To our knowledge, this is the first community-based study using a reproductive age mortality survey (RAMOS) in any part of Sudan on the state level. The maternal mortality ratio was found to be 714\100,000 live births. The proportion of maternal deaths among female deaths (PMDF) in our study was 43.3%, which is high compared to the reported national estimate of 19.4% . This high rate of maternal mortality reflects poor maternity services. We found a wide discrepancy between urban and rural areas (369 & 872/100,000 LB, respectively). Even in urban areas, sectoral differences in maternal mortality were reported  in which slum dwellers and internally displaced people camps around cities had a high maternal mortality ratio compared to their neighboring inhabitants. More than 75% of deaths occurred during childbirth and postpartum, which is consistent with the pattern of causes in Sub-Saharan Africa. For example, in Eritrea, Ghebrehiwot  found that 16% of maternal deaths occurred during pregnancy, 48% occurred during childbirth, and 36% occurred postpartum.
Indirect causes account for 20 to 25% of maternal deaths and are attributable to illnesses aggravated by pregnancy  such as anemia, malaria and acquired immune deficiency syndrome (HIV/AIDS). In our study, indirect causes constituted more than 40% of deaths, with severe anemia as the major cause. Anemia is highly prevalent in Africa, with up to three-fifths of pregnant women having some degree of anemia and approximately one-third classified as having severe anemia [14–17]. Anemia is common in Eastern Sudan due to malaria, chronic illnesses and poverty  and is the main risk factor for stillbirth in maternity hospitals in Kassala . Women, especially those in rural areas in Kassala State, enter pregnancy in a state of nutritional deficit and therefore are unprepared to cope with the extra physiological demands of pregnancy. Mason et al. found anemia in 45% of pregnant women and 49% of children under age 5 in developing countries . Anemia is also an underlying risk factor for 18% of maternal and 24% of perinatal mortality [21, 22].
The direct obstetrical causes of death in our study, obstetric hemorrhage, obstructed labor, sepsis and hypertensive disease with pregnancy, were similar to those of developing countries . Abortion accounts for a small percentage of the deaths in this study. This is primarily because the respondents have difficulty recognizing the early menstrual history of the deceased. Also induced abortion for unwanted and out of wedlock pregnancy were usually not revealed due to influence of culture and abortion law.
The most common cause of maternal death was bleeding, which can kill even a healthy woman within two hours if unattended. Anemia in pregnant women reduces a woman's ability to survive bleeding during and after childbirth. In this study, hemorrhage is the second most common cause of death. These women likely faced bleeding in a state of anemia.
The "Three Delays" model , which includes delays in the decision to seek care, delays in reaching care, and delays in receiving care at the facility, were evident in our study. Although problem recognition was high, we found that the delay in seeking care was also high. In this study, 67.2% of deaths occurred at home. These patients were either unable or did not want to reach the health facility.
During the period of the study, the state was affected by war, and roads were closed by night. These factors likely contributed to transportation problems, which were found in 54.7% of the cases.
Delay in receiving medical care at health facilities, primarily due to lack of emergency obstetric care (Emoc) in almost all the health facilities in the rural settings, was found in 68.8% of cases.
Domestic delivery is common in Sudan, which was estimated at 76.5% and 82% nationally and in Kassala State, respectively . In rural settings in Kassala State, almost all deliveries occur at home, and patients seek medical care if life-threatening complications arise. More than half of those who died after delivery were delivered by non-skilled birth attendants. Sudan Household Health Survey (SHHS) showed that 49.2% of births in the two years prior to the survey were delivered by qualified health personnel. The majority of the qualified health personnel are village midwives, who are on the margin of the health system in Sudan. They have no fixed jobs in the health system and depend on the incentives given by parturient mothers. Inability of health facilities to deliver effective Emoc led the community to lose trust in these hospitals, and these health facilities became a distinct cause of delay. Other interactive factors such as poverty, illiteracy and transportation problems were contributed to deaths in this study.
This high percentage of maternal deaths could be effectively reduced by improving the availability and use of Emoc in all health facilities. There is a need to expand midwifery coverage by availability of a certified midwife in every village. We recommend expansion of midwifery training by opening midwifery schools in remote areas. Establishing maternity waiting home near the tertiary hospitals will enable patients from rural areas with obstetric complications to stay in the town and avoid long cost hospitalization. To overcome transportation problems, rural hospital needs to be equipped with ambulances. Furthermore, improving non-health sector factors such as poverty, female education and infrastructure is important to reduce maternal mortality in the state.