Pediatric morbidity and mortality remain major indicators for evaluating the performance and quality of health systems. Despite a global decline in child mortality over the past two decades, the burden remains disproportionately high in low- and middle-income countries. In 2022, nearly five million children worldwide died before reaching their fifth birthday, equivalent to one death every six seconds, with the majority of these preventable deaths occurring in sub-Saharan Africa and South-East Asia [1].

Hospital mortality data provide crucial insights into the quality of pediatric care. They allow healthcare managers and providers to identify the main causes of death and adjust clinical and public health strategies accordingly. While many countries in sub-Saharan Africa have reported pediatric hospital mortality trends, there is a paucity of multicenter studies in the Democratic Republic of Congo (DRC), where most available data are single-center and often outdated [2–6].

The DRC faces multiple health system challenges, including high prevalence of infectious diseases, delays in health-seeking behavior, limited diagnostic capacities, and insufficient coverage of preventive programs such as immunization. Understanding the clinical patterns and determinants of pediatric hospital mortality in this context is essential for guiding national strategies aimed at reducing preventable deaths and achieving the Sustainable Development Goals (SDGs) related to child survival.

The objective of this study was therefore to provide an updated profile of pediatric hospital mortality across five reference hospitals in the DRC and to identify key risk factors associated with in-hospital deaths among children.

Methods

Study design and setting

We conducted a multicenter, retrospective cross-sectional study based on pediatric deaths recorded between January 2019 and December 2021. Data were collected from five reference hospitals located in four major regions of the Democratic Republic of Congo (DRC):

• North: Kisangani (Makiso General Referral Hospital)
• South: Lubumbashi (University Clinics of Lubumbashi)
• East: Kindu (Kindu General Referral Hospital)
• West: Kinshasa (National General Referral Hospital, formerly Mama Yemo, and Kimbanguist Hospital of Kimbanseke).

Study population

The study population included all children aged 1 month to 20 years who were admitted to the pediatric wards of the participating hospitals during the study period and subsequently died. Neonates (0–28 days) and children admitted for severe acute malnutrition were excluded. These exclusions were made because neonatal and malnutrition-related deaths are typically analyzed separately, given their distinct etiological profiles and specialized management protocols.

Sample size and sampling

A total of 2,097 pediatric admissions were recorded across the five hospitals, of which 146 deaths were documented and included in the analysis. All eligible cases during the study period were considered, thus no sampling technique was applied.

Data collection

Data were extracted from hospital records using a standardized questionnaire developed on the KoboCollect platform by the Unit of Prevention and Disease Control at the University of Kindu. Variables collected included:

• Sociodemographic: age, sex, month of admission.
• Clinical and anamnesis: presence of chronic disease, presenting complaints, delay before consultation.
• Therapeutic and outcome: admission diagnosis, length of hospital stay, and vital status.

Definitions of indicators

• Crude hospital mortality rate = (total number of pediatric deaths / total pediatric admissions) × 100.
• Proportional mortality rate (PMR) = (number of deaths from a specific disease / total pediatric deaths) × 100.
• Specific mortality rate (SMR) = (number of deaths from a specific disease / total admissions for that disease) × 100.
• Case fatality rate (CFR) = (number of deaths from a specific disease / total cases of that disease) × 100.

Data analysis

Data were analyzed using SPSS version 25 and Microsoft Excel 2017. Descriptive statistics were computed as frequencies, percentages, means, and standard deviations. Bivariate analyses were performed using Pearson’s chi-square test to assess associations between independent variables and mortality. A p-value <0.05 was considered statistically significant.

Ethical considerations

Prior authorization was obtained from the administrative boards of all participating hospitals before data collection. Ethical approval was granted by the institutional ethics committee of the University of Kindu. All data were anonymized, and confidentiality was strictly maintained throughout the study.

Table I: Proportional mortality, specific mortality and lethality.

NA: not applicable

Table I: Sociodemographic, anamnestic, therapeutic and outcomes characteristics of deceased children.

Table III: Common Complaints of deceased children.

NA: Not applicable

Results

Overall hospital mortality

During the study period, 2,097 pediatric admissions were recorded across the five hospitals. Among these, 146 deaths occurred, corresponding to a hospital mortality rate of 7.0%. The mean age at death was 3.7 years (±2.4), with 71.2% occurring in children under five years (OR=2.4; 95% CI: 1.5–3.9; p=0.001). This distribution is shown in Figure 3, where children under five represented more than two-thirds of all deaths. The male-to-female ratio was 1.14, with no significant difference observed (p=0.959).

Causes of death

Five conditions accounted for more than four out of five pediatric deaths: malaria (52.1%), anemia (21.9%), meningitis (18.5%), pneumonia (15.8%), and sepsis (12.3%). When case fatality was considered, the most lethal conditions were acute kidney injury (CFR=40.6%), meningitis (15.1%), pneumonia (11.3%), and malaria (5.7%). These results highlight that while malaria contributed the largest share of deaths, acute kidney injury, although less frequent, was associated with the highest lethality. These findings are illustrated in Figure 1, which compares the proportional mortality rate and case fatality rate across the leading causes of death.

Temporal distribution

Analysis of the monthly distribution of deaths revealed three distinct seasonal peaks: April–May, September–October, and

December (p=0.008). These peaks coincided with periods of high malaria transmission in the DRC. As shown in Figure 2, mortality curves followed a bimodal pattern, reinforcing the influence of environmental and seasonal factors on pediatric hospital deaths.

Clinical presentation

The most common presenting complaints among deceased children were fever (50%), dyspnea (19.9%), physical asthenia (15.8%), convulsions (15.1%), vomiting (12.3%), loss of consciousness (5.5%), and abdominal pain (4.8%) (all p<0.05). Fever was almost universal and often associated with severe malaria or sepsis, while dyspnea and convulsions reflected the burden of severe respiratory and neurological infections.

Health-seeking behavior and chronic conditions

The average delay before consultation was 6 days (±3). Notably, 60.3% of deceased children presented after more than three days of illness onset (OR=3.1; 95% CI: 1.9–5.2; p<0.0001). Chronic disease was identified in 13% of cases and was significantly associated with mortality (p=0.005). This suggests that both delayed health-seeking behavior and underlying chronic morbidity substantially contributed to the risk of in-hospital death.

Length of hospital stay

The mean duration of hospitalization among deceased children was 5 days. However, the majority of deaths (63.7%) occurred within the first 72 hours of admission (OR=2.9; 95% CI: 1.7–4.8; p<0.0001), reflecting both the severity of cases at admission and delays in accessing care.

Figure 1: Graphical comparison of mortality indicators

Figure 2: Illustration of the monthly evolution of mortality

Figure 3: Illustration of the distribution of mortality by age group.