Prevalence of Obstetric Referrals in the Dassa-Glazoué Health District, Benin, from 2020 to 2024 ()
1. Introduction
Pregnancy complications sometimes result in obstetric emergencies requiring evacuation and referral to hospitals. These obstetric emergencies necessitate prompt and effective management to prevent maternal and fetal deaths. In Africa, gynecological and obstetric emergencies are estimated to account for 30% - 98% of overall maternal mortality, with hemorrhage being the leading cause [1]. In Benin, obstetric referrals are estimated to be associated with 70% - 90% of maternal deaths in hospitals [2]. They represent a significant public health issue, predominantly affecting developing countries, and reflect the development level of their healthcare systems [3]. Obstetric referrals are characterized by high frequency [4] [5] and often require evacuation from peripheral maternity centers, with both maternal and fetal prognosis being uncertain. This prognosis is worsened by poor prenatal follow-up, inadequacies in the management of parturients at health centers, geographic inaccessibility of referral centers, and lack of logistical, material, and financial resources. No study has been conducted on referrals in the DASSA-Glazoué health district. The objective of this study was to assess the prevalence of obstetric referrals in the DASSA-Glazoué health district.
2. Methods
This was a cross-sectional analytical study with retrospective data collection from January 1, 2020, to December 31, 2024, a period of five years. The study population included all women referred to two hospitals in the Dassa-Glazoué health district (Zone Hospital and l’Abbraccio Hospital) during the study period. All women referred between January 1, 2020, and December 31, 2024, were included. Excluded were incomplete or unusable medical records of non-referred women. sampling consisted of a census of all eligible referred patient records. The sample size corresponded to the total number of records meeting the inclusion criteria. The dependent variable was the prevalence of referred women during the study period. Independent variables included socio-demographic characteristics, referral-related factors, admission status, and diagnoses at referral and admission. A digital questionnaire on KoboCollect, configured on a smartphone, was used for data collection after reviewing medical records. Data were collected from June 18 to July 20, 2025, and analyzed using Epi Info 7 software. Frequency measures and bivariate analyses were used to describe and identify associated factors. Qualitative variables were expressed as proportions, and quantitative variables as means with standard deviations. Text processing was performed using Microsoft Word 2016, and tables and graphs were created using Microsoft Excel 2016. Comparisons of qualitative variables were performed using Chi-square or Fisher’s exact tests depending on theoretical counts. Statistical significance was set at p < 0.05.
Approval was obtained from the protocol review committee of the National School for Training Public Health Technicians and Epidemiological Surveillance (ENATSE), the Local Ethics Committee for Biomedical Research of the University of Parakou (CLERB-UP), and local health authorities. Informed consent and confidentiality were ensured.
3. Results
A total of 2771 women were included, of whom 827 were referred, corresponding to a prevalence of 29.84%. Among these, 20 maternal deaths and 18 fetal deaths were recorded, representing a frequency of 0.72%.
3.1. Socio-Demographic Characteristics of Referred Women
The mean age of referred women was 26.56 ± 6.62 years, ranging from 12 to 48 years. Most women were aged 24 - 31 years (36.52%), followed by the 17 - 24 years age group (35.67%), 32 - 38 years (18.62%), over 38 years old (6.17%) and under 17 years old 3.02%.
The mean gravidity of the women was 3.02 ± 2.11, with paucigravidae accounting for 50.52%, primigravidae 30.31%, multigravidae 20.21%, and grand multigravidae 13.24%.
The mean parity was 1.87 ± 1.97, with nulliparous women at 32.44%, pauciparous 27.54%, primiparous 21.26%, multiparous 12.99%, and grand multiparous 5.77%.
Most referred women resided in urban areas (70.16%), and 40.09% were homemakers. The table presents the distribution of referred women according to place of residence and socio-professional occupation (Table 1).
Table 1. Distribution of referred women in the Dassa-Glazoué Health District from 2020 to 2024 according to place of residence and socio-professional occupation.
|
Numbers |
Percentage (%) |
Place of residence |
|
|
Urban |
580 |
70.13 |
Rural |
233 |
28.17 |
Not specified |
14 |
1.70 |
Socio-professional occupation |
|
|
Housewife |
331 |
40.02 |
Civil servant |
70 |
8.46 |
Artisan |
154 |
18.62 |
Farmer |
29 |
3.51 |
Trader/Vendor |
127 |
15.36 |
Pupil/Student/Apprentice |
96 |
11.61 |
Not specified |
20 |
2.42 |
3.2. Characteristics Related to the Referral
Most women (91.41%) came from a health center (Table 2).
Table 2. Distribution of referred women according to the referral center in the Dassa-Glazoué Health District from 2020 to 2024.
|
Numbers |
Percentage (%) |
Referral center |
|
|
Health center |
756 |
91.41 |
Private clinic |
21 |
2.54 |
District hospital |
45 |
5.44 |
Medical practice/Care facility |
01 |
0.12 |
Not specified |
04 |
0.48 |
Referral with form, intravenous access, qualification of referring personnel, and duration of stay at the referring center
Among the women received, 58.52% were in the active phase of labor, of whom 48% had a partograph. They were referred with a referral form and intravenous access in 93.83% and 74.49% of cases, respectively. Referrals were mainly made by midwives (69.65%), and the duration of stay at the referring center was one day (1 day) in 88.97% of cases (Table 3).
Table 3. Distribution of referred women in the Dassa-Glazoué Health District from 2020 to 2024 according to referral with form and intravenous access, qualification of the referring personnel, and duration of stay at the referring center.
|
Numbers |
Percentage (%) |
Referred with referral form |
|
|
Yes |
776 |
93.83 |
No |
51 |
6.17 |
Referred with intravenous access |
|
|
Yes |
616 |
74.49 |
No |
211 |
25.51 |
Referred with partograph |
|
|
Yes |
739 |
89.36 |
No |
88 |
10.64 |
Qualification of the referring healthcare provider |
|
|
General practitioner |
03 |
0.36 |
Gynecologist |
06 |
0.73 |
Midwife |
576 |
69.65 |
Nurse |
93 |
11.25 |
Nursing assistant |
88 |
10.64 |
Not specified |
61 |
7.38 |
Length of hospital stay |
|
|
0 day |
66 |
8.23 |
1 day |
734 |
88.75 |
2 days |
25 |
3.02 |
At the time of referral, 42.20% of women arrived at one of the two hospitals within less than 2 hours, with motorcycles being the main mode of transport (58.77%). Inadequate technical facilities were the most common reason for referral from 2020 to 2024, accounting for 46.43% according to the time interval (Table 4).
Table 4. Distribution of referred women in the Dassa-Glazoué Health District according to time between referral and arrival, referral conditions, and reason for referral.
|
Numbers |
Percentage (%) |
Time interval between referral and arrival |
|
|
<2 hours |
349 |
42.20 |
2 - 4 hours |
278 |
33.62 |
>4 hours |
155 |
18.74 |
Not specified |
45 |
5.44 |
Distance traveled before arrival at the receiving
center |
|
|
<10 km |
102 |
12.33 |
11 - 15 km |
186 |
22.49 |
16 - 20 km |
227 |
27.45 |
21 - 25 km |
149 |
18.02 |
26 - 30 km |
99 |
11.97 |
31 - 40 km |
64 |
7.74 |
Mode of referral/transportation |
|
|
Ambulance |
134 |
16.20 |
Motorcycle |
486 |
58.77 |
Non-medical vehicle |
127 |
15.36 |
Not specified |
80 |
9.68 |
Reason for referral |
|
|
Lack of medication |
09 |
1.09 |
No response to treatment |
25 |
3.02 |
Lack of diagnostic facilities |
73 |
8.83 |
Lack of qualified personnel |
108 |
13.05 |
Inadequate technical facilities |
384 |
46.43 |
Others* |
05 |
0.60 |
Not specified |
223 |
26.96 |
Km: kilometer Others*: Better management, patient request, no issues reported.
3.3. Referral Diagnosis
Pregnant women were mainly referred for excessive uterine height (18%) and for a scarred uterus (16%) (Figure 1).
Others: cord prolapse, polyhydramnios; bleeding during pregnancy, perinatal asphyxia; macrosomia; labor complications; anhydramnios.
Figure 1. Distribution of women in the Dassa-Glazoué Health District from 2020 to 2024 according to the referral diagnosis.
3.4. General Condition at Admission
At admission, the general condition was preserved in 98.52% of women, with a good level of consciousness in 98.99% of these women.
The main diagnoses at admission were gestational hypertension (9%), post-term pregnancy (8%), and premature rupture of membranes (8%) and Labor in a scarred uterus (Figure 2).
Others: cord prolapse, polyhydramnios; bleeding during pregnancy, perinatal asphyxia; labor complications; anhydramnios.
Figure 2. Distribution of referred women in the Dassa-Glazoué Health District from 2020 to 2024 according to the diagnosis at admission.
3.5. Relationship between Maternal Mortality and
Referral-Related Characteristics
The reason for admission was statistically associated with maternal deaths. Indeed, pregnant women admitted for pain (p < 0.001) had a 0.16 times lower risk of death (Table 5).
Table 5. Relationship between maternal mortality and obstetric referral characteristics in the Dassa-Glazoué Health District from 2020 to 2024.
|
|
Death |
|
|
|
N |
Yes |
No |
PR |
IC95% |
P-value |
827 |
N |
% |
n |
% |
Reason for admission |
|
|
|
|
|
|
|
|
Pallor |
09 |
00 |
0.0 |
09 |
100 |
- |
- |
- |
Pain |
280 |
07 |
2.5 |
273 |
97.50 |
0.16 |
[0.06 - 0.45] |
<0.001 |
Altered level of consciousness |
27 |
00 |
0.0 |
27 |
100 |
- |
- |
- |
Headache |
100 |
03 |
3.0 |
97 |
97.0 |
1.37 |
[0.36 - 5.2] |
0.641 |
Hemorrhage |
89 |
03 |
3.37 |
86 |
96.63 |
1.54 |
[0.41 - 5.84] |
0.522 |
Eclampsia |
02 |
00 |
0.0 |
02 |
100 |
- |
- |
- |
Premature rupture of membranes (PROM) |
320 |
07 |
2.19 |
313 |
97.81 |
01 |
|
|
Reason for referral |
|
|
|
|
|
|
|
|
Lack of medication |
109 |
00 |
0.0 |
09 |
100 |
- |
- |
|
No response to treatment |
53 |
00 |
0.0 |
25 |
100 |
- |
- |
|
Lack of diagnostic facilities |
173 |
00 |
0.0 |
73 |
100 |
- |
- |
|
Lack of qualified personnel |
108 |
00 |
0.0 |
108 |
100 |
- |
- |
|
Inadequate technical facilities |
384 |
04 |
|
480 |
|
01 |
|
|
PR*= Prevalence report, CI** = Confidence Interval.
4. Discussion
4.1. Prevalence of Obstetric Referrals
The prevalence of obstetric referrals was 29.84%, indicating a high proportion of referrals, which reflects congestion at referral centers or inadequate capacity of primary health care facilities to manage obstetric complications. This rate is comparable to that reported in other settings in Benin by Tshabu Aguemon et al. at the University Clinic of Gynecology and Obstetrics (CUGO) of Benin in 2012, where the referral rate was 30.38% [6]. This finding highlights the limitations of primary health care structures in managing a substantial proportion of obstetric complications and underscores the absolute necessity of an effective referral system.
A total of 20 maternal deaths (0.72%), 18 fetal deaths, and a case fatality rate of 0.024 were recorded. Although these rates (mortality and case fatality) appear relatively low, they remain concerning in a context where the majority of maternal deaths are preventable through timely and appropriate care. The World Health Organization emphasizes that the persistence of maternal deaths within health facilities often reflects delays in the decision to refer, delays in transportation, or delays in the provision of adequate care upon arrival commonly described as the “three delays” model [1].
The concomitant occurrence of 18 fetal deaths highlights the negative impact of severe obstetric complications not only on maternal survival but also on fetal outcomes. This association between obstetric referral and adverse perinatal outcomes has been widely documented by Say et al. confirming that complicated pregnancies requiring referral constitute a high-risk group for both maternal and fetal morbidity and mortality [7].
4.2. Sociodemographic Profile
The mean age was 26.56 years, with most women aged 24 - 31 years. Ghardallo in Tunisia found a mean age of 29.5 years [8], and Ibrahima in Guinea reported 25.1 years [9], closely matching our study and reflecting the reproductive-age population.
The mean parity was 1.87, with a majority being nulliparous or pauciparous, which could indicate a high-risk population or inadequate management of early pregnancies. Ghardallo et al. reported 47.6% nulliparous and 22.8% primiparous women [8], similar to our findings.
Most women resided in urban areas (70.16%), reflecting better accessibility, but urbanization alone did not reduce referrals or complications. Ibrahim in Guinea reported 77% of women coming from urban areas [9], consistent with our context, though some studies, such as Ghardallo et al. in Tunisia, report a higher proportion of rural women, depending on geographic context and inclusion criteria [8].
4.3. Referral-Related Characteristics
Most women came from health centers (91.41%), indicating a centralized referral structure, which may cause delays and overcrowding. Takai et al. in Nigeria (2023) reported that 70% - 90% of referrals came from primary-level centers [10], noting that centralization overloads referral hospitals. This reflects strong dependence on secondary or tertiary centers, as in pyramidal health systems in many African countries.
Most women were referred with a referral form (93.83%) and intravenous access (74.49%). Neelam et al. in Bangladesh (2015) reported frequent gaps in pre-transfer preparation [11]. The presence of a formal referral form and initial treatment (intravenous access, stabilization, emergency care) remains insufficient in many African health systems, increasing the risk of severe complications upon arrival. The high rates of referral documentation and intravenous access on admission are rather favorable and reflect a good level of formal organization and structured management of the referral process. This represents a strength of the health system in Benin. However, these indicators do not necessarily guarantee the quality or timeliness of care, as highlighted in the work of Leonardo et al. and Takai et al. [9] [11].
A short duration of stay at the referring center (1 day for 88.97% of women) may reflect emergency management but also potential gaps in post-referral follow-up.
4.4. Referral Delays and Conditions
Women (42.20%) arrived within 2 hours, a relatively short delay, though this does not ensure quality of care. Neelam (2008) reported delays exceeding 3 hours [10]. Time is a crucial factor in managing obstetric emergencies. The majority used motorcycles (58.77%), reflecting logistical and infrastructure challenges and increased transport risk (accidents, delays). Extensive use of motorcycles, taxis, or community transport has been described by Roger in Ouagadougou [12]. Leonardo [11] noted that even when transport reaches the facility, institutional delays may still result in deaths. Kyei-Nimakoh highlighted two of the “three delays” in obstetric care, emphasizing transport time and delay at arrival for receiving appropriate care [13]. These obstacles are often interdependent and can cause significant delays in emergency obstetric management [14] [15].
The main reason for referral was inadequate technical facilities (46.43%), reflecting structural weaknesses in primary health centers and the need for equipment and training improvements.
4.5. Referral Diagnosis
The main indications for referral were excessive uterine height (18%) and scarred uterus (16%), representing frequent obstetric complications requiring specialized care. Neelam in Bangladesh reported prematurity (64.5%) and hypertension (27.6%) as referral diagnoses [10]. Most women did not present with altered general condition at admission (98.52%), suggesting relatively early intervention or case selection.
4.6. Relationship between Admission Reason, Referral, and
Mortality
Pain as the reason for admission was significantly associated with an increased risk of mortality (p < 0.001), suggesting that pain may represent a marker of clinical deterioration or severe obstetric complications [16]. Leonado reported that among maternal deaths, direct causes including hemorrhage, sepsis, uterine rupture, and complications of prolonged or obstructed labor were predominant [11]. Takai, in Nigeria, identified indications such as pre-eclampsia, eclampsia, and hemorrhage as major contributors to maternal mortality [9]. In a study conducted by Ugochukwu, the leading causes of maternal death were eclampsia/pre-eclampsia (approximately 41%), obstetric hemorrhage (37.5%), and respiratory distress, with an overall mortality rate of 48% [17]. These findings corroborate that hypertensive disorders of pregnancy remain among the leading causes of maternal mortality, particularly in sub-Saharan Africa [18].
5. Conclusion
The prevalence of referrals was high in the Dassa-Glazoué Health District from 2020 to 2024. The formal referral system is relatively well-structured (health centers, hospital, referral forms, intravenous access) but weakened by technical insufficiencies, non-medicalized transport, organizational delays, and limited technical capacity. The strong association between specific diagnoses (uterine rupture, infection, severe hypertension) and mortality underscores the critical importance of quality and rapid management from the time of referral.
6. Study Limitations
This study has methodological limitations (retrospective design, reliance on medical records, absence of multivariate analysis). Additionally, important aspects are often missing in such studies: actual quality of transfers, time from referral decision to departure, stabilization before transport, accompaniment during transport, availability of resources (blood, medications, personnel), quality of care on arrival, and post-operative follow-up. Prospective, ideally multicenter studies with comprehensive data collection, including transport, stabilization, care, and follow-up, are needed to better assess the impact of referral on maternal and fetal morbidity and mortality.
Ethical Approval
Informed consent for all participants.