This was an analytical cross-sectional study. Patients in precarious security areas were compared to those in safer areas for certain characteristics.

Our study took place in the gynecology and obstetrics department of the ORTH. This is the referral center of the health facilities of the Northern region, which include 5 medical centers with surgical antenna offering comprehensive obstetric and newborn care, and 227 primary health care facilities that are supposed to provide basic emergency obstetric and newborn care. Unfortunately, 69.2% of these health facilities did not meet the minimum standards for qualified health workers in 2020 [8] . Emergency evacuations to the referral center are carried out using ambulances, most of which are stationed in medical centers. Access to peripheral health centers is often impossible due to poor road conditions, the risk of improvised explosive devices or targeted attacks on ambulances. Patients are then evacuated by motorbike to other health centers accessible by ambulance, or directly to the referral center.

The study focused on patients admitted to the gynecology and obstetrics department of the ORTH and who presented an obstetric emergency.

• Inclusion criteria: we included in our study, patients in the gravido-puerperal period with an obstetric emergency and admitted to the gynecology and obstetrics department of the ORTH and who consented to participate to the study.

• Non-inclusion criteria: patients with a pregnancy of less than 28 weeks were not included in our study.

• Sampling: all patients meeting the study inclusion criteria were conducted.

The data was collected prospectively, from June 10 to October 10, 2020, meaning a period of 4 months. They were collected through individual patient interviews, and a review of patient records. The following variables were collected:

• Socio-demographic characteristics: age, marital status, education, and occupation.

• The area of residence.

• Clinical data: reason of admission, personal history of gynecology and obstetrics, diagnoses.

• Management and evolution: the route of delivery, the condition of the mother and that of the newborn at the discharge of the hospital.

The data collected was analyzed using the R software. The chi-squared test and, where applicable, the Fisher test were used for comparison between qualitative variables at significance level p-value less than or equal to 5%.

Obstetric emergencies are any circumstances occurring during pregnancy, childbirth and postpartum that involve the vital or functional prognosis of the mother and/or fetus in the short term [1] . A normal delivery was not considered an obstetric emergency.

A precarious security zone is any locality:

• Where armed groups impose laws on the population.

• Where there have been one or more armed attacks or clashes involving armed groups.

• Where did internally displaced people leave from.

• Where travel is risky due to the presence of armed men or improvised explosive devices they may have on the roads.

• Where public services (schools, town halls...) receive threats from armed men, are closed or partially functioning because of the security risk.

A safer area: any locality that is not under the direct and imminent threat of armed groups and where people can move with a minimum sense of security. In these localities the existing public structures are fully functional.

During our study we recorded 1103 inpatients from which 426 were received for an obstetric emergency, making an incidence of 38.6%.

Of these registered obstetric emergencies, 109 were from precarious security zones, accounting for 25.6% of obstetric emergencies and 9.9% of all admissions to the service.

The mean age was 26.1 years with a standard deviation of 7.7 years. The extreme ages were 15 and 45 years. In the population of patients from more secure areas, the mean age was 26.5 years, compared to 25.7 years in safer areas (p = 0.15).

Table 1 shows the distribution of patients according to other socio-demographic characteristics and by the two groups.

The patient’s history, as well as their prenatal care status, is summarized in Table 2.

Methods and indications of admission

The patients who came on their own were 45 meaning 10.59% of the total. In the population of patients from precarious security areas, direct admissions accounted for 3.67%, compared to 12.93% among patients from safer areas. Evacuations and referrals accounted for 89.41% of the total population, 87.07% in patients from safer areas and 96.33% in those from precarious areas (p < 0.01). Table 3 shows the distribution of patients according to their reasons for admission.

Condition of newborn and fetuses at admission

At admission, 350 patients were still pregnant, with twin pregnancies in 20 of them, including 16 patients from safer areas, and 4 from precarious security

Some patients had multiple motives at once.

areas. The patients who arrived in the immediate postpartum period were 76 including a patient who gave birth to twins. The condition of fetuses and newborns is summarized in Table 4.

Diagnostic

The diagnoses selected after the examination of patients received in emergency are summarized in Table 5.

Depending on the pathologies presented by the patients, the management used different means including obstetric means. A total of 408 patients making 95.77% of the total population gave birth after their symptoms. Table 6 presents the distribution of patients by mode of delivery and area of residence of parturients.

In mothers

During our study, 6 maternal deaths (1.41% of the total population) were recorded: 2 among patients from insecure areas and 4 from safer areas (p = 0.65). The causes were: hemorrhage, severe malaria, severe anemia.

Perinatal condition

A total of 408 parturients gave birth to 429 newborns, including 42 twins (21 twin pregnancies). The number of newborns born from mothers from precarious

security zones was 109, including 10 twins (5 twin pregnancies), and the one from mothers from safer areas was 320, including 32 twins (16 twin pregnancies). The status of newborns at birth and in the first week of life is shown in Table 7.

About a quarter (25.6%) of obstetric emergencies originated in precarious security zones during the study period. These areas corresponded to the health areas of health facilities closed or functioning at a minimum, but also to other localities that had experienced security incidents and were still fully functional health facilities. These security incidents result from the presence of mobile unidentified armed groups that travel through different localities and are responsible of targeted armed attacks, kidnappings of civilians, expulsion of civilian populations from their localities, planting of improvised explosive devices on roads [8] [9] [10] [11] .

The study population is characterized by a similar average age in both comparison groups. Residence in rural areas was significantly more common among women from insecure areas. Indeed, urban areas were relatively spared from armed attacks. Women from insecure areas residing in urban areas represent women who have recently moved from their villages to the city of Ouahigouya due to insecurity. They lived in this city in precarious conditions. Professional status was marked by a large proportion of unpaid occupations in the general population. This category of unpaid occupation was significantly more represented

in the population of women from precarious security zones. This difference could be explained by the fact that the socio-economic tissue is often disorganized in areas under the control of armed groups, due to markets and businesses ransacked by armed groups, difficulties in supplying equipment and various foodstuffs, the prohibition of agricultural activities or the raiding of livestock and property by armed men [12] [13] . In addition, the latter often enact laws with religious and cultural overtones that tend to confine women to their traditional roles as housewives without income-generating activities. This contributes to making women more vulnerable.

Obstetric history was marked by a greater number of large multigravidae (22.02%) and large multiparous (10.09%) in the group of women from precarious security zones. High multigravidae and high multiparity are, however, recognized as maternal-fetal risk factors. They are due to the fact that a contraceptive prevalence of 28.6% is still insufficient in the northern region, with disparities between rural and urban areas [14] . Rural areas most often have a lower prevalence rate. The majority of patients in the two groups we identified came from the rural areas, which are also the most affected by the phenomenon of insecurity. Discourses and prohibitions based on religious and cultural considerations and ideologies specific to armed groups in their areas of influence would reduce the demand for modern methods of contraception as also noted by Degni et al. in northern Mali [5] . For their part, providers of health facilities that are still functional may be reluctant to promote family planning methods to avoid being singled out. For example, the advanced strategy of health workers bringing promotional care activities such as contraceptive supplies and community-based distribution closer to their targets has been abandoned in areas with high security challenges [7] , which could increase unmet need for family planning. Finally, the closure of health facilities has deprived women of family planning services.

Pregnancy monitoring was not optimal for 25.53% of the population with a significantly higher proportion in the group from precarious security zones. The closure or minimal operation of certain peripheral health facilities due to insecurity has been a factor. In addition, the security risks during travel [7] [13] due to armed men who are mobile and because of the risks associated with improvised explosive devices were well-founded and would discourage the use of remote health centers for pregnancy monitoring. The context of reduced income [13] due to the slowdown in income-generating activities could also be a cause of the large proportion of suboptimal follow-up of pregnancies. Indeed, even if antenatal consultations are free, it takes logistical means (motorcycle or other vehicle, fuel) to go to a health facility for this purpose. Yet quality antenatal care is an important pillar in the fight against maternal and perinatal morbidity and mortality [15] .

Most patients (89.44%) were evacuated or referred from peripheral health centers to the Regional University Hospital of 0 uahigouya which is the referral center of last resort in the Northern region. Patients are coming directly from safer areas more frequently than from precarious security zones (p < 0.01). In these precarious security zones, pregnant women must travel to other health centers further away when health centers in their localities are closed or inaccessible. Otherwise, when it comes to eutocic delivery, they are helped in an inadequate environment, by community health workers who have all received some training as part of a national plan for the resilience of health structures [7] . When the use of a referral center for surgery or blood transfusion is necessary, the journey to reach these referral centers is often long and perilous, as other authors have described in other circumstances of armed conflict [5] [16] . However, it is recognized that in order to reduce maternal and perinatal morbidity and mortality, the assistance of a qualified provider and an adequate referral system are necessary [17] [18] .

Clinically, the grounds for admission to the obstetrics and gynecology department of the ORTH were dominated by dystociclabour, preeclampsia/eclampsia, acute fetal distress, and postpartum hemorrhage. No motive was significantly more common in either comparison group.

The diagnoses made at admission to the ORTH were dominated by those related to direct causes of maternal and perinatal mortality, including dystocia, preeclampsia/eclampsia, acute fetal distress, postpartum hemorrhage. Only uterine rupture was significantly more common among women in precarious security areas (p = 0.02).

After their symptoms, most patients (95.77%) gave birth. The route of delivery was the caesarian section in more than half of the cases in the total population. When comparing the two groups, there were significantly more caesarean sections in the group of women coming from safer areas. Indeed, these women arrived most often on time, and the indication of an emergency caesarean section kept all its meaning because the fetus was still most often alive, and it was necessary to extract it urgently to save it, or because a maternal or adnexal reason for caesarean section was still present. On the contrary, women from precarious security zones arrived later, as ambulances were often targets of armed groups [7] [11] , so they could not arrive in these areas. Patients are then forced either to come using unsuitable means (motorcycles, tricycles) or to go first to a health center where ambulances can arrive with a minimum of security. This long journey and inadequate means of transport are the cause of a long delay in evacuation to the reference center, and also the cause of postures (most often sitting on a motorcycle or in an undeveloped tricycle). The consequences were uterine ruptures, deaths in utero, delivery along the way without assistance for mothers and newborns. Thus, on arrival, the indication for an emergency caesarean section often became obsolete because of these consequences.

The result of this situation is stillbirth and higher perinatal mortality among patients from precarious security zones. Apart from the known causes of stillbirths (maternal, adnexal, dystocia and other pathologies) [15] [19] , some of which are found in the diagnoses used in the patients studied, insecurity related to armed conflict is a contributing factor to stillbirths. In fact, Buitrago et al. showed in their study in Colombia that a reduction in violence was associated with a reduction in stillbirth and perinatal deaths [20] .

As for maternal mortality, it was noted in about the same proportions in both groups, probably because our study did not recruit patients admitted to medical centers with surgical antennas or those who could not go to a health center. But it is recognized that timely access to comprehensive emergency obstetric and neonatal care is key to reducing maternal mortality.