Surgical Task Shifting Helps Reduce Neonatal Mortality in Ethiopia: A Retrospective Cohort Study

Background To improve access to surgical service and to reduce neonatal mortality Ethiopia implemented surgical task shifting to nonphysician surgeons (NPSs). We aim at assessing surgical outcomes between NPSs and physician surgeons working in two hospitals. Methods We collected data from two hospitals on 474 maternal medical records. Completed maternal medical records were included in this study. Data were entered into Epi Info version 7 and analyzed by SPSS version 20 software. Both descriptive and inferential statistics were done. If the 95% confidence interval values exclude the null value, the factor was considered as a significant factor. Result Totally, 3429 mothers delivered in two hospitals. Of them, 840 (24.5%) delivered by caesarian section (CS), but 474 mothers' records meet the inclusion criteria included in this study. Of 474 CS deliveries, the majority (82%) of them were performed by NPS. Maternal or fetal emergency conditions were the main reasons (92.0%, n=436) for CS. Task shifting does not affect immediate newborn outcomes (ARR, 1.24 (0.55, 2.78)), but duration of hospitalization (ARR: 4 (2.3, 7.5)), condition of the fetus during admission (ARR: 5.22 (2.9, 9.2)), and type of anesthesia used (ARR: 0.2 (0.1, 0.4)) significantly determine the outcome. Conclusion Surgical task shifting to NPS does not affect the immediate newborn outcome. But general anesthesia is one of the major factors that affects the outcome.


Introduction
Surgical care task shifting to nonphysicians is a common practice in most developing countries [1]. Ethiopia implemented task shifting to NPSs, (they are BSc health officers or BSc nurses plus a 4-year training on integrated emergency surgery and obstetrics) to overcome the dire shortage of physicians, to improve access to medical and surgical care service, and to reduce maternal and neonatal mortality [2][3][4][5][6].
Neonatal mortality is one of the major health problems of Ethiopia. Surgical task shifting is an intervention supposed to reduce the rate of neonatal mortality [7][8][9]. Emergency Obstetric and Newborn Care (EmONC) is another strategy to reduce neonatal mortality [10].
EmONC is best implemented in developing countries through surgical task shifting.
ere is evidence from different countries on significant reduction of neonatal mortality due to EmONC implementation [7,11]. is is because surgical intervention of EmONC creates a window to save the life of both mother and fetus due to emergency conditions during pregnancy. An emergency condition is a major reason for CS [12][13][14][15][16][17][18].
Different scholars reveal that there is an improvement on access to CS service and reduction of neonatal mortality after implementation of surgical task shifting [1,[12][13][14]19]. ere are findings from Ethiopia that assess the quality of CS service and its rate; but this research assesses the effect of surgical task shifting on neonatal birth outcomes [12,15,16,20,21].

Methods and Materials
A retrospective cohort study design was used to assess the rate of CS delivery and the relationship between NPS and CS outcomes. Data were collected from July 1 to 30, 2017.
e study was conducted in Arba Minch General Hospital and Sawla District Hospital. ese hospitals were selected because they qualify Comprehensive EmONC standard [20].
Mothers who gave birth by CS from July 2015 to June 2016 and all those who meet inclusion criteria (fully recorded medical record, mothers who give birth after arrival to the hospital, and both the mother and fetus are alive at the time of arrival to the hospitals) were included in this study. Data were collected by data abstraction tools adopted from averting maternal death and disability program module [10]. We collected data on the condition of mother and fetus during admission, who conducts CS, different tasks done during the procedure, and status of the newborn until the time of discharge. Experienced midwives and anesthetists were used to collect data after attending one-day training on how to collect the data. Ethical clearance was obtained from the South Region Health Research Ethics Review Committee.
Data were entered into Epi Info version 7 and analyzed using SPSS version 20. We did both descriptive and analytic statistics. Initially, binary logistic regression was performed to assess the association between each exposure and outcome variables. During bivariate analysis, variables with P value 0.2 were considered for multivariate binary logistic regression. When 95% confidence interval values exclude null value the factors considered as a significant factor. e immediate newborn outcome was categorized into good and bad. e good outcome means when the newborn is alive during birth without distress, and bad outcome means when there is a prenatal death or lives birth with distress.

Participants in the Study.
Totally, 3429 deliveries were attended in two hospitals, and of them 840 (24.5%) were delivered by CS. But only 474 mothers who fulfilled the inclusion criteria were included in this study ( Figure 1).

Characteristics of Mothers.
From the total population, the majority (58%) of mothers were in the age group between 15 and 25: 189 (39.5%) mothers were nulliparous (not given birth before), and 64 (13.5%) mothers have more than 5 children. Greater than 60% of mothers were rural by residence, and 222 (48.2%) mothers come to hospital referred from rural health centers. At the time of admission, 159 (33.5%) mothers and 157 (33.1%) fetuses had an unstable vital sign. Only 152 (32.1%) mothers' medical registrations have data on the time interval from decision to perform CS to actual incision time.
Often, spinal anesthesia and lower uterine segment transverse CS were used in 379 (79.1%) and 464 (96.9%) mothers, respectively. e majority (82%) of procedures were performed by nonphysician surgeons. e average time of the procedure was 51 (SD 18.62) minutes ( Figure 3).

Discussion.
is study reveals that the rate of CS, the rate of bad CS birth outcome, major indications for CS, the effect of surgical task shifting, and factors determine CS birth outcome. Incompleteness of maternal data registration imposes gap in the study, but strong association between factors and CS strengthens our findings. e rate of CS delivery (24.5%) obtained in this study is greater than the normal range set by WHO, which is between 5 and 15% and rates reported nationally and abroad [10,12,[20][21][22][23]. Similar to different findings, high emergency conditions (92%) are the major reasons for CS delivery 2 Surgery Research and Practice [13,15,22]. is shows that surgical task shifting is playing a great role in Ethiopia to reduce maternal and fetal mortality because the commonest emergency conditions, such as CPD, fetal distress, malpresentation, and antipartum hemorrhage, similar to findings from different countries [12,13,15,17,18], can lead to fetal or maternal mortality if they are not managed through surgical interventions. However, in order to reduce risk of surgery, the increasing rate of CS delivery needs deep investigation. In Ethiopia and other countries where both surgeons and NPSs are active, NPSs are performing majority (82%) of CS procedures, [17,19,[24][25][26]. is means NPSs contribute in saving the lives of mothers and fetuses during emergency conditions. Moreover, they make CS service more accessible to people in need. However, out of 73 (15%) bad CS birth 21

Conclusion
NPSs are dominant hand for saving the life of children of Ethiopia. Using spinal anesthesia and short-time hospitalization will improve CS birth outcomes.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that there are no conflicts of interest.

Authors' Contributions
YT designed the study, supervised data collection, analyzed the data, and drafted the manuscript. TG participated in the design of the study and supervised data collection. MM reviewed the manuscript. HT supervised data collection and critically reviewed the manuscript. All authors read and approved the final manuscript.