Trends in health facility based maternal mortality in Central Region, Kenya: 2008-2012

Introduction WHO classifies Kenya as having a high maternal mortality. Regional data on maternal mortality trends is only available in selected areas. This study reviewed health facility maternal mortality trends, causes and distribution in Central Region of Kenya, 2008-2012. Methods We reviewed health records from July 2008 to June 2012. A maternal death was defined according to ICD-10 criterion. The variables reviewed included socio-demographic, obstetric characteristics, reasons for admission, causes of death and contributing factors. We estimated maternal mortality ratio for each year and overall for the four year period using a standard equation and used frequencies means/median and proportions for other descriptive variables. Results A total 421 deaths occurred among 344,191 live births; 335(80%) deaths were audited. Maternal mortality ratios were: 127/100,000 live births in 2008/09; 124/100,000 live births in 2009/2010; 129/100,000 live births in 2010/2011 and 111/100,000 live births in 2011/2012. Direct causes contributed majority of deaths (77%, n=234) including hemorrhage, infection and pre-eclampsia/eclampsia. Mean age was 30(±6) years; 147(71%) attended less than four antenatal visits and median gestation at birth was 38 weeks (IQR=9). One hundred ninety (59%) died within 24 hours after admission. There were 111(46%) caesarian births, 95(39%) skilled vaginal, 31(13%) unskilled 5(2%) vacuum deliveries and 1(<1%) destructive operation. Conclusion The region recorded an unsteady declining trend. Direct causes contributed to the majority deaths including hemorrhage, infection and pre-eclampsia/eclampsia. We recommend health education on individualized birth plan and mentorship on emergency obstetric care. Further studies are necessary to clarify and expand the findings of this study.


Introduction
The fifth millennium development goal aimed to improve maternal health with a target of reducing the maternal mortality ratio by 75% between 1990 and 2015. The total global maternal deaths declined from 523,000 in 1990 to 289,000 in 2013. The lifetime risk of dying during pregnancy and child birth in Africa is 1 in 40. The most dangerous place for a woman to have a baby is in sub-Saharan Africa [1]. According to WHO, the main causes of maternal deaths in the world include pre-existing medical conditions (diabetes, malaria, HIV, obesity) exacerbated by pregnancy, severe bleeding, pregnancy-induced high blood pressure, infections mostly after child birth, obstructed labor and other direct causes, abortion complications and blood clots [2]. In comparison, the main causes of maternal death in Africa are hemorrhage, other indirect causes, sepsis/infections, hypertensive disorders, HIV/AIDS, unclassified deaths, other direct causes, obstructed labor, abortion, anemia, embolism, and ectopic pregnancy [3]. In Kenya, Maternal Mortality Ratio (MMR) was 390 deaths per 100,000 in 1993 and increased to 488 deaths per 100,000 in 2008/09 [4][5][6][7]. Kenya is therefore classified by WHO as making 'insufficient progress' towards achievement of the MDG since the average annual decline has been less than 2% [1]. According to a report by Kenyan Ministry of Health on reviewing causes of maternal deaths from 1994-2010, the leading causes are antepartum and postpartum hemorrhage. Other common causes include eclampsia, sepsis, ruptured uterus and obstructed labor in that order [8]. There has been several interventions and strategies in Kenya's health policy since 1994 intended to reverse the deteriorating health indicators including maternal health [8,9]. In Kenya, maternal death reviews were introduced in 2004. However, the reviews faced challenges including serious underreporting of maternal deaths, incomplete data, lack of understanding of health staff on the value of reviews and lack of evidence of specific actions after review meetings [10]. Nationally, there exists a maternal and perinatal death surveillance system, which was practiced in central region in the period under review.

Methods
The study took place in Central region, one of the eight regions in Kenya ( Figure 1). The region constitutes of five Counties including Kiambu, Murang'a, Nyeri, Kirinyaga and Nyandarua. The overall total fertility rate for the region was 3.4, rate of skilled deliveries was 73.8%, family planning coverage was 66.7% and antenatal attendance was 92.7% [7]. The study population comprised of women who had given birth in health facilities in the region during the study period. We conducted a descriptive retrospective study of maternal deaths reported from the health facilities in the region between 1st July 2008 and 30th June 2012. A maternal death was defined as any death of a woman while pregnant or within 42 days of termination of pregnancy, from any cause related to or aggravated by the pregnancy or its management, but not from accidental or incidental causes (ICD-10). We retrieved data from to undertake the study was granted by the regional health management team.  (Table 1). Two hundred and thirty-four (77%) of the deaths reported were due to direct causes, while 70(23%) were due to indirect causes. The direct causes of death included hemorrhage (40%, n=135), infection (13%, n=44), preeclampsia/eclampsia (12%, n=41), abortion (3%, n=10), amniotic embolism (3%, n=9) and disseminated intravascular coagulation (3%, n=9). The indirect causes included HIV/AIDS (5%, n=20), anemia (4%, n=12), cardiovascular disorders (4%, n=12), pulmonary embolism or embolism (3%, n=9), anesthesia complications (3%, n=9) and tuberculosis (2%, n=8) ( Table 2).
However, in Moi teaching and referral hospital and in the Nakuru general hospital higher deaths in younger age group were documented [15,16]. Maternal deaths occurred mainly among women between 37 and 42 weeks of pregnancy. In contrast, the findings in the Moi Teaching and Referral Hospital found that majority of the women were less than 36 weeks gestation. In the same hospital most mothers had attended less than four antenatal visits [15]. More than half the women had less than three living children. The deaths occurred mainly in Kiambu County. This can be explained by the fact that 37% of the population in the region resides in the County [19]. In addition, it could probably mean that in comparison to other Counties in the region, the County had a higher burden of health facility-based maternal mortality. This study had several limitations. Secondary data used in the analysis, was prone to variations in data quality across facilities including incomplete entries. We also relied on the accuracy in the doctor's assessment of the cause of death and documentation of the time of death. In this study, hospital stay was measured to the nearest one hour. A date that did not have time recorded was assumed to have

Conclusion
We therefore concluded that Central region had a moderate health facility based maternal mortality with an unsteady declining trend.

Competing interests
The authors declare no competing interests.