Determinants of Second-Trimester Safe Termination of Pregnancy in Public Health Facilities of Amhara Region, Northwest Ethiopia: An Unmatched Case-Control Study

Background . Second-trimester medical abortion is the termination of pregnancy between 13 and 28 weeks of gestational age. Although the majority of abortions are performed in the ﬁrst trimester, 10–15% of terminations of pregnancies have taken place in the second trimester globally. Objective . To identify the determinant factors of second-trimester safe termination of pregnancy in public health facilities of the Amhara region, northwest Ethiopia. Methods . An institution-based unmatched retrospective case-control study conducted from 01/10/2019–30/02/2020. A systematic random sampling technique was used to select 119 cases and 238 controls. An interviewer-administered questionnaire was used to collect the data. A binary logistic regression model was ﬁtted to identify determinant factors. The odds ratio with 95% CI was computed to assess the strength and signiﬁcance of the association between dependent and independent variables. Result . Rural resident (adjusted odds ratio (AOR) � 1.9; 95% CI 1.07–3.25), irregular menses (AOR � 1.8; 1.06–3.13), had no known symptoms of pregnancy (AOR � 1.9; (95% CI 1.06–3.46)), not knowing the abortion law (AOR � 3.0; (95% CI 1.63–5.60)), low level of education (1st–8th grade) (AOR � 2.7; (95% CI 1.06–6.60), opposition against abortion care (AOR � 2.6; (1.22–5.42)), delayed referral (AOR � 10.1 (95% CI 4.02–29.18)), and not undertaking pregnancy test (AOR � 2.2; (95% CI (1.21–4.04)) were determinants of second-trimester safe termination of pregnancy. Conclusion . Women being rural residents, irregular menses, not undertaking pregnancy test, not knowing the abortion law, low-level educational status, delayed referral, no knowledge about signs and symptoms of pregnancy, and opposition of safe abortion were determinants of second-trimester safe termination. The Regional Health Bureau and Health Facilities should give emphasis to women living in rural areas, and they should increase awareness towards abortion law and sign


Introduction
Abortion is the termination of pregnancy before 20 weeks of age in developed countries but up to 28 weeks of gestational age in Ethiopia with a fetus born weighing less than 500 gm by removal or expulsion of a conception tissue [1]. Unsafe abortion is a procedure for terminating unwanted pregnancy by persons lacking the necessary skills or in an environment not in conformity with minimal medical standards, or both [2]. Second-trimester medical abortion is the termination of pregnancy between 13 and 28 weeks of gestational age [3].
Globally, 42 million abortions are performed annually and 10-15% of the cases take place in the second trimester period, over half of which are considered unsafe [4]. Globally, abortion-related maternal deaths account for 13%, of which the majority is caused by unsafe abortions [5].
Ninety-two percent of African women of childbearing age live in countries with restrictive abortion laws [3]. Safe termination of pregnancy is now permitted where continuation of the pregnancy endangers the health or life of the woman or the child or where pregnancy is a result of rape or incest for up to 28 weeks of gestational age in Ethiopia [6].
Globally, from 2010-2014, about 55.9 million abortions occur each year of which 49.3 million are in developing countries [7]. e burden of abortion is still high, being one of the leading causes of maternal mortality and morbidity and millions are injured [8]. About 22 million women performed unsafe abortion every year [9]. Second-trimester abortion is associated with a higher risk of complication compared with first-trimester abortion. It may cause cervical laceration, cervical perforation, heavy bleeding, infection, and uterine perforation which may require hysterectomy [10].
Evidence showed that the prevalence of induced secondtrimester abortion was 11% in Ethiopia. e majority (97%) of these abortions are unsafe [11]. From 2010-2013, a total of 7484 women accessed second-trimester abortion, and over fifty percent of women presenting in the second trimester had incomplete abortion in Ethiopia [12].
Maternal mortality is still high in Ethiopia and the five major causes of maternal mortality (19.6%) occur due to abortion complications [13]. Between 2008 and 2014, the percentage of abortions performed in health facilities increased from 27-53% but still 294,100 abortions occurred outside of health facilities every year [14].
In Ethiopia, 50% of women presenting in the secondtrimester were suffering from incomplete abortion [15]. In the Amhara region, an estimated 45,812 women performed abortion in health facilities in 2014, while around 106,325 women terminated their pregnancy outside health facilities in the same year. In the study area, the prevalence of secondtrimester abortion was 19.2% [16]. erefore, the objective of this study was to assess determinant factors contributing to second-trimester safe termination of pregnancy in public health facilities, Amhara region, northwest Ethiopia.

Study Area and Design.
A facility-based unmatched retrospective case-control study design was conducted in Felege Hiwot Comprehensive Specialized Hospital (FHCSH) in Bahir Dar city, Gondar University Comprehensive Specialized Hospital (GUCSH) in Gondar town and Debre Markos Referral Hospital (DMRH) in Debre Markos town of Amhara region from November, 2019, to February, 2020. FHCSH is found 560 km northwest of Addis Ababa, capital city of Ethiopia. It provides health service to more than 5 million people and serves on average 120 safe abortion care per month. GUCSH is located in North Gondar Administrative Zone, 727 km far from Addis Ababa. It is a teaching hospital and serving a catchment area of approximately 7 million people and on average for 350 (110, 2nd trimester) and (240, 1st trimester) abortion care service per month. DMRH serves more than five million people and on average 105 safe terminations of pregnancy service per month of the East Gojjam Zone, located 295 kilometers from Addis Ababa.

Study Participants.
All women who came for safe abortion care in Felege Hiwot Comprehensive Specialized Hospital (FHCSH), Gondar University Comprehensive Specialized Hospital (GUCSH), and Debre Markos Referral Hospital (DMRH) during November 2019, to February 2020, were included in the study.
All women who came in the second trimester (13-28 wks) for safe abortion care were included as cases and women who came in the first trimester (up to 13 weeks) for safe abortion care were included as controls. ese participants were interviewed when they came for abortion care at health facilities.

Sample Size and Procedure.
e calculated sample size including 10% nonresponse rate was 357 (cases � 119 and controls � 238) with 1 : 2 case-to-control ratio and using 80% power to detect an odds ratio of 2.02. Odds ratio and percent of prevalence were taken from the associated factors of the previous study [6]. Participants were allocated proportionally in the three hospitals based on the average monthly performance of first and second-trimester abortion care service. A systematic random sampling method was used to select study participants. e sample was taken every two intervals for controls and consecutively for cases.

Data Collection Procedure and Quality Assurance.
A structured interviewer-administered questionnaire developed from a review of relevant literature was utilized to collect data on sociodemographic characteristics, institutional factors, community factors, partner, and obstetric and personal factors adapted from other literatures [6,16]. e questionnaire was prepared in English and translated to the local language (Amharic) and translated back to English by another individual to check the consistency. e questionnaire was pretested on 5% cases and controls of the related population before the actual data collection period. e data were collected by three midwives who have a B.Sc. degree in midwifery and a public health officer as a supervisor.

Data Analysis.
e data were checked, coded, and entered using Epi-Data software version 3.1 (Epi-Data Association, Odense, Denmark) and exported to SPSS software Version 23 for further analysis. e outcome variable was dichotomized into 1 � cases and 0 � controls. e descriptive result was presented using frequency tables and proportions for all variables. To identify the independent predictors of delayed safe termination of pregnancy, multivariable logistic regressions were performed and variables with p value <0.05 were considered as statistically significant predictors of outcome variable. e odds ratio with 95% CI was computed to assess the strength and significance of the association between dependent and independent variables.

Sociodemographic Characteristics.
A total of 347 (110 cases and 237 controls) participants were included in this study, with a response rate of 97.2%. Among these, 44 (40%) of cases and 161 (67.9%) of controls were urban residents. Seventy-two (32.3%) of the controls and 50 (43.4%) of the cases were less than 19 years old. Regarding the educational status, 64 (58.2%) of the cases and 87 (36.7%) of the controls were less than primary school (Table 1).

Maternal and Obstetric Characteristics of the Women.
Forty (43.5%) cases and 52 (56.5%) controls had not checked for pregnancy tests during the first trimester period. Twentyfive (22.7%) cases and 116 (82.3%) controls knew the law of safe termination of pregnancy ( Table 2).

Determinants of Second-Trimester Safe Termination of Pregnancy.
We performed multivariable logistic regression analysis to identify determinant factors of second-trimester safe termination of pregnancy. After controlling for possible confounding factors, rural residents, irregular menses, did not know signs and symptoms of pregnancy, did not know abortion law, and low level of education status were significant determinant factors of second-trimester safe termination of pregnancy. Participants who were rural residents were nearly two times more likely to delay for safe termination of pregnancy as compared to those in urban [AOR � 1.87, 95% CI (1.072-3.247)]. Participants who had irregular menses were nearly two times more likely to delay for safe termination than those who had regular menses [AOR � 1.82, 95% CI (1.06-3.125)] and participants who did not know the sign and symptoms of pregnancy were nearly two times more likely to delay for safe termination than who knew [AOR � 1.91, 95% CI (1.056-3.459)]. Participants who did not know abortion law were three times more likely to delay for safe termination of pregnancy than those who knew [AOR � 3.02, 95% CI (1.631-5.595). Participants who had a low level of educational status (primary school) were nearly three times more likely to delay for safe termination than participants who had a diploma and above [

Discussion
is study provides additional insights into determinants of second-trimester safe termination of pregnancy in the Amhara Region, northwest Ethiopia. In this study, about 45% of women conducted second-trimester pregnancy termination at the age of 15-19 years. e reason for the second-trimester termination of pregnancy at this age could be fear of abortion at an early age. Women who were living in rural areas were nearly two times more likely to be delayed for safe termination of pregnancy as compared to those who were living in urban. is result is supported by studies done in Ethiopia [6,16,17] and India [18]. Inaccessibility to health facilities in rural areas and risk perception of participants to abortion could be the main justifications for second-trimester safe termination. In this study, second-trimester safe termination of pregnancy was more likely in those who had low level of educational status (primary School) than diploma and above. is result is in line with that of a study conducted in Ethiopia [6,16], Kenya [19], Zambia [20], and Burkina Faso [21]. e reason could be, as educational level increases, the knowledge towards safe abortion also increases [17].
is study revealed that second-trimester safe termination of pregnancy was more likely in women who had irregular menses than those who had regular menses supported by studies conducted in Ethiopia, Canada, and Vietnam [16,22,23]. Irregular menses may confuse women and make them late for second-trimester safe termination of pregnancy. But, a study done in a Zambian teaching hospital [20] showed irregular menses had no association with second-trimester safe termination. e other important variable of this study was women who did not know the law of abortion were significantly higher for second-trimester safe termination of pregnancy than those who knew. is result is supported by studies conducted in Harar and Oromia regions of Ethiopia [2,24]. But, a study done in South Africa does not collaborate with this study [25]. is might be due to the difference in the awareness level of women towards abortion law between countries. In addition, participants who did not know the signs and symptoms of pregnancy were significantly higher in the second-trimester safe termination.
is finding is consistent with a study in South Africa [25]. Lack of awareness towards pregnancy signs and symptoms may lead to late seeking termination of pregnancy.
In this study, women who delayed testing for pregnancy were more likely have second-trimester safe termination of pregnancy and it is consistent with a study done in Canada [22]. e reason could be that women may not know the presence of pregnancy tests. is reason is also mentioned in a study done in Britain [26].
In addition, this study also showed that participants who faced opposition for abortion care from families were significantly higher for second-trimester safe termination of pregnancy. is study also agreed with studies done in Brazil and Canada [22,27]. But, this finding was not in line with a study done in Zambia [20]. e possible reason for this difference might be due to educational status differences, behavior, and lifestyle of the society between countries.

Conclusion
is study found many determinants that contribute to second-trimester safe termination of pregnancy in Amhara region referral hospitals. Women who did not understand the symptoms of their pregnancy and did not know the law of abortion, being rural resident, delay referral, low level of educational status (primary education), irregular menses, delay testing for pregnancy, and discrimination (opposition) against abortion care service were determinant factors of second-trimester safe termination of pregnancy.
Amhara Regional Health Bureau and Hospitals should give emphasis to women living in rural areas to easily access safe termination of pregnancy. ey should increase awareness towards abortion law and about signs and symptoms of pregnancy and irregular menses. ey should work on discrimination (opposition) of safe abortion and avert this abnormal behavior. Furthermore, to explore determinant factors of second-trimester safe termination of pregnancy, more qualitative research is needed. Data Availability e datasets supporting the conclusions of this article are included within the article.

Ethical Approval
Ethical clearance was obtained from the Institutional Review Board of the College of Medicine and Health Sciences, Bahir Dar University. An official support letter was obtained from the Amhara Public Health Institute.

Consent
Written consent was obtained from the mothers/caregivers after informing all the purpose, benefits, and risks of the

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Authors' Contributions
TM, YW, and AA contributed to the proposal development, data collection, analysis, and result interpretation. YW, SM, YM, and NF contributed to the conceptualization and writing of the paper and edited the overall improvement of the manuscript. All authors read and approved the final submitted paper.