.4%)
	Private work	28 (11.7%)	7 (3%)	35 (14.7%)
	Other	12 (5%)	4 (1.7%)	16 (6.7%)

Note. Other: student, pension, and unemployed.

Indicates statistically significant results (p < 0.05).

Clinical-Related Characteristics and Diagnosis of PICU

The primary reasons for PICU admission were respiratory failure (n = 43, 17.9%), meningitis (n = 42, 17.5%), and septic shock (n = 41, 17.1%). The cumulative incidence of death was highest among children admitted with respiratory failure (n = 20, 8.3%). However, survival distribution a did not differ significantly by admission diagnosis of septic shock or meningitis (log rank P > .05). In contrast, survival distribution differed significantly according to neurological status at admission, with children presenting with GCS score of <8 (log-rank P < .001; Table 2).

Table 2.

Clinical-Related and Diagnostic Characteristics of the Cumulative Incidence of Death and Survival Probability of Patients.

Variables		Censored	Dead	Total	Log rank
Reason for admission	Asthma	6 (2.5%)	1 (0.4%)	7 (2.9%)	0.7
	Severe pneumonia	18 (7.5%)	3 (1.3%)	21 (8.8%)	0.4
	Respiratory failure	24 (10%)	19 (7.9%)	43 (17.9%)	0.01^*
	Cardiovascular cases^a	11 (4.6%)	6 (2.5%)	17 (7.1%)	0.6
	Meningitis	29 (12.1%)	13 (5.4%)	42 (17.5%)	0.13
	Traumatic brain injury	10 (4.1%)	4 (1.7%)	14 (5.8%)	0.7
	Gullian-Barré syndrome	5 (2.1%)	0	5 (2.1%)	0.2
	Septic shock	32 (13.3%)	9 (3.8%)	41 (17.1%)	0.5
	Surgery	21 (8.8%)	2 (0.8%)	23 (9.6%)	0.3
	Malaria	8 (3.3%)	0	8 (3.3%)	NA
	Burn	3 (1.3%)	0	3 (1.3%)	NA
	Acute glomerulonephritis	3 (1.3%)	0	3 (1.3%)	NA
	Miscellaneous condition^b	11 (4.5%)	2 (0.8%)	13 (5.4%)	0.8
Level of consciousness (GCS)	<8	28 (11.7%)	26 (10.8%)	54 (22.5%)	<0.001^*
Level of consciousness (GCS)	≥8	154 (64.2%)	32 (13.3%)	186 (77.5%)	<0.001^*
Oxygen saturation	<90%	94 (39.2%)	33 (13.8%)	127 (53%)	0.24
Oxygen saturation	90%-100%	88 (36.7%)	25 (10.3%)	113 (47%)	0.24
Comorbidity conditions	No	110 (45.8%)	18 (7.5%)	128 (53.3%)	0.02^*
Comorbidity conditions	Yes	72 (30%)	40 (16.7%)	112 (46.7%)	0.02^*
In PICU complication	No	157 (65.4%)	36 (15%)	193 (80.4%)	0.09
In PICU complication	Yes	25 (10.4%)	22 (9.2%)	47 (19.6%)	0.09
Reason for admission	Infectious	102 (42.5%)	31 (13%)	133 (55.5%)	0.3
Reason for admission	Non infectious	80 (33.2%)	27 (11.3%)	107 (44.5%)	0.3
Organ dysfunction (MODS)	Yes	123 (51.3%)	42 (17.5%)	165 (68.8%)	0.2
Organ dysfunction (MODS)	No	59 (24.5%)	16 (6.7%)	75 (31.2%)	0.2
Patient feeding	Oral feeding	43 (18%)	12 (5%)	55 (23%)	0.6
	Nasogastric tube	74 (30.8%)	22 (9.2%)	96 (40%)
	Maintenance fluid	65 (27.1%)	24 (10%)	89 (37.1%)
Hemoglobin (g/dl)	<10	122 (50.8%)	30 (12.5%)	152 (63%)	0.3
Hemoglobin (g/dl)	≥10	61 (25.4%)	28 (11.6%)	89 (37%)	0.3

Abbreviation: MODS, multi-organ dysfunction syndrome.

Indicates statistically significant results (p < 0.05).

Cardiovascular cases: congestive heart failure, cardiac arrest, and cardiogenic shock.

Miscellaneous: neutropenic fever, diabetic ketoacidosis, upper respiratory obstruction, and status epileptics.

Comorbid conditions were recorded in 46.7% of patients (n = 122) and were present among 40 deaths (16.7%). Of the 183 documented comorbidities, severe acute malnutrition (ICD-11:5B52) was the most prevalent, accounted for 27.7%, followed by anemia and other erythrocyte disorders (ICD-11:3A00-3A9Z), observed in 20.2%. HAI constituted the predominant in-PICU complication, occurring in 53.1% of complicated cases (Supplemental Table S1). Among 25 patients with HAI, 13 died (52.0%). Kaplan–Meier analysis showed lower survival in the HAI group, though the log-rank test did not reach statistical significance (P = .06), indicating a non-significant trend toward higher mortality (Supplemental File S2). Moreover, survival distributions did not differ significantly according to the presence of organ dysfunction (log-rank P = .20) or the administration of maintenance intravenous fluids (log-rank P = .60; Table 2).

Management of Pediatric Patient Admitted to the PICU

Upon admission, antibiotics were administered to 91.3% patients (n = 219). MV was required in 29.2% patients (n = 70). Survival distribution differed significantly between ventilated and non-ventilated patients (log-rank P < .001), with the approximately 12.5% higher cumulative incidence of death among ventilated patients. Among those receiving MV, ventilation lasting >7 days was observed in 5.8% deaths (n = 14). Inotropic agents were administered to 30.8% patients (n = 74), and survival distributions differed significantly according to inotrope use (log-rank P = .01), with 12.9% deaths (n = 31) occurring in this group (Table 3).

Table 3.

Types of Treatments Given for Patients Admitted to PICU.

Characteristics		Censored	Dead	Total	Log rank
Type of intervention given during PICU stays	Antibiotic	165 (68.7%)	54 (22.5%)	219 (91.3%)	0.3
	Mechanical ventilation	40 (16.7%)	30 (12.5%)	70 (29.2%)	<0.001^*
	Anticonvulsant	25 (10.4%)	4 (1.7%)	29 (12.1%)	0.5
	Analgesics	104 (43.4%)	31 (12.9%)	135 (56.3%)	0.2
	Steroid	56 (23.3%)	14 (5.8%)	70 (29.2%)	0.4
	Fluid therapy	114 (47.5%)	38 (15.8%)	152 (63.3%)	0.8
	Inotropes	43 (17.9%)	31 (12.9%)	74 (30.8%)	0.01^*
	Transfusion	26 (10.8%)	8 (3.3%)	34 (14.1%)	0.7
	Respiratory support	114 (47.5%)	39 (16.3%)	153 (63.8%)	0.8
	Nutritional treatment	57 (23.8%)	21 (8.7%)	78 (32.5%)	0.3
	Diuretics	26 (10.8%)	11 (4.6%)	37 (15.4%)	0.6
	ACE inhibitors	18 (7.5%)	4 (1.7%)	22 (9.2%)	0.9
	Anti-asthmatic	40 (16.7%)	8 (3.3%)	48 (20%)	0.2
	Stress prophylaxis	47 (19.6%)	16 (6.7%)	63 (26.3%)	0.8
	Anti-malaria	8 (3.3%)	0	8 (3.3%)	NA
	Sedative	36 (15%)	9 (3.7%)	45 (18.7%)	0.2
	Miscellaneous drugs^a	19 (7.9%)	7 (2.9%)	26 (10.8%)	0.9
Length of MV stay in days	<2	7 (2.9%)	5 (2.1%)	12 (5%)	0.1
	2-7	15 (6.2%)	11 (4.6%)	26 (10.8%)
	>7	18 (7.5%)	14 (5.8%)	32 (13.3%)
Catheters	Central line catheters	9 (3.8%)	4 (1.6%)	13 (5.4%)	0.6
Catheters	Urinary catheter	27 (11.2%)	11 (4.6%)	38 (15.8%)	0.8

Miscellaneous drugs: insulin, thrombosis prophylaxis, antifungal, beta-blockers, anti-hypertensive, and anti-helminthics. Respiratory support: nasal cannula, facemask, and noninvasive.

Indicates statistically significant results (p < 0.05).

Incidence Rate and PICU Outcomes

During the study period, 24.2% patients (n = 58) died in the PICU; however, 75.8% (n = 182) were censored. Of the censored cases, 47.5% (n = 114) improved and were transferred to the wards. The overall incidence rate of death was 5.37 deaths per 100 person-days (95% CI: 2.87-9.8; Figure 1).

Figure 1.

Outcome at discharge among pediatric patients admitted to the PICU.

Kaplan-Meier Curve for Variable Groups

According to this study, pediatric patients from rural areas had shorter survival estimates compared to their urban counterparts (Figure 2). Patients with a low consciousness level (≤8) exhibited a higher probability of mortality than those with a higher consciousness level (>8; Figure 3). Additionally, the survival probability for patients who did not receive inotropic agents was better than for those who did (Figure 4). Furthermore, patients with coexisting diseases experienced shorter survival probabilities than those without such conditions (Figure 5).

Figure 2.

Kaplan-Meier curves showing survival estimates for pediatric patients by residence.

Figure 3.

Kaplan-Meier curves showing survival estimates for pediatric patients regarding to conscious level.

Figure 4.

Kaplan-Meier curves showing survival estimates for pediatric patients regarding to inotropic agent.

Figure 5.

Kaplan-Meier curves showing survival estimates for pediatric patients with the presence of comorbidity.

Predictors of Mortality in the PICU

A Cox proportional hazard regression was performed to identify predictors of in-PICU mortality. Accordingly, residence, nutritional status, meningitis diagnosis, respiratory failure, the use of inotropic agents, need for a MV, consciousness level, in-PICU complications, and the presence of comorbid conditions were first assessed in bi-variable analysis. Variables with P < .20 were entered into the multi-variable model. Model adequacy was assessed using Cox–Snell residuals, which demonstrated that the observed cumulative hazard closely aligned with the expected line, indicating an acceptable fit of the Cox proportional hazards model (Supplemental Figure S2).

In the final multi-variable model, 4 variables were independently associated with an increased risk of death. Patients from rural areas had a 2.0-fold increased risk of death compared to those from urban areas (AHR: 2.0, 95% CI: 1.2-3.5; P = .013). The use of inotropic agents was associated with a 2.0-fold higher risk of mortality compared to those who did not receive them (AHR: 2.0, 95% CI: 1.2-3.5; P = .012). Unconscious patients (GCS < 8) had a 2.3-fold increased hazard of dying compared to conscious patients (AHR: 2.3, 95% CI: 1.2-4.4; P = .011). The presence of comorbidities increased the mortality risk by 1.9 times compared to patients without comorbidities (AHR: 1.9 95% CI: 1.1-3.4; P = .033; Table 4).

Table 4.

Bivariate and Multivariable Cox Proportional Hazard Regression to Identify Predictors of In-ICU Mortality at PICU.

Variables		CHR (95% CI)	P-value	AHR (95% CI)	P-value
Residence	Urban	1	.01	1	.013^*
Residence	Rural	2.1 (1.2-3.5)	.01	2.0 (1.2-3.5)	.013^*
Nutritional status	Normal	1	.03	1
Nutritional status	Malnourished	1.8 (1.0-3.1)	.03	1.1 (0.6-2.0)	.8
Meningitis	Yes	1.6 (0.8-3.0)	.13	1.7 (0.8-3.4)	.1
Respiratory failure	Yes	2.0 (1.1-3.4)	.02	1.5 (0.7-3.0)	.3
Inotropic agent	No	1	.001	1	.012^*
Inotropic agent	Yes	2.5 (1.5-4.1)	.001	2.0 (1.2-3.5)	.012^*
Mechanical ventilation	No	1	<.001	1	.2
Mechanical ventilation	Yes	2.7 (1.6-4.5)	<.001	1.5 (0.9-2.7)	.2
In PICU-complication	No	1	.1	1	.6
In PICU-complication	Yes	1.5 (0.9-2.7)	.1	1.2 (0.6-2.3)	.6
Consciousness level (GCS)	≥8	1	<.001	1
Consciousness level (GCS)	<8	2.6 (1.5-4.4)	<.001	2.3 (1.2-4.4)	.011^*
Presence of comorbidities	No	1	.006	1	.033^*
Presence of comorbidities	Yes	2.2 (1.3-3.8)	.006	1.9 (1.1-3.4)	.033^*

Note. The martingale residuals goodness-of-fit test shows χ²(1) = .1 and P-value = .5, indicating a good fit for the multivariate model. The global Schoenfeld test results are chi χ²(8) = 3.9 with a P-value of .9. There is no multicollinearity concern, with a maximum VIF of 1.24.

Indicates statistically significant results (p < 0.05).

Discussion

This multicenter prospective study provides the first comprehensive evidence on survival outcomes and predictors of mortality among pediatric patient admitted to tertiary hospital PICUs in Eastern Amhara. The main findings reveal a high PICU mortality rate, with rural residence, comorbidities, severe neurological impairment (GCS <8), and the need for inotropic support emerging as independent predictors of death. These findings highlight systemic challenge in pediatric critical care in resource-limited settings, where delayed referral, inadequate monitoring, and high infection burden contribute to poor outcomes.

The observed PICU mortality align with reports from other LMICs, including Ethiopia (32.6%), Pakistan (26.6%), and Mozambique (25%),^15,30,37 but is lower than studies from Tanzania (41.4%), Uganda (40.1%), and Egypt (37.95%).^14,16,38 Differences likely reflect variation in patient case mix, illness severity at admission, and the availability of critical care resources. For instance, a cohort in Jimma, Ethiopia, with a higher proportion of trauma cases,¹⁷ reported 40% mortality, particularly severe head injuries.³⁹ Conversely, our mortality is higher than reports from Iran (16.5%), Nepal (12.6%), and India (14%).^13,40,41 This elevated mortality in our cohort might attributed to systemic factors, including delayed presentation, limited emergency service, staff shortages, and insufficient equipment.^9
-12 In high income countries, greater staffing and advanced facilities are associated with <5% PICU mortality,^7,8 emphasizing the influence of systemic factors on outcomes

The elevated mortality observed in our cohorts may be partly attributable to the predominance of children under 5 years of age, who are particularly vulnerable to infection, malnutrition, and other preventable condition due to immature immune systems and lower physiological reserve. In addition, nearly half of admissions were emergencies, reflecting delayed presentation and limited access to intermediate level care, which underscores the need for early identification and timely referral.^42,43 Similarly, deaths were common among children with infectious diseases, consistent with other LMICs reports,^15,44 highlighting the continued burden of preventable infectious and antimicrobial resistance.

In this study, respiratory and central nervous system (CNS) disorders were the leading causes of PICU admission, consistent with previous studies.^15,40,45 This conditions frequently are associated with a high deterioration risk. These observations underscore the importance of timely ventilator support, early antimicrobial therapy, and intensive neurological monitoring. The relatively prolonged PICU stays observed in this cohort, align with similar settings,^33,44 likely reflect both the severity of illness at admission and development of in PICU complications such as HAI collectively contributing to the high mortality in this context.

Key predictors of mortality provide insight into both clinical severity and contextual challenges. Children from rural areas faced a higher mortality. This is consistent with the findings of studies done in another region of Ethiopia.²⁹ A study in Lebanon, rural-residing children had a 44% increased risk of in-hospital mortality.⁴⁶ Likewise, a study conducted in the United States found also a similar finding, partly due to increased exposure to drug-resistant bacteremia.⁴⁷ This possibly due to delays care-seeking, limited transportation, reliance on traditional healers, and higher prevalence of malnutrition and chronic infections.^48,49

Underlying comorbidities independently increased mortality risk, as seen in other LMICs studies.^25,50
-52 Conditions such as chronic malnutrition, HIV infection and tuberculosis reduce physiological reserve, impair immune function, and limit the ability to tolerate acute critical illness.⁵³ In resource limited settings, delayed diagnosis and sub-optimal management of these conditions often result in advanced diseases at PICU admission,⁵⁴ contributing to rapid deterioration and poor outcomes.

Severe neurological impairment at admission was a strong mortality predictor; consistent with prior studies.^55
-57 Depressed consciousness reflects advanced diseases severity, risk of aspiration, respiratory failure and hemodynamic instability.⁵⁸ In resource limiting settings, where capacity for continuous neurological monitoring and rapid airway intervention is constrained, these risks are further amplified.^54,59 Early recognition and supportive care are therefore critical.

The requirement for inotropic support was also associated with increased mortality, indicating profound circulatory compromise. This finding is supported by previous published researchs.^44,60 Inotropic therapy typically reflects severe shock or cardiac dysfunction, often secondary to sepsis or advanced infection.⁶¹ In resource constrained PICUs, delay in shock recognition, limited access to invasive monitoring, and high patient -to-staff ratios may hinder timey circulatory support.^59,62 These findings highlight the importance of prompt recognition and management of hemodynamic instability

Although log rank analysis showed that survival distribution significantly differed in both children requiring MV and admitted with respiratory failure, these factors were not independent predictor in multivariable models. However, previous studies have reported both as independent predictor of mortality.^17,24,33 This possibly due to limited sample size, case heterogeneity, or residual confounding. Larger multicenter studies are needed to clarify prognostic factors and guide context-specific interventions for early recognition and supportive care in PICUs.

Evidence based implication from our finding include the need for close monitoring and timely management of children with severe neurological impairment, comorbidities, or requiring inotropic support. Addressing underlying comorbidities, preventing and promptly managing HAI, and ensuring access to effective antimicrobials are supported by the data.

This study’s multicenter design strengthens generalizability and allowed assessment of variability across PICUs of Eastern Amhara. Limitation include missing caregiver socioeconomic data, inability to PIM2 score for all patients, relatively small sample size reducing multivariable power, observational design preventing causal inference, and potential residual confounding from unmeasured factors such as pre-hospital care differences or PICU resource constraints. Additionally, although vaccination records were prospectively verified, age-specific timing was not consistently available, so some children up-to-date for their age may have been misclassified as not fully vaccinated.

Conclusion

This multicenter study found a high incidence of in-PICU mortality among pediatric patients admitted to tertiary hospitals in Eastern Amhara. Mortality was independently associated with rural residence, severe neurological impairment (GCS <8), need for inotropic support, and the presence of underlying comorbidities, with children from rural areas and those under 5 years particularly vulnerable. These findings highlight the importance of early identification and close monitoring of high-risk children, timely referral from rural areas, and targeted management of comorbidities such as malnutrition and chronic illnesses. Prospective studies are warranted to evaluate interventions aimed at improving survival and clinical outcomes in PICUs in resource-limited settings.

Supplemental Material

sj-docx-3-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental material, sj-docx-3-gph-10.1177_30502225261424327 for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study by Abel Temeche Kassaw, Eliyas Addisu Taye, Desalegn Addis Mussie, Getachew Yitayew Tarekegn, Samuel Berihun Dagnew and Tigabu Eskeziya Zerihun in Sage Open Pediatrics

Supplemental Material

sj-docx-4-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental material, sj-docx-4-gph-10.1177_30502225261424327 for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study by Abel Temeche Kassaw, Eliyas Addisu Taye, Desalegn Addis Mussie, Getachew Yitayew Tarekegn, Samuel Berihun Dagnew and Tigabu Eskeziya Zerihun in Sage Open Pediatrics

Supplemental Material

sj-docx-5-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental material, sj-docx-5-gph-10.1177_30502225261424327 for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study by Abel Temeche Kassaw, Eliyas Addisu Taye, Desalegn Addis Mussie, Getachew Yitayew Tarekegn, Samuel Berihun Dagnew and Tigabu Eskeziya Zerihun in Sage Open Pediatrics

Supplemental Material

sj-pdf-1-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental material, sj-pdf-1-gph-10.1177_30502225261424327 for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study by Abel Temeche Kassaw, Eliyas Addisu Taye, Desalegn Addis Mussie, Getachew Yitayew Tarekegn, Samuel Berihun Dagnew and Tigabu Eskeziya Zerihun in Sage Open Pediatrics

Supplemental Material

sj-pdf-2-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental material, sj-pdf-2-gph-10.1177_30502225261424327 for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study by Abel Temeche Kassaw, Eliyas Addisu Taye, Desalegn Addis Mussie, Getachew Yitayew Tarekegn, Samuel Berihun Dagnew and Tigabu Eskeziya Zerihun in Sage Open Pediatrics

Footnotes

Acknowledgements

The authors extend their sincere gratitude to the pediatric staff at Debre Tabor, Tibebe Ghion, and Felege Hiwot Comprehensive Specialized Hospitals for their invaluable cooperation and support throughout the study. Finally, we offer our heartfelt appreciation to the families of all participating patients for their trust and consent.

ORCID iDs

Abel Temeche Kassaw

Desalegn Addis Mussie

Getachew Yitayew Tarekegn

Samuel Berihun Dagnew

Tigabu Eskeziya Zerihun

Ethical Considerations

Ethical approval was obtained from the Institutional Research Ethics Review Committee (IRERC) at Debre Tabor University (DTU) (Approval No: DTU/IRERC/007/2018), and permissions were obtained from the Chief Clinical Directors of all participating hospitals.

Consent to Participate

Written informed consent was secured from legal guardians after explanation of study objectives and procedures. Participation was voluntary, with the right to withdraw at any time without consequence. All data were anonymized and handled confidentially in accordance with the Declaration of Helsinki.

Author Contributions

AT: Conceptualization, Formal Analysis, Methodology, Interpretation, Writing–original draft, Writing–review and editing, EA and DA: Data curation, Formal Analysis, Writing–review and editing, SB, GY, and TE: contributed to the source, design, and drafting of the research manuscript. All authors read and approved the revised manuscript for publication.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Data will be disclosed on a reasonable request from the corresponding author.*

Supplemental Material

Supplemental material for this article is available online.

References

Marshall

Bosco

Adhikari

, et al. What is an intensive care unit? A report of the task force of the world federation of societies of intensive and critical care medicine. J Crit Care. 2017;37:270-276. doi:10.1016/j.jcrc.2016.07.015

Clancy

Johnson

Design, organization and staffing of the intensive care unit. Surgery. 2021;39:7-14. doi:10.1016/j.mpsur.2020.11.007

Zimmerman

Al-Saadi

Critical care services: scope of practice. In: Hidalgo

Pérez-Fernández

Rodríguez-Vega

, eds. Critical Care Administration: A Comprehensive Clinical Guide. Springer International Publishing; 2020;1-20.

Pronovost

Fau-Dorman

Fau-Morlock

, et al. Building safety into ICU care. J Crit Care. 2002;17(2):78-85. doi: 10.1053/jcrc.2002.34363.

Riviello

Letchford

Fau-Achieng

Fau-Newton

, et al. Critical care in resource-poor settings: lessons learned and future directions. Crit Care Med. 2011;39(4):860–867. doi: 10.1097/CCM.0b013e318206d6d5.

Gemke

Bonsel

van Vught

AJ.

Long-term survival and state of health after paediatric intensive care. Arch Dis Child. 1995;73:196-201.

Ohbe

Sasabuchi

Matsui

Fushimi

Yasunaga

Resource-rich intensive care units vs. standard intensive care units on patient mortality: a nationwide inpatient database study. J Med Assoc J. 2021;4:397-404. doi:10.31662/jmaj.2021-0098

Moynihan

Alexander

PMA

Schlapbach

, et al. Epidemiology of childhood death in Australian and New Zealand intensive care units. Intensive Care Med. 2019;45:1262-1271. doi:10.1007/s00134-019-05675-1

English

. Child survival: district hospitals and paediatricians. Arch Dis Child. 2005;90(9):974-978.

10.

Riviello

Kiviri

Fowler

, et al. Predicting mortality in low-income country ICUs: the Rwanda mortality probability model (R-MPM). PLoS One. 2016;11:e0155858.

11.

Tripathi

Kaur

Kashyap

Dong

Gajic

Murthy

A survey on the resources and practices in pediatric critical care of resource-rich and resource-limited countries. J Intensive Care. 2015;3(1):40. doi: 10.1186/s40560-015-0106-3.

12.

Diaz

Riviello

Papali

Adhikari

NKJ

Ferreira

JC.

Global critical care: moving forward in resource-limited settings. Ann Global Health. 2019;85(1):1-11. doi:10.5334/aogh.2413

13.

Shah

Thapa

Admission patterns and outcome in a pediatric intensive care unit in Nepal. Br J Med Res. 2014;4:4939-4945. doi:10.9734/BJMMR/2014/10318

14.

Sawe

Mfinanga

Lidenge

Fau-Mpondo

BCT

, et al. Disease patterns and clinical outcomes of patients admitted in intensive care units of tertiary referral hospitals of Tanzania. BMC Int Health Hum Rights. 2014;14:26. doi: 10.1186/1472-698X-14-26

15.

Punchak

Hall

Seni

, et al. Epidemiology of disease and mortality from a PICU in Mozambique. Pediatr Crit Care Med. 2018;19:e603-e610.

16.

Kwizera

Dünser

Nakibuuka

National intensive care unit bed capacity and ICU patient characteristics in a low income country. BMC Res Notes. 2012;5(1):475.

17.

Abebe

Girmay

, et al. The epidemiological profile of pediatric patients admitted to the general intensive care unit in an Ethiopian university hospital. Int J Gen Med. 2015;8:63-67. doi:10.2147/IJGM.S76378

18.

Tilahun

Heye

Admission pattern and outcome in a pediatric intensive care unit of Gondar university hospital. Ethiop Med J. 2019;57(2).

19.

Adedokun

Yaya

Childhood morbidity and its determinants: evidence from 31 countries in sub-Saharan Africa. BMJ Glob Health. 2020;5:e003109. doi:10.1136/bmjgh-2020-003109

20.

Ezbakhe

Pérez-Foguet

Child mortality levels and trends: a new compositional approach. Demogr Res. 2020;43:1263-1296. doi:10.4054/DemRes.2020.43.43

21.

World Health Organization (WHO). Newborns: improving survival and well-being. 2020. Accessed December 05, 2024. https://www.who.int/en/news-room/fact-sheets/detail/newborns-reducing-mortality

22.

Kliegman

Geme

JWS

. Nelson Textbook of Pediatrics E-Book. Elsevier; 2019.

23.

Ethiopian Public Health Institute (EPHI), Federal Ministry of Health (FMoH) and ICF. Ethiopia mini demographic and health survey 2019. 2021. Ethiopia: EPHI/FMoH/ICF

24.

Seifu

Eshetu

Tafesse

Hailu

Admission pattern, treatment outcomes, and associated factors for children admitted to pediatric intensive care unit of Tikur Anbessa specialized hospital, 2021: a retrospective cross-sectional study. BMC Anesthesiol. 2022;22:13. doi:10.1186/s12871-021-01556-7

25.

Haftu

Hailu

Medhaniye

G/Tsadik

. Assessment of pattern and treatment outcome of patients admitted to pediatric intensive care unit, Ayder referral hospital, Tigray, Ethiopia, 2015. BMC Res Notes. 2018;11:339. doi:10.1186/s13104-018-3432-4

26.

Abhulimhen-Iyoha

Pooboni

Vuppali

NKK

. Morbidity pattern and outcome of patients admitted into a pediatric intensive care unit in India. Indian J Clin Med. 2014;5:1. doi:10.4137/IJCM.S13902

27.

El Halal

MGDS

Barbieri

Filho

Trotta

EDA

Carvalho

PRA.

Admission source and mortality in a pediatric intensive care unit. Indian J Crit Care Med. 2012;16:81-86. doi:10.4103/0972-5229.99114

28.

Onyekwulu

Anya

. Pattern of admission and outcome of patients admitted into the intensive care unit of university of Nigeria teaching hospital Enugu: a 5-year review. Niger J Clin Pract. 2015;18(6):775–779. doi: 10.4103/1119-3077.163291.

29.

Dendir

Awoke

Alemu

, et al. Factors associated with the outcome of a pediatric patients admitted to intensive care unit in resource-limited setup: cross-sectional study. Pediatric Health Med Ther. 2023;14:71-79. doi:10.2147/PHMT.S389404

30.

Teshager

Amare

Tamirat

KS.

Incidence and predictors of mortality among children admitted to the pediatric intensive care unit at the university of Gondar comprehensive specialised hospital, northwest Ethiopia: a prospective observational cohort study. BMJ Open. 2020;10:e036746. doi:10.1136/bmjopen-2019-036746

31.

Vittinghoff

McCulloch

CE.

Relaxing the rule of ten events per variable in logistic and cox regression. Am J Epidemiol. 2007;165:710-718.

32.

Lyimo

Said

Kivuyo

Nkya

Furia

FF.

Mortality and associated factors among children admitted to an intensive care unit in Muhimbili national hospital, from the time of admission to three months after discharge: a prospective cohort study. BMC Pediatr. 2024;24:170. doi:10.1186/s12887-024-04620-6

33.

Bacha

Gadisa

Gudeta

Beressa

Negera

GZ.

Survival status and predictors of mortality among patients admitted to pediatric intensive care unit at selected tertiary care hospitals in Ethiopia: a prospective observational study. Clin Med Insights Pediatr. 2023;17:11795565231169498. doi: 10.1177/11795565231169498.

34.

Ilori

Kalu

QN.

Intensive care admissions and outcome at the university of Calabar teaching hospital, Nigeria. J Crit Care. 2012;27:105.e1-105.e4.

35.

Harrison

Weber

Jakob

Chute

CG.

ICD-11: an international classification of diseases for the twenty-first century. BMC Med Inform Decis Mak. 2021;21:206. doi:10.1186/s12911-021-01534-6

36.

Mehta

Chinthapalli

Glasgow coma scale explained. BMJ. 2019;365:l1296. doi:10.1136/bmj.l1296

37.

Khan

Basheer

Nayyar

, et al. Clinical profile and outcome in a paediatric intensive care unit of a tertiary care hospital in Pakistan. Pakistan Armed Forces Med J. 2020;70(6):1657-1661.

38.

Mahmoud

Abdelatif

Bakheet

MAM

Mohamed

MM.

Predictors of outcome in an Egyptian pediatric intensive care unit. Egypt J Hosp Med. 2023;90:1558-1569.

39.

McMillan

Teasdale

Weir

Stewart

Death after head injury: the 13 year outcome of a case control study. J Neurol Neurosurg Psychiatry. 2011;82:931-935.

40.

Valavi

Aminzadeh

Shirvani

Jaafari

Madhooshi

The main causes of mortality in pediatric intensive care unit in south west of Iran. Zahedan J Res Med Sci. 2018;20(4):e63006.

41.

Roy

Basu

Roy

BC.

Clinical profile and outcome of patients admitted to pediatric intensive care unit of a tertiary care teaching hospital in eastern India. J Med Sci Clin Res. 2018;6:1071-1075.

42.

Purcell

Prin

Sincavage

Kadyaudzu

Phillips

Charles

Outcomes following intensive care unit admission in a pediatric cohort in Malawi. J Trop Pediatr. 2020;66:621-629. doi:10.1093/tropej/fmaa025

43.

Huang

Yang

, et al. Shock index, pediatric age-adjusted predicts morbidity and mortality in children admitted to the intensive care unit. Front Pediatr. 2021;9:9. doi:10.3389/fped.2021.727466

44.

Nyirasafari

Corden

Karambizi

, et al. Predictors of mortality in a paediatric intensive care unit in Kigali, Rwanda. Paediatr Int Child Health. 2017;37:109-115.

45.

Mukhija

Chandra

Prasad

DP.

Clinical profile of patients admitted to the PICU of a tertiary care teaching hospital. Pediatr Rev Int J Pediatr Res. 2017;4:125-127. doi:10.17511/ijpr.2017.i02.06

46.

Leyenaar

Freyleue

Arakelyan

, et al. Rural-urban disparities in hospital services and outcomes for children with medical complexity. JAMA Network Open. 2024;7:e2435187. doi:10.1001/jamanetworkopen.2024.35187

47.

Whittington

Malone

Hogan

, et al. Staphylococcus Aureus bacteremia in pediatric patients: uncovering a rural health challenge. Open Forum Infect Dis. 2023;10:296. Doi:10.1093/ofid/ofad296

48.

Ventura

Hauschild

Barbosa

, et al. Undernutrition at PICU admission is predictor of 60-day mortality and PICU length of stay in critically ill children. J Acad Nutr Diet. 2020;120:219-229. doi:10.1016/j.jand.2019.06.250

49.

Simbila

Kilindimo

Sawe

, et al. Predictors and outcome of time to presentation among critically ill paediatric patients at emergency department of Muhimbili national hospital, Dar Es Salaam, Tanzania. BMC Pediatr. 2022;22:22. doi:10.1186/s12887-022-03503-y

50.

Botan

Gün

Şden

, et al. Characteristics and timing of mortality in children dying in pediatric intensive care: a 5-year experience. Acute Crit Care. 2022;37:644-653.

51.

Low

JZH

Lim

JKB

Tan

, et al. Epidemiology of paediatric intensive care unit admissions, deaths and organ donation candidacy: a single-centre audit. Ann Acad Med Singap. 2025;54:17-26.

52.

Daher

Al-Ammouri

Ghanem

Abu Zahra

Al-Zayadneh

Al-Iede

All-cause mortality in a pediatric intensive care unit at a teaching hospital in Amman, Jordan. Pediatr Int. 2022;64(1):e14940. doi:10.1111/ped.14940.

53.

Esper

Martin

. The impact of comorbid [corrected] conditions on critical illness. Crit Care Med. 2011;39(12):2728–2735. doi:10.1097/CCM.0b013e318236f27e.

54.

Kortz

Nielsen

Mediratta

, et al. The burden of critical illness in hospitalized children in low- and middle-income countries: protocol for a systematic review and meta-analysis. Front Pediatr. 2022;10:756643. DOI: 10.3389/fped.2022.756643.

55.

Abeytunge

Miller

Cameron

, et al. Development of a mortality prediction tool in pediatric severe traumatic brain injury. Neurotrauma Rep. 2021;2:115-122.

56.

Bao

, et al. Associated factors of trauma severity and mortality in pediatric patients admitted to intensive care unit; a 10-year retrospective study. Arch Acad Emerg Med. 2024;13 (1):e7. doi: 10.22037/aaem.v12i1.2413.

57.

Khajeh

Fayyazi

Miri-Aliabad

, et al. Comparison between the ability of Glasgow coma scale and full outline of unresponsiveness score to predict the mortality and discharge rate of pediatric intensive care unit patients. Iran J Pediatr. 2014;24(5):603–608.

58.

Rosenfeld

Mathiesen

TI.

The clinical and ethical challenges of treating comatose patients following severe brain injury. Acta Neurochir. 2023;165:829-831. doi:10.1007/s00701-022-05376-7

59.

Slusher

Kiragu

Day

, et al. Pediatric critical care in resource-limited settings—overview and lessons learned. Front Pediatr. 2018;6:49. doi: 10.3389/fped.2018.00049.

60.

Kumar Dey

Ghosh

Hemram

Mukherjee

Annigeri

Nair

Morbidity pattern with treatment outcome and predictors of mortality of children admitted to pediatric intensive care unit in a peripheral medical college in India. Acta Med Iran. 2021;59(8):491-498.

61.

Jordan

Nguyen

Fletcher

Garbern

SC.

Clinical prognostic models in children with sepsis in low- and middle-income countries: a systematic review and meta-analysis. Front Pediatr. 2024;12:1463986.

62.

Shahid

Rana

Maqsood

The evolving role of inotropes in pediatric septic shock: shifting paradigms from fluid first to vasoactive early. J Pediatr Surg. 2025;60:162448. doi:10.1016/j.jpedsurg.2025.162448

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.07 MB

0.03 MB

0.27 MB

0.20 MB

Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara,Ethiopia: A Prospective Multi-Center Cohort Study

Abstract

Background:

Objective:

Methods:

Results:

Conclusion:

Keywords

Introduction

Methods and Material

Study Setting, Design, and Period

Populations and Eligibility Criteria

Sample Size Calculation and Sampling Technique

Study Variables

Operational Definitions

Data Collection Procedures and Quality Assurance

Data Processing and Analysis Procedure

Ethical Approval and Informed Consent

Result

Socio-Demographic Characteristics of Study Participants

Clinical-Related Characteristics and Diagnosis of PICU

Management of Pediatric Patient Admitted to the PICU

Incidence Rate and PICU Outcomes

Kaplan-Meier Curve for Variable Groups

Predictors of Mortality in the PICU

Discussion

Conclusion

Supplemental Material

sj-docx-3-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental Material

sj-docx-4-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental Material

sj-docx-5-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental Material

sj-pdf-1-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Supplemental Material

sj-pdf-2-gph-10.1177_30502225261424327 – Supplemental material for Predictors of Mortality Among Pediatric Patients Admitted to Intensive Care Units in Eastern Amhara, Ethiopia: A Prospective Multi-Center Cohort Study

Footnotes

Acknowledgements

ORCID iDs

Ethical Considerations

Consent to Participate

Author Contributions

Funding

Declaration of Conflicting Interests

Data Availability Statement

Supplemental Material

References

Supplementary Material