Study area and design
The study was conducted in Wolaita Zone, southern Ethiopia, 330 km southwest of Addis Ababa, the capital of Ethiopia. The Zone’s population was projected to be more than 2.6 million in 2020 [28, 29]. The 2020 Wolaita Zone Health Department report indicated ten hospitals (one referral hospital, two general hospitals, seven primary hospitals), 70 health centres, and 326 health posts (Wolaita Zone Health Department: Annual Progress Report, Unpublished). The facilities provide preventive, curative, and rehabilitative health services for over 2 million people in the Zone and neighboring zones . Accordingly, the Zone had 80 EmONC facilities. Of them, two non-governmental and eight government hospitals provided Comprehensive Emergency Obstetric and Newborn Care (CEmONC) services, and 70 health centres provided Basic Emergency Obstetric and Newborn Care (BEmONC) services in 2019 (Wolaita Zone Health Department: Annual Progress Report, Unpublished).
A facility-based cross-sectional study was conducted to assess the quality of EmONC service and its predictors in Wolaita Zone, southern Ethiopia. The study was conducted from October 01 – December 31, 2020.
The source population are all EmONC facilities and women and newborns who came for EmONC services in Wolaita Zone during the study period. Randomly selected health facilities and women and their newborns were the study populations of the study.
Women aged 18 years and above who visited EmONC facilities during the study period were included in the study. In contrast, women referred to another health facility and those who had major obstetric/gynecological surgical procedures (cesarean section, hysterectomy, colporrhaphy, cervical cerclage, etc.) were excluded from the study.
Sampling and sample size
Sampling of health facilities
Of the 22 districts of Wolaita Zone, seven (30% of the total) districts were randomly selected. There were 27 eligible health facilities in the randomly selected districts. To take a representative sample from the selected districts, we included 14 facilities (more than 50%) from the total eligible health facilities in the districts. Accordingly, two EmONC facilities from each district were randomly selected, making the total number of health facilities selected for the study 14.
Sampling of women
The sample size for observation of EmONC services and exit interview was calculated using the single population proportion formula based on the following assumptions: a-47% proportion (p) of women who received quality EmONC services in Tanzania , the normal distribution of z at 95% confidence interval, and 5% margin of error (d). To adjust for non-responses, the sample size was increased by 10%, making the required sample size for the study 422. This sample size was intended to assess the process and outcome components of EmONC service quality. The calculated sample was allocated to the facilities considering the previous year’s volume of EmONC services’ utilization. Accordingly, the proportional allocation ranged from 82 in Wolaita Sodo University comprehensive specialized hospital to 9 in Wadu health centre as per their volume of EmONC service utilization. The sample size calculation using the single population formula is shown below
After adding 10% non-response rate, the final sample size required for the study was 383 + 39 = 422.
On the other hand, all selected women who came for EmONC services were recorded. Using the Kth number generated, women were selected systematically as they came for EmONC services. In the previous year, 9211 women (2303 patients in 3 months) visited the selected facilities for obstetric emergencies, presenting grounds to calculate the kth number to choose eligible women systematically from the facilities. Accordingly, every fifth woman was selected for the study until the required sample size was met in each facility.
Data collection instrument
The triad of structure, process, and outcome of the Donabedian Framework for Health Care Quality , was used to assess the utilization of quality of EmONC services. The structural (input) quality of care was measured using a facility audit checklist that was also developed after reviewing the literature [15, 18,19,20, 31]. A structured EmONC services observation checklist was developed after reviewing different guidelines and instruments [16, 18, 19, 31] and was used to observe the EmONC processes, i.e., to measure the observed quality. The EmONC service delivery (process) observation and facility audit checklists were prepared and used in the English language. The exit interview tool (questionnaire) was developed by the investigators after reviewing the literature [15, 18,19,20, 31]. The exit interview tool contained items regarding the socio-demographic characteristics of women, factors associated with the quality of EmONC services, and output quality assessment items. The exit interview tool was developed in English, translated into the local language (Wolaita Dona), and re-translated into English to check the consistency.
A two-day training was given to the data collectors. The data were collected by 14 data collectors who had a BSc in nursing (midwifery) and had experience with collecting data with the Open Data Kit (ODK) application and had no history of working in the assigned health facility. Similarly, seven supervisors, who had MPH and experience with data collection and supervision were hired to collect the data. The data were collected using the ODK mobile application with android tablet phones. The data collectors filled the facility audit, EmONC care observation checklist, and exit interview questions loaded in the ODK. ODK submitted the data to an online server in real-time. One supervisor was assigned to two health facilities, checked the data collection processes, provided support for data collectors on-site, and provided feedback to them in real-time.
Facility audit data collection
The facility audit was conducted 1 week before the observation of EmONC services and exit interviews. Seven data collectors conducted the facility audit. They completed the different sections of the audit checklist by contacting the heads of the units of the health facilities such as the health facility manager/director, maternal and child health unit heads, pharmacy and laboratory unit heads, and document reviews. Additional staff was consulted for information that was not available by the above persons or on their referral.
Observation of EmONC services data collection
The data collectors enrolled the woman if she met the inclusion criteria and documented the care provided to the woman with the EmONC services observation checklist. The observation of EmONC care started at the initial patient assessment, followed by all the stages of labor, and ended up at discharge from the facility. This approach was supported by other studies [15, 18, 19, 31].
Exit interview data collection
The data collectors interviewed the woman after 6 hours of postpartum, or a discharge summary was issued to her; whichever came first was sufficient to initiate the interview. The exit interview was done privately in a room in the facility.
Data management and quality control
Before the data collection, a pre-test was conducted in a similar setting (out of the study area) to check for the appropriateness of the study tools. Regular supervision was provided by the principal investigator, co-researchers, and supervisors to the data collectors to check for completeness, and confusion was cleared at the end of each data collection day.
Since the study involved observation of the care process by health workers, ruling out the Hawthorne effect was impossible. However, several considerations were made to minimize the effect of the presence of observers on the providers’ behavior. Initially, the data collectors assured the care providers that the purpose of the study was not for evaluating their performance or reporting it to their supervisors. Besides, observers had informed care providers that individual data will not be shared publicly (published reports only refer to aggregate data). The investigators discarded the first five observations of each health care provider because studies reported that care providers reverted to their normal behaviors after being observed a few times (observations) by the same observers [32,33,34]. In addition, care providers were not aware of the items on the checklist, so they could not prepare in any way. For further caution, the data collectors were not assigned to facilities where they currently or previously worked.
The data were exported to Stata v17 (College Station, Texas) to clean, re-code, explore and do advanced analysis. The descriptive statistics were done using frequency tables, charts, and summary statistics. The principal component analysis (PCA) was conducted to determine the household wealth index of study participants using the DHS approach . The simple and multiple linear regression analyses were done to identify candidate and predictor variables of the index (discussed below) of the observed quality of EmONC services. Coefficients with a 95% confidence interval were used to declare the significance and strength of association. Variables with a p-value less than 0.25 in the simple linear regression were taken as a candidate for multiple linear regression, and those with a p-value below 0.05 in the final model (multiple linear regression) were declared independent predictors of the observed quality of EmONC services.
Linear assumptions, such as homogeneity of variances and normality were checked and fulfilled. In the multi-collinearity test, all predictor variables had a variance inflation factor (VIF) value below 5. The final model was found significant with the adjusted R2 value of 0.344, explaining 34.4% of the variation.
The study was conducted after receiving ethical approval from the University of KwaZulu-Natal Biomedical Research Ethics Committee (BREC) (Ref: BREC/00001744/2020), South Africa, and the Institutional Review Board (IRB) of the College of Health Sciences and Medicine Wolaita Sodo University (Ref: CARD 4/979/20), Ethiopia. Furthermore, permission to conduct the study was obtained from Wolaita Zone Health Department and all participating health facilities. Written informed consent was obtained from all participants. The participants were informed that they had full right to participate or not in the study. Furthermore, the objectives, benefits, and harms of research were communicated. Respondents were also informed that their responses would be kept confidential. During observation of care provision, the data collectors were passively observing (did not intervene) the EmONC care provided to women.
Emergency Obstetric and Newborn Care: is the care given to mothers and neonates during pregnancy, child delivery, and the postpartum period when she faces serious and life-threatening obstetric complications [15, 35].
Obstetric complication: a woman is classified as having obstetric complication if she had at least one of these; a) hemorrhage (antepartum and postpartum), b) prolonged and/ or obstructed labor, c) postpartum sepsis, d) complications of abortion, e) severe pre-eclampsia and eclampsia, f) ectopic pregnancy and g) ruptured uterus [15, 35].
Input quality: was measured using an index created by calculating the mean of performance of 75 items (66 items of structure and nine items of signal function tests) and computed out of 100%. The index included items that assess power and water supply, waste management, drugs, supplies, equipment, storage, examination room, delivery room, waiting area, and standard precautions. It also includes the nine items (seven for BEmONC and nine for CEmONC facilities) of signal functions tests to assess the readiness of EmONC facilities in the past 3 months preceding the study [31, 35, 36].
Observed quality: was measured using an index created by calculating the mean of performance of 42 items related to EmONC services (standard clinical actions) and computed out of 100%. The index includes items that assess the observed quality of care EmONC service, including general patient assessment and danger signs, standard precautions, standard procedures in the stages of labor, and communication .
Output quality: was assessed by using the woman’s satisfaction with the EmONC services she utilized. A total of 12 items using a 5-scale Likert scale tool (ranging from “strongly disagree” to “strongly agree”) were used to assess the woman’s satisfaction with the EmONC services given at the health facility and computed out of 100% .
Basic Emergency Obstetric and Newborn Care (BEmONC) services: are expected to provide the seven signal function tests, namely: parenteral antibiotics, parenteral uterotonics, parenteral anticonvulsants, manual removal of placenta, removal of retained products, assisted vaginal delivery, and neonatal resuscitation .
Comprehensive Emergency Obstetric and Newborn Care (CEmONC) services: are expected to provide the seven services given by BEmONC facilities and the additional two services, namely, caesarian section and blood transfusion services .
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.