Distance from health facility and mothers’ perception of quality related to skilled delivery service utilization in northern Ethiopia

Abstract

Background

Poor maternal health service utilization is one of the contributing factors to a high level of maternal and newborn mortality in Ethiopia. The factors associated with utilization of services are believed to differ from one context to another. We assessed the factors associated with skilled delivery service utilization in rural northern Ethiopia.

Subjects and methods

A community-based survey was conducted among mothers who gave birth in the 12 months preceding the study period, from January to February 2015, in the Tigray region of Ethiopia. Multistage sampling technique was used to select mothers from the identified clusters. Households within a 10 km radius of the health facility were taken as a cluster for a community survey. Data were collected using face-to-face interview at the household level. We compared the mothers who reported giving birth to the index child in a health facility and those who reported delivering at home, in order to identify the predictors of skilled delivery utilization. Multivariable logistic regression model was used to determine the predictors of skilled delivery service utilization. The results are presented with odds ratio (OR) and 95% confidence interval (CI).

Results

A total of 1,796 mothers participated in the study, with a 100% response rate. Distance to health facilities (adjusted odds ratio [AOR] =0.53 [95% CI: 0.39, 0.71]), perception of mothers to the availability of adequate equipment in the delivery service in their catchment area (AOR =1.5 [95% CI: 1.11, 2.13]), experiencing any complication during childbirth, using antenatal care, lower birth order and having an educated partner were the significant predictors of skilled delivery service utilization.

Conclusion

Implementing community-based intervention programs that will address the physical accessibility of delivery services, such as the ambulance service, road issues and waiting rooms, and improving quality maternity service will likely reduce the current problem.

Keywords:

• distance from health facility
• perception of quality
• skilled delivery service utilization
• Northern Ethiopia

Introduction

Pregnancy, birth and postpartum periods pose greater potential risk to maternal mortality and morbidity and a significant risk to the fetus and newborn as well.¹ The use of skilled delivery service during this period is found to be an effective intervention in attaining significant improvement in maternal and newborn health outcomes.^2,3

Globally, an estimated 303,000 maternal death occurs each year, of which 99% occurs in developing countries.⁴ Sub-Saharan Africa has the highest maternal mortality rate at 546 maternal deaths per 100,000 live births with poor progress in reduction.⁵ Ethiopia has the highest maternal mortality statistics globally (353/100,000) and is one of the 10 countries that accounted for 59% of the global maternal deaths in 2015.⁴ Overall, Ethiopia somehow shows a progress in reduction of maternal mortality, but still with high mortality and poor service utilization.^4,5

Improving the maternal health service and increasing the number of deliveries attended by skilled personnel to 80% by 2020 is one of the strategies in the health sector plan of Ethiopia.⁶ By 2016, 62% pregnant mothers visited for antenatal care and 28% of births were attended by skilled provider nationally; regional variations were very wide though, and there was a wide gap in skilled birth attendant use between urban and rural areas (from 80% to 21%).⁷ Skilled birth attendant utilization in northern Ethiopia increased from 6% in 2005 to 59% in 2016.^7,8 However, the current annual rate of improvement in skilled attendant use nationally is only 0.4%, which is the lowest among the six countries with the highest annual number of maternal deaths.⁹

The impact of mothers’ factors on skilled delivery utilization has been studied by many scholars,^8,10–14 but has not been comprehensively assessed in Ethiopia,^11,13,14 in relation to the current available theoretical models developed by Gabrysch and Campbell.^15,16 Many of the studies have not linked the perception of mothers to quality delivery service with skilled delivery use. Therefore, this study aims to assess the factors associated with skilled delivery utilization in rural parts of northern Ethiopia.

This framework stated the four factors (sociocultural, perceived benefit/need, economic and physical accessibility) that affect skilled delivery use, which are the main causes of Phase I and II delays. The framework further classified the factors as individual- and health facility-related factors. The individual-related factors include the mothers’ sociocultural factors and the perceived benefit/need of facility use in which both factors directly influence the mothers’ decision to seek care (Phase I). Economic and geographic accessibility mainly influences whether the woman actually identifies and reaches the health facility for delivery (Phase II), as shown in Figure 1.¹⁵

Figure 1 Conceptual framework of the factors affecting the use of skilled delivery service.

Notes: From the three delays for seeking care, only the first and second phases are relevant for receiving normal delivery care. Adapted from Thaddeus S, Maine D. Too far to walk: maternal mortality in context. Soc Sci Med. 1994;38(9):1091–1110.¹⁶
Abbreviation: ANC, antenatal care.

Subjects and methods

This study was part of a large community-based survey conducted among mothers who had given birth in the year preceding the survey in the Tigray region of Ethiopia. In this study, we compared mothers who reported giving birth to the index child at a health facility with a skilled birth attendant and those who reported delivering at home, in order to identify the predictors of skilled delivery utilization.

A multistage sampling technique was used. First, from the 90 Kebeles (the smallest administrative unit in Ethiopia) in the study area, 27 Kebeles were selected using lottery methods. Then, to select the study participants from the 27 rural communities linked with health facilities that provide maternity service in their catchment, we used cluster sampling techniques. All Kebeles within a radius of 10 km of each selected index health facility were taken as a cluster. A total sample size of 1,796 mothers was included in this study by including 67 mothers from each cluster. All eligible women who gave birth in the year preceding the survey were included in the study. Then mothers were classified as those who utilized skilled delivery or not, based on their report.

Data collection procedures

Data were collected through interviewer-guided survey, with mothers at their household. A structured questionnaire was adapted from the Demographic Health Survey and other similar literatures.^12,13,17 The questionnaire includes variables such as mothers’ sociodemographic and sociocultural factors, perceived need (birth order, wanted pregnancy, recent antenatal care use, history of complications during a recent pregnancy and childbirth, perceived quality care), economic accessibility (occupation of mothers and partners) and physical accessibility (distance to health facilities and ease of getting transport service to the nearest health facility). As a measure of data quality control, tools were translated into local language Tigrigna and back-translated into English by blind translators to check consistency. Pretest was also done outside the study areas and modification was made. Data were collected by 27 experienced health professionals who were recruited out of the selected community. They were trained prior to the study period. Data collection was supervised on a daily basis. Every day, the filled questionnaires were cross-checked for their completeness.

Measurements

The outcome variable of this study was skilled delivery service utilization and was coded 1 if mothers reported that they had attended delivery service by skilled health personnel (midwife, nurse or medical doctor) at the health facility for their recent birth; otherwise, it was coded 0. To assess the factors affecting the utilization of skilled delivery service, we adopted a conceptual framework from Gabrysch and Campbell,¹⁵ and we assessed each independent variable based on the framework.

Distance to nearby health facility

This was measured from the report of mothers on the walking hours to the health facilities. This was coded 1 if mothers reported the walking hours to reach the nearby health facility to be ≥30 min; otherwise, it was coded 0.

Ease of getting transport service from home to the nearest health facility

This was measured from the report of mothers. It was coded 1 if the mothers reported and assumed that getting transport service from home to the nearest health facility was easy; otherwise, it was coded 0.

Mothers’ perception about the capability of service providers in managing complicated childbirths

This was measured from the report of mothers. It was coded 1 if the mothers reported and perceived that the service providers at the nearby health facility were capable of managing complicated childbirths; otherwise, it was coded 0.

Mothers’ perception of availability of adequate equipment in delivery service

This was measured from the report of mothers. It was coded 1 if the mothers reported and assumed that there was availability of adequate equipment in the delivery rooms; otherwise, it was coded 0.

Knowledge of mothers on danger signs during pregnancy and childbirth

This was coded 1 if the mothers reported they knew at least one danger sign during pregnancy and childbirth (eg, bleeding, fever, foul vaginal discharge, headache, vision problem, prolonged labor, convulsion); otherwise, it was coded 0.

Complications during childbirth

This was coded 1 if the mothers reported at least one complication during recent childbirth that occurred to themselves or newborns (eg, bleeding, prolonged labor >12 hours, leak of amniotic fluids before labor started, vaginal discharge, fever, retained placenta, convulsion, severe headache, epigastric pain, slow fetal movement, fetal distress); otherwise, it was coded 0.

Wealth index

This was measured using ownership of household assets: infrastructure (eg, homes and land ownership for agriculture, electricity), small equipment (eg, telephone, TV, radio) and domestic animals (cows/oxen, mule/donkey, sheep/goat). Items were coded into a relative index of household wealth using principal component analysis and calibrated into quintiles, with each representing 20% of the score, from 1 (very poor) to 5 (very wealthy) quintiles. Mothers were ranked by wealth quintiles using this index.

Statistical analysis

All the questionnaires were checked visually, coded and entered into EpiData version 3.1, and exported to Stata version 12 for analysis. Then, the data were cleaned to check for errors, and missed values and any error identified were corrected. Descriptive statistics were used to calculate the frequency distribution and proportions for categorical variables. For the normally distributed continuous variables, mean with SD was also used. Data related to the wealth of the mothers were assessed using household assets and were measured using principal component analysis. The bivariate logistic regression model was used to assess the factors associated with skilled delivery service utilization. Variables with a p-value of <0.25 were entered into multivariable model. A p-value of <0.05 was considered as the cutoff point for statistical significance. The degree of association between dependent and independent variables was assessed using odds ratio with 95% CI. Hosmer–Lemeshow test was used to compare and rule out the goodness of fit of the models. Multicollinearity was also examined, and all covariates having a value of variance inflation factor of <10 were tolerated.¹⁸ The results are presented in the form of tables and text.

Ethical considerations

The study was approved by Mekelle University Institutional Review Board (ERC0427/2014), and a formal letter of permission was obtained from the Tigray Regional Health Bureau. Informed verbal consent was obtained from each mother.

Results

Sociodemographic characteristics

A total of 1,796 mothers participated in the study. There was no refusal to participate in the study. The average age of mothers was 27 (SD ±6) years. Most of the mothers were married (1,578 [87.9%]) and Orthodox Christian followers (1,620 [90.2%]). More than half of the mothers did not attend formal school (824 [45.9%]) and were housewives (717 [39.9%]) and farmers (617 [34.4%]). Regarding the wealth index, most of the mothers were categorized as poor (731 [40.7%]), as shown in Table 1.

Table 1 Sociodemographic characteristics of mothers who participated in the study in northern Ethiopia, 2015

Sociodemographic characteristics	Frequency	Percentage
Age (years)
≤20	225	12.5
21–29	928	51.7
30–38	571	31.8
≥39	72	4.0
Marital status
Single	66	3.6
Married	1,578	87.9
Widowed	21	1.2
Divorced	73	4.1
Separated	58	3.2
Religion
Orthodox Christian	1,620	90.2
Muslim	176	9.8
Educational status of mothers
No schooling	824	45.9
Junior (1–4 grades)	236	13.1
Elementary (5–8 grades)	392	21.8
Secondary (9–12 grades)	281	15.7
Diploma and above	63	3.5
Occupational status of mothers
Housewife	717	39.9
Merchant	314	17.5
Employed	148	8.2
Farmer	617	34.4
Wealth index
Very poor	360	20.0
Poor	371	20.7
Medium	347	19.3
Wealthy	359	19.0
Very wealthy	359	19.0

Mothers’ perception to quality delivery service and accessibility

Accessibility to health facilities and the perceived quality were associated with the use of facilities delivery services. About half, 897 (49.9%), of the mothers reported that the distance from home to facility took >30 min on foot. About 644 (35.9%) mothers reported that it was difficult to get the transportation service from home to the nearby health facility. About 741 (41.3%) mothers perceived that the service providers at the nearby health facility were not capable of solving birth-related complications. More than half, 940 (52.3%), of the mothers perceived that there was inadequate equipment for delivery service at the nearby health facility (Table 2).

Table 2 Perception of mothers toward the accessibility and quality of maternity services in their catchment area in northern Ethiopia, 2015

Perception of mothers	Total (%)	Skilled delivery use		p-value
		Yes (%)	No (%)
Distance to the nearest health facility				<0.001
<30 min walking time	899 (50.1)	768 (85.4)	131 (14.6)
≥30 min walking time	897 (49.9)	643 (71.7)	254 (28.3)
Easy transport service from home to the nearest health facility				0.009
Yes	1,152 (64.1)	484 (75.2)	160 (24.8)
No	644 (35.9)	927 (80.5)	225 (19.5)
Service providers are capable of handling birth-related problems				<0.001
Yes	1,055 (58.7)	873 (82.7)	182 (17.3)
No	741 (41.3)	538 (70.1)	203 (29.9)
There is availability of adequate equipment in delivery service at the nearest facility				<0.001
Yes	856 (47.7)	720 (84.1)	136 (15.9)
No	940 (52.3)	691 (73.5)	249 (26.5)

Skilled delivery service utilization

In this study, 1,411 (78.6% [95% CI: 0.77–0.80]) mothers’ recent delivery was assisted by skilled birth attendants. About 659 (36.7%) mothers had at least four antenatal care visits, while majority of the mothers visited only once (1,581 [88%]). Almost all mothers did not attend postnatal care during the first week of birth (1,735 [96.6%]).

Factors associated with skilled delivery service utilization

After controlling for confounders, those living far from health facility were less likely to utilize skilled delivery services (adjusted odds ratio [AOR] =0.53 [95% CI: 0.39, 0.71]) and those who perceived facilities as having adequate equipment in the delivery were more likely to utilize skilled delivery services (AOR =1.5 [95% CI: 1.11, 2.13]). In addition, antenatal care use (AOR =0.17 [95% CI: 0.12, 0.25]), birth order (AOR =0.29 [95% CI: 0.16, 0.51]), experiencing of any complications during childbirth in the recent pregnancy (AOR =1.98 [95% CI: 1.44, 2.71]) and having an educated partner (AOR =2.15 [95% CI: 1.31, 3.52]) were the significant predictors of skilled delivery service utilization (Table 3).

Table 3 Factors associated with skilled delivery service utilization in northern Ethiopia, 2015

Variables	Skilled delivery utilization		COR (95% CI)	AOR (95% CI)
	Yes (%)	No (%)
Age of the mother, years
≤20	203 (90.2)	22 (9.8)	1	1
21–29	771 (83.1)	157 (16.9)	0.53 (0.33, 0.85)a	0.94 (0.51, 1.71)
30–38	384 (67.3)	187 (32.7)	0.22 (0.14, 0.36)b	0.65 (0.32, 1.28)
≥39	53 (73.6)	19 (26.4)	0.30 (0.15, 0.60)b	0.93 (0.38, 2.26)
Marital status
Single	55 (83.3)	11 (16.7)	1	1
Married	1,243 (78.8)	335 (21.2)	0.74 (0.38, 1.43)	0.68 (0.33, 1.50)
Widowed/divorced/separated	113 (74.3)	39 (25.7)	0.58 (0.27, 1.22)	0.45 (0.20, 1.11)
Educational status of mother
No schooling	579 (70.3)	245 (29.7)	1	1
Junior (1–4 grades)	185 (78.4)	51 (21.6)	1.53 (1.09, 2.17)a	0.91 (0.59, 1.43)
Elementary (5–8 grades)	330 (84.2)	62 (15.8)	2.25 (1.65, 3.07)b	0.83 (0.54, 1.28)
Secondary (9–12 grades)	259 (92.2)	22 (7.8)	4.98 (3.14, 7.89)b	1.17 (0.64, 2.16)
Diploma and above	58 (92.1)	5 (7.9)	4.9 (1.95, 12.39)b	0.81 (0.24, 2.69)
Education of partner/husband
No schooling	494 (68.2)	230 (31.8)	1	1
Junior (1–4 grades)	159 (77.9)	45 (22.1)	1.65 (1.14, 2.37)a	1.26 (0.83, 1.93)
Elementary (5–8 grades)	297 (82.0)	65 (18.0)	2.13 (1.56, 2.90)b	1.68 (1.12, 2.51)a
Secondary school and above	461 (91.1)	45 (8.9)	4.77 (3.38, 6.73)b	2.15 (1.31, 3.52)a
Occupation of mother
Housewife	585 (81.9)	130 (18.1)	1	1
Merchant	263 (83.8)	51 (16.2)	1.41 (0.80, 1.63)	1.11 (0.74, 1.67)
Employed	130 (87.9)	18 (12.1)	1.6 (0.94, 2.71)	1.42 (0.72, 2.78)
Farmer	431 (69.9)	186 (30.1)	0.51 (0.40, 0.66)b	0.97 (0.70, 1.33)
Birth order
First	436 (92.4)	36 (7.6)	1	1
Second	296 (82.2)	64 (17.8)	0.38 (0.25, 0.59)b	0.40 (0.24, 0.68)b
Third	231 (78.6)	63 (21.4)	0.30 (0.19, 0.47)b	0.38 (0.22, 0.67)b
Fourth and above	448 (66.9)	222 (33.1)	0.17 (0.11, 0.24)b	0.29 (0.16, 0.51)b
Recent pregnancy
Wanted	1,311 (79.2)	344 (20.8)	1	1
Unwanted	100 (70.9)	41 (29.1)	0.64 (0.44, 0.94)a	0.74 (0.47, 1.18)
Antenatal care utilization for the recent birth
Yes	1,313 (83.1)	268 (16.9)	1	1
No	98 (45.6)	117 (54.4)	0.17 (0.13, 0.23)b	0.17 (0.12, 0.25)b
Mother experienced any complication during childbirth for recent pregnancy
No	1,018 (76.5)	313 (23.5)	1	1
Yes	393 (84.5)	72 (15.5)	1.68 (1.27, 2.22)b	1.98 (1.44, 2.71)b
Mother knows a place for delivery service
Yes	1,112 (80.3)	273 (19.7)	1	1
No	299 (72.7)	112 (27.3)	0.65 (0.51, 0.84)b	0.79 (0.57, 1.49)
Mothers were aware of the danger signs of pregnancy, labor and delivery
Yes	993 (80.7)	238 (19.3)	1	1
No	418 (74.0)	147 (26.0)	0.68 (0.54, 0.86)b	1.08 (0.79, 1.49)
Mothers perceived that service providers are capable of handling birth-related problems
No	538 (70.1)	203 (29.9)	1	1
Yes	873 (82.7)	182 (17.3)	1.81 (1.44, 2.27)b	1.32 (0.97, 1.80)
Mothers perceived that there is availability of adequate equipment in delivery service
No	691 (73.5)	249 (26.5)	1	1
Yes	720 (84.1)	136 (15.9)	1.91 (1.51, 2.41)b	1.54 (1.11, 2.13)a
Distance to the nearest health facility
<30 min walking time	768 (85.4)	131 (14.6)	1	1
≥30 min walking time	643 (71.7)	254 (28.3)	0.43 (0.34, 0.55)b	0.53 (0.39, 0.71)b
Easy transport service from home to the nearest health facility
No	484 (75.2)	160 (24.8)	1	1
Yes	927 (80.5)	225 (19.5)	1.36 (1.08, 1.71)a	0.57 (0.42, 0.77)b

Note:

a p<0.05,

b p≤0.001.

Abbreviations: CI, confidence interval; COR, crude odds ratio.

Discussion

The aim of the study was to identify the predictors of skilled delivery utilization. Based on this, distance to health facilities, perception of mothers on availability of adequate equipment in the delivery service in their catchment area, having an educated partner, using antenatal care, lower birth order and experiencing any complication during childbirth were the significant predictors of skilled delivery service utilization.

The longer the walking distance to reach the nearest health facility, the lesser was the mothers likely to utilize skilled delivery service. Lack of easy transportation service to the nearest health facility was also reported as an important challenge. These findings are consistent with the findings of studies conducted in Nepal,¹⁹ Zambia²⁰ and Ethiopia,²¹ wherein physical distance to the maternity facility and lack of geographic access were reported as barriers for the place of delivery. On the other hand, freely available transport service for the pregnant mothers was reported to cause substantially reduced pregnancy-related mortality in Ethiopia.²² It is, however, important to note that the distance to a health facility in this study was assessed based on the mothers’ report, which is subjective; it may over- or underestimate the actual distance to the health facilities and may affect the associations. If a mother perceived that the distance to a health facility is considered to be a barrier, they are less likely to utilize services.

Mothers’ perception of how well a facility is equipped was significantly associated with utilization of skilled delivery services. This kind of perception could be based on previous self and friends/relative exposures to the service. Perception of poor quality reduces the likelihood of utilizing maternal service and pushes them to take risks by delaying service-seeking even in the face of serious complications.^19,20,23 Previous studies in Ethiopia have reported poor and inadequate supply of essential equipment in health facilities.²⁴ Similarly, a study in Zambia indicated that poor perception has an effect on poor utilization of skilled delivery service.²⁰ As a limitation of this study, measuring the perception using only quantitative approach may not be adequate to know the feeling of mothers.

Those mothers who reported they did experience any complications during labor and delivery in their recent pregnancy were more likely to use skilled delivery service. This indicated that mothers are not likely to use skilled delivery service unless they had experienced complications. This could be due to the poor knowledge of mothers on the complications and birth preparedness plan, cultural influence or poor counseling during pregnancy.¹⁶

We found that mothers giving birth have repeatedly been less likely to use skilled delivery service, compared to those giving birth for the first time. This indicates multipara mothers neglect the service for their subsequent pregnancy and births. This may be due to their previous bad experience of health outcome at the facility, cultural influence or poor perception of the complications. The finding is consistent with those of studies done in Ethiopia¹⁶ and the Democratic Republic of Congo,²⁵ wherein it was reported that mothers do not use skilled birth care as the birth order increases. This shows emphasis is needed to educate multipara mothers during antenatal care visits. Moreover, mothers who did not attend antenatal care for the index pregnancy were less likely to use skilled delivery service. Having pervious exposure to health facility may influence mothers to continue the service.^14,15 Mothers having an educated partner were more likely to use skilled delivery service. The association between education and service utilization is also in line with previous reports from Ethiopia and other African countries.^14,15,26

Similar to previous observations, mothers who did not know about the danger signs during pregnancy, birth and postpartum period were less likely to use delivery service.²⁷ Birth preparedness has a clear positive effect on the decision making for skilled providers.²⁷ Thus, awareness creation and offering client-centered counseling to mothers and families at an early stage of pregnancy during antenatal care or home visits at community are important.

Conclusion

Distance to the health facility, perception of mother to availability of adequate equipment in the delivery service in their catchment area, experiencing any complication during childbirth, using antenatal care, lower birth order and having an educated partner were the significant predictors of skilled delivery service utilization. Implementing community-based intervention programs that will address the physical accessibility of delivery service, such as the ambulance service and road issues; improving quality maternity service; revising the protocol for admission of mothers during the latent phase of labor, especially for those mothers who come from distant places; preparing adequate waiting rooms at the facility; and awareness creation, especially for mothers with high birth order, will likely reduce the current problem.

Author contributions

GF, YB, AW and WT conceived and designed the study, performed data analysis and wrote the paper. All authors read and approved the final manuscript.

Acknowledgments

Our heartfelt thanks go to Mekelle University and Addis Continental Institute of Public Health for all-round support. We thank the African Population and Health Research Center (APHRC) for the African Doctoral Dissertation Research Fellowships (ADDRF) toward financial and technical support (grant number ADF005).

Disclosure

The authors report no conflicts of interest in this work.