This was a mixed-methods study with a comprehensive design. The quantitative component comprised two parts. The first part was cross-sectional and analytical. The second part consisted of a case-control study that established links between perinatal deaths and underlying factors. The study covered a three-year period, from January 1, 2022, to December 31, 2024.

This study was conducted in the 18 health zones of Kasai Province, in the Democratic Republic of Congo (DRC).

The study population consisted of newborns who died and those who experienced complications during the first 28 days of extrauterine life during the study period.

Regarding the sampling method, for the quantitative component, the study adopted an exhaustive approach by including all perinatal deaths recorded in the documentary sources of the 18 health zones of the Kasai Provincial Health Directorate (DPS). We matched one case to three controls (1:3). This was a one-to-one match, where the cases were perinatal deaths occurring in the departments and the controls were complications recorded in the same departments and during the same period. We have collected 1059 files.

For the qualitative component, a purposive selection process was applied to choose participants. They included secondary actors (healthcare providers) and primary actors (community members, community health workers, and community leaders) divided into 10 focus groups of 8 to 10 people.

Regarding the inclusion criteria, for the quantitative component, all complete and usable medical records of newborns who experienced complications or died during their first 28 days of life in the targeted health zones were retained. The qualitative component included healthcare providers and community members who voluntarily agreed to participate in the study.

Regarding exclusion criteria, for the quantitative component, incomplete medical records or those lacking the variables necessary for analysis were excluded. For the qualitative component, any person who did not formally give their informed consent or who decided to withdraw from the process was excluded from the study.

The dependent variable was perinatal death. The independent variables concerned maternal characteristics (age, place of residence, marital status, level of education, occupation, parity, inter-pregnancy interval, ANC follow-up) and newborn characteristics (gestational age, age, sex, birth weight [according to gestational age], presence of asphyxia, neonatal sepsis/infection, congenital malformation, twin pregnancy).

For the quantitative component, data collection was carried out through a review of hospital records. The tools used were maternity registers, prenatal care (ANC) consultation forms, delivery registers, partograms, hospitalization registers, and perinatal death review forms. A standardized survey questionnaire was pre-tested on 20 records to adjust the tool in terms of clarity and usability. Each record was assigned a unique code. The data were entered into a secure Excel database and then saved to an external storage device. For the qualitative component, data were collected using a semi-structured interview guide administered until saturation, through interviews and focus groups with healthcare providers, health zone managers, and community leaders/liaisons.

Quality assurance ensured data reliability. Interviewers and supervisors were rigorously recruited based on their education level and experience, and received prior training. Systematic triangulation was performed between data from health facilities, data transmission tools, DHIS2 data, and verbatim transcripts from the various focus groups (community and management teams).

The research protocol was formally approved by the Ethics Committee of the Faculty of Medicine at the University of Lubumbashi (No. UNILU/CEM/036/2025). Authorization from the provincial health authorities, informed consent, confidentiality, and fairness were required before data collection began.

Table 1. General characteristics of newborns and mothers included in the study (Kasai Province, 2022-2024).

The study included a total sample of 1059 newborns in Kasai Province between 2022 and 2024. Regarding maternal age, 2.46% of the newborns were born to mothers under 18 years old. Males represented 39.57% of the sample. A large majority (74.32%) had a normal birth weight, ranging from 2500 to 3500 grams (See Table 1).

The profile of the mothers in the study was characterized by a majority residing in rural areas (71.29%) and an age group largely concentrated between 18 and 40 years (95.47%). More than two-thirds (71.95%) of the babies were born after a pregnancy of more than 36 weeks of amenorrhea.

Perinatal Mortality Rate in Kasai Province (2022-2024)

Between 2022 and 2024, the DHIS2 system recorded 730,934 live births and 1303 perinatal deaths, representing a perinatal mortality rate of 2%. Targeted analysis of our surveys reveals a much more critical situation, with 353 perinatal deaths out of 1059 recorded obstetric complications, representing a case fatality rate of 33.3%. (See Figure 1Figure 1)

Figure 1.Perinatal mortality rate in Kasai Province from 2022 to 2024. Source: (DHIS2) and our surveys.

A comparative analysis between deceased and surviving newborns highlights several factors associated with perinatal deaths. For each of the following criteria, the observed differences are statistically significant (p ≤ 0.05): Low birth weight was the most pronounced factor (52.69% of deceased newborns weighed less than 2500 grams, compared to only 2.12% of survivors in this category). Rural areas showed a significantly higher mortality rate (93.20% of deaths occurred in rural areas, compared to 60.34% of survivors in these areas). Among premature infants (born between 31 and 36 weeks of gestation), the mortality rate was 44.48%, compared to 15.44% of survivors. The same difference was observed among premature infants born before 31 weeks of gestation. The absence of prenatal care negatively impacts survival (35.69% of babies who died were born to mothers who had not received any prenatal care, compared to 23.65% of survivors). Boys were more affected, representing 52.69% of deaths compared to 34.14% of survivors. Conversely, the presence of qualified personnel is a major protective factor: in this category, the survival rate rises to 66.15% (See Table 2).

Table 2. Factors associated with neonatal deaths among newborns interviewed in Kasai Province from 2022 to 2024.

The study of causes of death in the 353 deceased newborns revealed statistically significant differences (p ≤ 0.05) with respect to dystocia (50.71% in deceased newborns vs. 30.45% in survivors), asphyxia (57.79% in deceased newborns vs. 27.20% in survivors), and neonatal sepsis (35.98% in deceased newborns vs. 18.98% in survivors) (See Table 3).

Table 3. Causes of newborn deaths in Kasai province from 2022 to 2024.

Multivariate analysis revealed a statistically significant association (p ≤ 0.05) between perinatal death and the following seven factors: dystocia (p = 0.0000), asphyxia (p = 0.0000), lack of skilled assistance during childbirth (p = 0.0000), living in a rural area (p = 0.0000), low birth weight (p = 0.0000), and sepsis (p = 0.0000) (See Table 4).

According to the testimonies gathered, responsibility for perinatal deaths is

Table 4. Adjustment for confounding variables.

perceived as shared and multidimensional. It simultaneously involves three levels of actors: the community, the administration of healthcare facilities, and healthcare providers. This approach highlights the highly systemic nature of perinatal mortality, fitting directly within the analytical framework of the three delays (access to care, referral, and management).

At the community level, the interviews revealed major shortcomings related to pregnancy monitoring and the recognition of warning signs in newborns. The main factors identified were the absence or inadequacy of prenatal consultations (PNC), failure to follow medical instructions given during these consultations, and families’ lack of awareness of danger signs.

The participants’ accounts noted the following:

“Many new mothers don’t recognize the warning signs their babies are showing. They bring them to the hospital late, reducing their chances of recovery.”

“Some women don’t attend prenatal appointments, and others don’t follow the instructions they are given.”

“Failure to follow instructions [...] prevents women from protecting their babies during birth. Often, no precautions are taken beforehand to minimize the risks.”

The role of administrative management was also highlighted, particularly regarding logistics and material and human resources in the delivery room: stockouts of essential medications and blocked access to the maternity ward’s emergency kits; lack of equipment maintenance and absence of practical training on resuscitation equipment; and the constraints related to free healthcare (which is insufficiently funded) directly impact staff motivation and behavior.

The participants’ testimonies indicated the following:

“The managers of the facilities often don’t make available the supplies and equipment needed to resuscitate newborns with complications.”

“Sometimes we lack an emergency kit in the maternity ward, even though there are medications available at the hospital’s central pharmacy.”

“We have plenty of equipment and resuscitation kits. When we receive them, there isn’t a technician to teach us how to use them.”

“Here, maternity care is free, yet the staff are poorly paid. This leads to negligence and mistreatment of women and their babies by the providers.”

Clinically, the shortcomings of healthcare professionals are evident in their technical, ethical, and organizational aspects. Delays in initiating care, errors in assessing the severity of cases, and belated referrals to higher-level hospitals are all signs of inadequate clinical management. Attitudes perceived as harsh, capricious, or stigmatizing, which discourage women from returning or seeking timely care, point to a strained patient-provider relationship. Lack of planning (absence of individualized birth plans), excessive workloads, insufficient continuing education, and demotivation due to low wages are all contributing factors to poor working conditions and skills.

Participant testimonies are included below:

On planning and ethics:

“Some providers don’t develop a birth plan with women during prenatal consultations [...]. There’s a lack of professionalism among providers.”

“We providers are also too capricious. We waste a lot of time judging women [...] instead of starting care to save the distressed fetus. Other women don’t come back because of this behavior.”

On emergency management and referrals:

“The perinatal deaths recorded here are often cases referred late by Registered Nurses (RNs), with no fetal heart sounds.”

“The case is there, but the providers intervene late. The ignorance of some providers is also at the root of these deaths.”

On motivation:

“People say we work hard, but we aren’t paid enough. This is the root cause of neglect.”

In summary, perinatal mortality cannot be attributed to a single factor. It results from a chain reaction involving a combination of factors: community delays in seeking care (first delay), institutional management barriers (second delay), and weaknesses in the quality of clinical care within healthcare facilities themselves (third delay). Addressing this problem therefore requires cross-cutting interventions that simultaneously address community awareness, healthcare system financing, and continuing education for healthcare professionals.