Article Text

Original research
Quality of childbirth care and its determinants along the continuum of care among pregnant women who gave birth vaginally in Gondar town public health facility, Northwest Ethiopia, 2022: generalised structural equation modelling
  1. Wondwosen Abey Abebaw1,
  2. Haileab Fekadu Wolde2,
  3. Werkneh Melkie Tilahun3,
  4. Zenebe Abebe Gebreegziabher4,
  5. Destaw Fetene Teshome2
  1. 1Department of Epidemiology and Biostatistics, Woldia University, Woldia, Ethiopia
  2. 2Epidemiology and Biostatistics, University of Gondar College of Medicine and Health Sciences, Gondar, Ethiopia
  3. 3Department of Public Health, Debre Markos University College of Health Science, Debre Markos, Ethiopia
  4. 4Department of Public Health, Debre Birhan University, Debre Birhan, Ethiopia
  1. Correspondence to Wondwosen Abey Abebaw; wondeabey{at}gmail.com

Abstract

Objective To assess the quality of childbirth care and its determinants along the continuum of care in Gondar town public health facility in Ethiopia.

Design An institution-based, cross-sectional study was employed. Completed data were imported to Stata V.16 for cleaning and analysis. A generalised structural equation model was employed to examine the relationships along the continuum of childbirth care and to determine the factors affecting the quality of childbirth care.

Setting and participants This study was conducted among a total of 865 women who delivered in the public health facility of Gondar, Ethiopia, from 19 May to 30 June 2022.

Results The study revealed the proportion of good-quality childbirth care during admission, intrapartum and immediate postpartum period was 59% (95% CI 55.7, 62.4), 76.8% (95% CI 73.8, 79.5) and 45% (95% CI 41.7, 48.5), respectively. Postsecondary educational status of mothers (β=0.60, 95% CI 0.16, 1.04) and maternal age of 25–35 (β=0.68, 95% CI 0.33, 1.02) were predictors of quality of care at admission. Referral hospital (β=0.43, 95% CI 0.10, 0.76), presence of guidelines (β=1.36, 95% CI 0.72, 1.99) and provider age of 25–35 (β=0.61, 95% CI 0.12, 1.10) affected the quality of care during the intrapartum period. Urban residence (β=0.52, 95% CI 0.12, 0.93), skilled birth attendant experience (β=0.19, 95% CI 0.11, 0.28) and number of delivery couches (β=−0.29, 95% CI −0.44, –0.13) had significant associations with the quality of childbirth care during the immediate postpartum period.

Conclusions Although our study found improvements in the quality of childbirth care along the continuum compared with previous studies, more workers are needed to alleviate the problem of poor-quality service. Different maternal, provider and facility factors were found to be predictors of the quality of childbirth care.

  • Maternal medicine
  • Primary Health Care
  • Risk Factors

Data availability statement

Data are available upon reasonable request.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • There may be observer bias between data collectors, as well as a Hawthorne effect, during data collection.

  • The tool we have used is a process quality measurement that focuses only on routine care and excludes caesarean section delivery.

  • Quality of care was assessed using validated context-based indicators that provide a more detailed picture of the state of quality of care in the process of childbirth.

  • In addition, we studied the quality of care throughout the continuum (from admission to the immediate postpartum period), which is better than a single-point study.

Introduction

Quality of care is defined by the WHO as the extent to which health services for individuals and populations increase the likelihood of achieving the desired health outcomes.1 2 The quality of maternity care involves offering a minimum level of care to all pregnant women and their newborns, as well as providing a higher level of care to individuals who need it. The care should result in the best possible medical outcome and satisfies the clients and their families.3 Continuum of care refers to the entire clinical experience of a mother and a newborn, from initial assessment before delivery to the first 1 hour of childbirth care, which includes the initial assessment, intrapartum, and immediate postpartum and newborn periods.4

Child and maternal survival has been one of the most important advances and health priorities worldwide in recent decades, as evidenced by their adoption as the fourth and fifth Millennium Development Goals, respectively,5 and later the third Sustainable Development Goal.6 However, improvements in coverage of life-saving interventions for maternal, newborn and child health (MNCH) have not consistently translated into reductions in mortality from preventable conditions.7 In 2019, 295 000 maternal deaths,8 9 2.5 million neonatal deaths10 and 2 million stillbirths were reported.11 Since 2000, Ethiopia has successfully reduced maternal and child mortality rates by 50%. Nonetheless, the current maternal mortality rate, standing at 267 per 100 000 live births in 2020, remains unacceptably high,12 while neonatal mortality is 33 per 1000 live births, which is higher than the Demographic Health Survey 2016 report (29 deaths per 1000 live births).13 14 The majority of these deaths occur during labour, childbirth and the early postnatal period.15 Estimates suggest that poor-quality services account for 61% of neonatal conditions and half of the maternal deaths in low-income and middle-income countries (LMICs).16 A study done in Ethiopia showed that only 20% of mothers received good-quality care during routine childbirth in general.17

The availability of trained birth attendant services is the best strategy to reduce intrapartum-related and postpartum-related deaths.18 19 It is estimated that professional birth attendants can prevent approximately 13%–33% of maternal deaths and 25% of newborn deaths.15 20 Recently, delivering babies in healthcare facilities has not consistently demonstrated a correlation with decreased maternal or neonatal mortality. Moreover, it has not guaranteed the implementation of appropriate interventions during the intrapartum and immediate postpartum periods.4 21 22 Consequently, in recent years, increased emphasis has been placed on the quality of care, especially in LMICs. For instance, in 2017, a network comprising 10 countries, including Ethiopia, was established with the goal of enhancing the quality of care for MNCH. The objective is to reduce maternal and newborn deaths, as well as stillbirths, by half. Additionally, the aim is to enhance the overall care experience for pregnant women, mothers and their infants in healthcare facilities by the conclusion of 2022.23

It is essential to address multiple obstacles in order to implement effective strategies for improving the quality of delivery care. One such obstacle is the conceptual difficulty in assessing the multidimensional quality of life for both mothers and newborns. The intricacies and unpredictability of obstetric problems make assessing quality even more challenging.3 However, the most recent research has adopted Avedis Donabedian’s approach, which defines quality of care by three components: structure, process and outcomes.24 Process indicators have been proven to best reflect the quality of care for mothers and newborns in the component Donabedian framework.25 Although there is no consensus on the standard process quality indicators in LMICs, the approach proposed by Tripathi and her colleagues21 has gained significance. Their expert-based index tool measures process quality throughout the entire continuum of care, encompassing intake, intrapartum and immediate postpartum periods. It particularly emphasises routine care and importantly has been validated for application in Sub-Saharan Africa, including Ethiopia, and has recently been employed. This index encompasses the majority of dimensions related to quality of care and is effective in distinguishing between poorly performed and well-performed deliveries.21 The tool comprises three domains: quality of care at admission, quality of care during the intrapartum period and quality of care during the immediate postpartum care.

Despite an increase in the number of mothers who visit delivery facilities, there is a mismatch between supply and demand for quality of care due to a shortage of qualified providers, knowledge gaps, lack of access to evidence-based information and lack of motivation among healthcare workers.4 26 As a result, identifying the possible determinants of quality of care during this important period could have a significant impact on maternal and infant survival, as well as inform concerned bodies about where improvements must be made to enhance the effectiveness of health services. Previous research has mostly focused on the quality of emergency obstetric and newborn care only or the quality of routine care at a specific point (intrapartum or immediate postpartum period only) during the course of childbirth,17 27 and has rarely looked at the quality of routine care across the continuum of care.4 26 This study applied the methodology proposed by Tripathi et al21 to evaluate the quality of childbirth care across the continuum, using an appropriate statistical model known as generalised structural equation modelling (GSEM). This modelling approach considers the specificities of the quality indicators used at each point of care, rather than treating them as equal in weight. It also facilitates the exploration of relationships along the continuum, acknowledging that the quality of care at subsequent stages may be influenced by the quality received at previous stages. Moreover, this model enables the simultaneous inclusion of multiple dependent variables and the estimation of the indirect effects of the variables. Therefore, this study aimed to assess the quality of childbirth care and its determinants along the continuum in Gondar town public health facility, Northwest Ethiopia, 2022.

Methods and materials

Study design and setting

An institution-based, cross-sectional study was conducted from 19 May to 30 June 2022. The study was conducted at nine public health facilities in Gondar town, the capital of central Gondar Zone. Gondar is 727 km northwest of the country’s capital city, Addis Ababa, and 172 km northwest of the capital city of Amhara Region, Bahir Dar. Gondar has a total population of 443 156 based on the 2021 population estimation.28 The expected number of pregnant women in the town was estimated to be 15 315 in 2021/2022.29

The town has public and privately owned health institutions. The public ones include the University of Gondar Comprehensive Specialized Hospital and eight health centres (Maraki, Mintewab, Gebriel, Azezo, Poly, Tseda, Wolleka and Bilajig). There are also 25 private specialty clinics, 17 medium clinics and 15 primary clinics. Private health facilities do not provide normal labour services, except elective caesarean section.29

Population

All mothers who delivered at Gondar town public health facility, their newborns and their corresponding skilled birth attendants during the study period were selected as study participants after excluding pregnant mothers who were referred to or from other health facilities, admitted with false labour and delivered by caesarean section.

Sample size calculation

As a rule of thumb, the minimum sample size for structural equation modelling should be 5–20 times the number of free parameters to be estimated.30 For this study, the number of parameters to be estimated was 83 and we used the recommended 1:10 ratio of the free parameter and the number of respondents.31 Therefore, taking the participants to a free parameters ratio of 10, the sample size for this study was 830 (83×10). By adding a 5%32 non-response rate, the minimum sample size required for this study became 872 (830×0.05) pregnant mothers, their newborns and their corresponding skilled birth attendants.

Sampling techniques and procedures

We included all public health facilities of Gondar town. The total sample was proportionally allocated to each health facility based on the expected number of deliveries per 2 months using a formula for proportional allocation, Embedded Image, where ni is the sample to be selected from each health facility, ntotal is the total sample size of the study, Embedded Image is the total population and Ni is the total population of each health facility. A systematic random sampling method was used to select study participants. The sampling interval was calculated by dividing the expected number of deliveries per 2 months by the sample size for each health facility. The first study participant was selected by lottery method. Data were then collected from each study participant with skipping interval until the desired sample size was reached.

Variables and measurements

Structured and pretested electronic-based questionnaires were used to collect data, aided by an online open data collection kit application tool (ODK). We used the interview method to collect provider, facility and maternal characteristics, and observation methods to assess the quality of childbirth care along the continuum. The questionnaire contains maternal characteristics, provider characteristics, facility characteristics and a tool used to measure the quality of childbirth care.

12 data collectors (5 midwives, 5 nurses and 2 public health) who had undergone a 2-day training were assigned in the delivery room to directly observe and document healthcare providers’ adherence to a set of signal functions while caring for pregnant women and their future newborns at three critical points: admission, intrapartum and immediately post partum (soon after delivery up to ≤1 hour).

Tools to measure the quality of childbirth care along the continuum

We used the tool recommended in 2019 by Tripathi and her colleagues,21 which has been validated and recommended for use in Sub-Saharan Africa, including Ethiopia.

The quality of childbirth care at admission was assessed using seven observational indicators with two response options (yes and no) coded as 0 and 1, respectively. Thus, the total score of care provided to the delivering mothers ranges from 0 to 7, and a higher score indicates good-quality childbirth care. These indicators were the following: whether the health professional asked if the woman experienced any danger signs, assessed the mother’s HIV status, took the temperature, took the blood pressure, took the pulse, washed their hands before the examination and wore sterile gloves during vaginal examination.

The quality of childbirth care during the intrapartum period was assessed using seven observational indicators with two response options (yes and no) coded as 0 and 1, respectively. Thus, the total score of care provided to delivering mothers ranges from 0 to 7, and a higher score indicates good-quality childbirth care. These indicators were the following: whether the health professional explained all the procedures, prepared uterotonic drug for active management of the third stage of labour, used a partograph during labour, prepared newborn resuscitation equipment, correctly administered a uterotonic, assessed the integrity of the placenta/membranes and assessed for perineal/vaginal lacerations.

The quality of childbirth care during the immediate postpartum period was assessed using six observational indicators with two response options (yes and no) coded as 0 and 1, respectively. Thus, the total score of care provided to delivering mothers ranges from 0 to 6, and a higher score indicates good-quality childbirth care. These six indicators were the following: whether the health professional immediately dried the baby with a towel, initiated skin-to-skin contact, tied or clamped the cord but not immediately after birth, took the mother’s vital signs 15 min after birth, palpated the uterus 15 min after delivery and assisted the mother in initiating breast feeding within 1 hour. These quality measures reflect the minimum standards of obstetric care, irrespective of the type of health facility where the delivery service is performed.

Possible determinants of quality of care were categorised into three groups: maternal, provider and facility characteristics.

Maternal characteristics include the age of the mother measured in completed years, educational status (no education, primary, secondary and postsecondary), residence (urban, rural), mode of delivery (spontaneous and instrumental vaginal delivery) and number of antenatal care (ANC) visits.

Provider characteristics include age (in years), gender (male, female), years of experience (in years), qualification (midwifery and others) and training for compassionate, respectful care (CRC) (yes, no).

Facility characteristics include the level of health facility (primary, referral), presence of piped water supply (yes, no), presence of clinical guidelines (yes, no), number of delivery couches and number of deliveries in the past months. Considering these variables, a hypothesised model based on literature about the determinants of quality of childbirth care was formulated (figure 1). Quality of care at admission and quality of care during intrapartum are endogenous latent variables in our model because they are influenced by other exogenous variables in our model. In the context of mediation analysis, an endogenous variable is influenced by other variables in the model. A mediator is a variable that helps explain the process or mechanism through which an independent variable affects a dependent variable.33 In our specific model, quality of care at admission and quality of care during intrapartum function as mediator variables as they help clarify the pathways through which exogenous variables impact the quality of care at immediate postpartum care.

Figure 1

A hypothesised model on the quality of childbirth care and its determinants along the continuum, Gondar, Ethiopia. The broken line indicates all the variables on the tail side will not affect the variable on the head side. A rectangle or square indicates an observed variable, whereas an ellipse indicates a latent variable. CRC, compassionate, respectful care; QoC, quality of care.

Operational definition

The quality of childbirth care at admission was classified as good if the score was ≥75% (≥5 points) and poor if the score was less than 75% (<5 points).17 21 The quality of childbirth care during the intrapartum period was classified as good if the score was ≥75% (≥5 points) and poor if the score was less than 75% (<5 points).17 21 The quality of childbirth care during the immediate postpartum period was classified as good if the score was ≥75% (≥5 points) and poor if the score was less than 75% (<5 points).17 21

Data processing and analysis

Completed data were downloaded from the Kobo Toolbox platform, checked for completeness and then imported to Stata V.16 for further cleaning and analysis. Descriptive statistics were performed. Reliability was evaluated for each aspect of childbirth care quality using Cronbach’s α coefficient, with all values exceeding 0.7, indicating satisfactory reliability. Specifically, the α coefficient was 0.708 (with bootstrapped 95% CI 0.707, 0.767) for the overall scale, 0.733 (with bootstrapped 95% CI 0.668, 0.749) for quality of childbirth care at admission, 0.803 (with bootstrapped 95% CI 0.771, 0.829) for quality of childbirth care during delivery and 0.756 (with bootstrapped 95% CI 0.725, 0.776) for quality of childbirth care during the immediate postpartum period. Cronbach’s α coefficient values of 0.7 or higher were considered satisfactory.34

To achieve the objective of determining the factors that affect the quality of childbirth care, a mediation analysis was carried out using GSEM, and to provide references on this topic. Quality measures of childbirth care at admission, intrapartum and immediate postpartum periods are latent variables that constitute items with binary responses, so their measurement model was analysed with a binomial family distribution with the logit link function. The presence of more latent variables increased the model’s complexity, which took longer time to converge during analyses. To get a more parsimonious model, we conducted sensitivity analyses by comparing the fitness and simplicity of two competing models: a model when the quality of care at admission was taken as a latent variable that constitutes items with binary responses and another model when it was considered as a categorical observed variable; the quality of care at admission was changed into a categorical observed variable by adding the score of item responses for each participant and categorised into binary outcome variable using the cut-off point provided by the authors of the scale.21 Based on the sensitivity analyses conducted, it was found that considering quality of care at admission as a binary observed variable provided a good fit for the data. It also resulted in a simpler model since the number of parameters to be estimated would be reduced by the number of items on the scale.

Exogenous variables may not only have a direct effect on the quality of care, but may also have an indirect effect. Therefore, if the mediation of effects was statistically significant, then the direct, indirect and total effects were determined using the non-linear combination of the estimator technique (‘nlcom’ command).33 35 36 As there were few missing data for variables related to quality of care during intrapartum, equation-wise deletion was employed in the GSEM analyses.

After checking all of the above assumptions, the analysis was started with a hypothesised model (figure 1). The modification was taken iteratively by adding or removing a path link. Finally, an overidentified model with minimum information criteria (Akaike information criteria (AIC) and Bayesian information criteria (BIC)) was retained.36 37 For multivariable GSEM analysis, an adjusted unstandardised estimate (β) with its 95% CI was reported, and variables with a p value <0.05 were considered significant predictors of quality of childbirth care.

Patient and public involvement

Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.

Ethics approval

A letter of cooperation was sought from the selected health facility. Written informed consent was obtained from each study subject prior to the data collection process. During data collection, each study participant was informed by the data collectors about the purpose and the anticipated benefits of the research project. The study participants were also informed that they had the full right to refuse, withdraw or completely reject part or all of their part in the study. Additionally, they were informed that all data obtained will be kept confidential by using codes instead of any personal identifiers and that the data will be used only for the study.

Results

Background characteristics of mothers and skilled birth attendants

The study included 865 mothers who gave birth at a public health facility in Gondar town, with a response rate of 99.2%. Above half of the mothers (509, 58.84%) were within the age group of 25–34 years and ranged from 17 to 45 years (median age=28, IQR=7). 599 (69.25%) mothers lived in urban areas and 623 (72.02%) delivered spontaneously. Regarding educational status, 25.09% (n=217) had no formal education (table 1).

Table 1

Background characteristics of mothers and skilled birth attendants in Gondar town public health facility, Ethiopia, 2022

67 (84.81%) skilled birth attendants were aged between 25 and 35 years. The health providers at delivery were predominantly midwives (65, 82.28%). More than half of the skilled birth attendants (53, 67.09%) providing childbirth care were male health professionals (table 1).

Facility characteristics

Most (5, 83%) of the participating health facilities were primary health centres. Four out of six health facilities did not have a piped water supply. Maternal and newborn clinical guidelines were present in more than 80% of the healthcare facilities (table 2).

Table 2

Gondar town public health facility characteristics, Ethiopia, 2022

Quality of childbirth care along the continuum

We examined the disparity in quality of care between health centres and referral hospitals and identified no significant differences among the included health centres. Except for the indicator ‘assessed for HIV status’, all measures of quality of childbirth care at initial assessment and examination showed a statistically significant difference between primary and referral health facilities at a p value of 0.05. Similar to this, all indicators for ‘newborn and immediate postpartum care’—aside from the indicator ‘explaining all the procedures of delivery’—were statistically different at a p value of 0.05. Contrarily, for the two indicators of ‘intrapartum care’ (supporting the mother to initiate skin-to-skin contact and palpated the uterus 15 min after birth), the two categories of health facilities were not significantly different at p=0.05 (online supplemental table 1).

This study revealed that the percentage of childbirth care rated as good quality at admission, during intrapartum and in the immediate postpartum period was 59% (95% CI 55.7, 62.4), 76.8% (95% CI 73.8, 79.5) and 45% (95% CI 41.7, 48.5), respectively.

Factors affecting the quality of childbirth care along the continuum

The final model containing both the structural (relationships among latent or observed variables) and measurement (relationship between a latent variable and its indicator or items) components is shown in figure 2. The fitted model was relatively parsimonious and had the minimum information criteria (AIC and BIC) value when compared with other competing models (figure 1). Variables such as gender of birth attendant, number of ANC visits, mode of delivery, profession of the skilled birth attendant, presence of piped water supply and number of deliveries per month were excluded from the final model as their contributions were not statistically significant at an alpha level of 0.05 and have collinearity with other variables. In the fitted model, all path coefficients in the diagram were statistically significant at an alpha level of 0.05.

Figure 2

Determinants of quality of childbirth care along the continuum in Gondar town public health facility, Ethiopia. age_catm, age of mothers; age_catp, age of provider; crc_b, provider’s training for compassionate, respectful care (CRC); d1–d6,quality of childbirth care indicators during the intrapartum period; education_mo, educational status of the mother; level_f, level of health facility; p1–p6, quality of childbirth care indicators during the immediate postpartum period; QoC_Post, quality of childbirth care during the immediate postpartum period; QoC_Intra, quality of childbirth care during the intrapartum period; QoC_admission_cat, quality of childbirth care at admission; residence_m, residence of the mother.

Factors affecting the quality of care at admission

The quality of routine childbirth care at admission was strongly influenced by delivery in the referral hospital, the mother’s age and her educational status.

Deliveries in referral hospitals were significantly and positively (adjusted β=1.52, 95% CI 1.20, 1.83; p<0.001) associated with higher quality of routine childbirth care at admission compared with deliveries in primary health facilities, keeping other variables constant.

The quality of childbirth care at admission was positively and significantly associated with primary (adjusted β=0.65, 95% CI 0.26, 1.05; p<0.001) and postsecondary (adjusted β=0.60, 95% CI 0.16, 1.04; p<0.007) educational status of the mothers. However, there is no significant difference in the quality of childbirth care among mothers with no education and with secondary educational status. Mothers who were between 25 and 35 years of age had a direct effect (adjusted β=0.68, 95% CI 0.33, 1.02; p<0.001) on higher quality of childbirth care at admission compared with those who were between 15 and 24 years of age, keeping the other variables constant (table 3).

Table 3

Factors affecting the quality of childbirth care at admission in Gondar town public health facility, Ethiopia

Factors affecting the quality of care during intrapartum care

According to the results of the multivariable GSEM analysis, referral hospital, presence of maternal and newborn clinical guidelines, and age of the provider had a significant direct effect; age and educational status of the mother had a significant indirect effect mediated by the quality of childbirth care at admission; and the referral hospital had both direct and indirect effects on the quality of routine childbirth care during intrapartum (delivery) care.

Among the upstream variables (exogenous variables related to the mediator variable, quality of childbirth care at admission), the referral hospital had both direct (adjusted β=0.43, 95% CI 0.10, 0.76; p<0.011) and indirect (adjusted β=2.00, 95% CI 1.26, 2.75; p<0.001) positive effects on the quality of intrapartum care that resulted in a positive total effect (total β=1.44, 95% CI 1.68, 2.75; p<0.004). Facilities that had maternal and newborn clinical guidelines had a positive direct effect (adjusted β=1.36, 95% CI 0.72, 1.99; p=0.004) on good-quality childbirth care during the delivery of the newborn, holding other variables constant.

After controlling for all other variables, skilled birth attendants aged between 25 and 35 (adjusted β=0.61, 95% CI 0.12, 1.10; p<0.014) and above 35 years old (adjusted β=1.17, 95% CI 0.49, 1.84; p<0.001) had a direct positive effect on the quality of childbirth care during intrapartum care compared with a skilled birth attendant who was below 25 years of age.

Age and educational status of the mother were indirectly and positively associated with good-quality intrapartum care. More specifically, the quality of childbirth care during intrapartum care was positively and significantly associated with primary (adjusted β=0.87, 95% CI 0.29, 1.45; p<0.003) and postsecondary (adjusted β=0.80, 95% CI 0.16, 1.44; p<0.014) educational status of the mothers. Moreover, mothers who were between 25 and 35 years of age had a positive indirect effect (adjusted β=0.89, 95% CI 0.35, 1.44; p<0.001) on the quality of childbirth care during the intrapartum period compared with those who were between 15 and 24, keeping other variables constant (table 4).

Table 4

Factors affecting the quality of childbirth care during intrapartum care in Gondar town public health facility, Ethiopia

Factors affecting the quality of childbirth care during the immediate postpartum period

Age of the mother, residence of the mother, years of experience of the skilled birth attendant, training for CRC and number of delivery couches have statistically significant associations with the quality of routine childbirth care during the immediate postpartum period.

Urban residence had positive and significant direct effects (adjusted β=0.52, 95% CI 0.12, 0.93; p<0.012) on good-quality childbirth care during the intrapartum period. Pregnant mothers aged above 35 years were more likely to receive quality childbirth care (adjusted β=0.72, 95% CI 0.15, 1.30; p<0.013) during the immediate postpartum period compared with pregnant mothers who were 15–25 years of age.

The quality of routine childbirth care during immediate postpartum care significantly increased with the provider’s years of working experience (adjusted β=0.19, 95% CI 0.11, 0.28; p<0.001). Skilled birth attendants who received CRC training provided low-quality childbirth care during the immediate postpartum period (adjusted β=−0.67, 95% CI −1.10, –0.23; p<0.003).

The number of delivery couches had a negative direct effect (adjusted β=−0.29, 95% CI −0.44, –0.13; p<0.001) on the provision of quality childbirth care during the immediate postpartum period, holding other variables constant (table 5).

Table 5

Predictors of quality of childbirth care during the immediate postpartum period in Gondar town public health facility, Ethiopia

Relationships along the continuum of care

Along the continuum of care, quality of care at initial assessment has a positive and statistically significant association with quality of intrapartum care (β=1.22, 95% CI 0.82, 1.62; p<0.001).

Discussion

In this study, the prevalence of good-quality childbirth care at admission, during intrapartum and immediately post partum was 59%, 74.6% and 45.2%, respectively. The quality of childbirth care at admission was predicted by factors such as the level of health facility, the age of the mother and the mother’s educational status. Direct predictors of the quality of childbirth care during the intrapartum phase included the provider’s age, the level of the health facility and the presence of clinical guidelines, while the level of the health facility and the age and educational status of the mother were indirect predictors. Significant predictors of childbirth care during the immediate postpartum period encompassed the age of the mother, the mother’s residence, provider training for CRC, the provider’s years of experience and the number of delivery couches.

In this study, the proportion of good-quality childbirth care at admission was only 59% (95% CI 55.6, 62.3). This finding is in line with a previous study done in Ethiopia which suggested that the quality of childbirth care at admission was high (65.62%).38 However, another institution-based, cross-sectional study in Northern Ethiopia revealed that only 41.2%32 of mothers received good-quality childbirth care at admission. The reported figures cannot be directly compared due to differences in measuring standards (the instruments used for assessing care quality) across studies, variations in the period of data collection, discrepancies in the number of participants and distinctions in the types of facilities involved. For instance, the earlier study encompassed a primary health facility, whereas the present study incorporated both primary and referral hospitals. These disparities, either individually or in combination, contribute to the lack of full comparability between the numbers. This may also be a result of skilled birth attendants not adhering to established standards, ignoring services for mothers and babies, lacking the most fundamental interventions or lacking the necessary compassionate traits and skills to provide standard birthing care activities, for instance not washing their hands during examination, which is part of quality measure at admission.32 Hence, greater emphasis should be placed on ensuring that providers adhere to established guidelines. Additionally, stakeholders should fulfil essential requirements such as providing handwashing basins and gloves.

We observed a positive significant association between the quality of childbirth care at admission and the level of health facilities. Deliveries in referral hospitals were significantly and positively (β=1.52, 95% CI 1.20, 1.83) associated with a higher quality of routine childbirth care at admission compared with deliveries in primary health facilities. This result is supported by a study done in Kenya.4 The difference in performance at the different stages of health facilities (referral and primary) could be due to inadequate service providers’ skills and competencies, lack of qualified clinicians, heavy workload and poor job satisfaction at health centres as compared with referral hospitals.39 These findings underscore the importance of directing increased attention towards primary health facilities. This includes providing health professionals with enhanced skills, ensuring the availability of necessary materials and alleviating workload pressures.

The quality of childbirth care at admission was positively and significantly associated with the primary (β=0.65, 95% CI 0.26, 1.05) and postsecondary (β=0.60, 95% CI 0.16, 1.04) educational status of the mothers. This result was comparable with a study done in Uganda that showed that higher educational levels (β=6.81, 95% CI 0.85 to 15.46) were statistically associated with higher quality scores.40 Educated mothers are more likely to use health facilities that provide higher quality care; they can physically access and afford high-quality care; they know what type of care to seek and can advocate for it; they have higher expectations for care and insist on it; and they are more likely to have relationships with health professionals, which helps them obtain high-quality services. These may be a few potential explanations for the disparities in the quality of care by educational status. Additionally, the narrower social power gap between highly educated women and health personnel may also allow higher-status women to assert their preferences to obtain high-quality care.41–43

Mothers who were between 25 and 35 years of age had a direct effect (β=0.68, 95% CI 0.33, 1.02) on higher-quality childbirth care at admission compared with those who were between 15 and 24 years of age. This result indicates that when age increases, the respect and care for labouring mothers also increase, since respecting older people is part of the culture of most Ethiopian communities.44 However, the effect of maternal age on the quality of childbirth care was insignificant in previous studies.26 32 This finding suggests that due attention should be given to women who are below 24 years of age.

According to this study, more than half of the mothers (74.6%; 95% CI 73.5, 79.2) experienced a high standard of routine childbirth care during the intrapartum period. This finding aligns with similar studies conducted in Tanzania, where 60% of participants reported comparable results.45 This result is higher than other previous studies in Ethiopia, ranging from 20% to 34%.17 32 38 46 This finding is also higher than the study done in Tanzania (14%)47 and a study conducted in Ghana (27%).48 While the studies are not entirely comparable, the differences in results may be attributed to variations in measurement standards across studies, the timing of data collection and a heightened focus on maternal and childcare services by both government and non-governmental organisations over time.27 However, this study took a composite of variables to measure quality care and involved observations. Concerning our study, although over 50% of the mothers received routine quality care, there is a need for additional efforts to enhance the quality of care. This includes tasks such as ensuring the readiness of resuscitation equipment, proper completion of the partograph and thorough assessment of placental completeness, as these aspects exhibit relatively lower adherence in our context.

Referral hospital had both direct (β=0.43, 95% CI 0.10, 0.76) and indirect (β=2.00, 95% CI 1.26, 2.75) positive effects on the quality of intrapartum care that resulted in a positive total effect (β=1.44, 95% CI 1.68, 2.75). This result is consistent with the study done in Burkina Faso and Côte d’Ivoire,26 Sweden49 and Nigeria.50 The difference in performance at the different stages of health facilities (referral and primary) may be a result of hospitals housing senior professionals, many of whom may be connected to academic institutions. Healthcare professionals working in these facilities will frequently have the opportunity to update their knowledge via rounds on the wards, bedside conversations with students and a series of seminars that are typically planned as part of the institution’s procedures. This suggests that regular mentoring programmes could help health centres share their catchment hospitals’ experience, enabling providers to deliver high-quality services.32

Facilities that had maternal and newborn clinical guidelines had a positive direct effect (β=1.36, 95% CI 0.72, 1.99) on good-quality childbirth care during the delivery of a newborn. This is supported by a study done in Kenya which suggests that several potential interventions warrant investment to improve the quality of care. One such example is the use of maternal and neonatal clinical guidelines.4 The reason for this could be failure to adhere to or improper utilization of guidelines, such as partograph, which can delay treatment in cases of prolonged labour. these delay can subsequently result in obstructed labour, ruptured uterus, bleeding and infections in both the mother and the newborn.51 This study emphasises that proper utilisation of maternal and newborn clinical guidelines by skilled birth attendants has the potential to significantly enhance the quality of care.

Skilled birth attendants 25–35 years of age had a direct positive effect (β=0.61, 95% CI 0.12, 1.10) on the quality of childbirth care during intrapartum care compared with skilled birth attendants who were below 25 years of age. Older health professionals have the credibility to provide quality care since they have exposure to different patient conditions and clinical scenarios that contribute to the development of knowledge, technical skills and critical thinking, which are also assumed to influence the effectiveness of the quality of childbirth care.52 This finding reinforces us to give more attention to junior and young skilled birth attendants by giving more training and allowing them to share their experience with senior and adult skilled birth attendants.

The quality of childbirth care during intrapartum care was positively and significantly associated with primary (β=0.87, 95% CI 0.29, 1.45) and postsecondary (β=0.80, 95% CI 0.16, 1.44) educational status of the mothers. This study is in line with a study done in Uganda.40 According to the study, literate women are more likely to receive high-quality care than illiterate mothers. This might be because well-educated women are highly regarded and valued by everyone in the society, including health professionals.53

Mothers who were between the ages of 25 and 35 had a positive indirect effect (β=0.89, 95% CI 0.35, 1.44) on higher-quality childbirth care during the intrapartum period compared with mothers between 15 and 24 years of age. There is no study found that is in line with this result. However, a qualitative study in Ghana showed that health professionals give less respect to young mothers than older mothers, which may result in poor quality of childbirth care for those who are below 24 years of age.54

In this study, more than half of the study participants (54.74%; 95% CI 53.37, 58.07) received poor-quality childbirth care during the immediate postpartum period. Although the sample size and other issues like the tool used and the time of data collection must be considered when interpreting the results, this finding is lower than the study done in Ghana (67%) and different studies in Ethiopia (ranging from 68% to 72%).32 38 55 Even though there was an improvement in the quality of care, the quality of immediate postpartum care is still poor. Possible reasons for this poor quality of care can be work-related burnout; the gap between providers’ skills and knowledge; fear of litigation; poor motivation schemes and issues related to their retention; shortage of skilled birth attendants, mainly midwives; lack of authority to make decisions; inability to translate training into practice; and unavailability of adequate medications and necessary equipment.32 In our specific context, there is a need for us to focus on specific areas such as ensuring proper cord clamping, accurately measuring vital signs after birth and promptly initiating skin-to-skin contact.

Urban residence had a positive and significant direct effect (β=0.52, 95% CI 0.12, 0.93) on good-quality childbirth care during the intrapartum period. This result is supported by qualitative studies in Ghana which suggest that rural women are more likely to be treated poorly even when they seek care in urban facilities.41 54 This may be due to the low socioeconomic and educational status of women from rural areas, making them inferior compared with urban women.56

Mothers above 35 years of age were more likely to receive quality childbirth care (β=0.72, 95% CI 0.15, 1.30) during the immediate postpartum period compared with pregnant mothers who were between 15 and 25 years of age. Comparable results regarding this finding have not been found. The reason behind the high quality of care received by those who are above 35 years of age may be the high respect they receive in the hospital community.44

The quality of routine childbirth care during the immediate postpartum care significantly increased with the provider’s years of working experience (β=0.19, 95% CI 0.11, 0.28). This finding is supported by a study conducted in Burkina Faso and Côte d’Ivoire26 which revealed that a year of experience was significantly associated with having good-quality childbirth care. This may be due to more experienced health workers being more likely to have benefited from quality improvement initiatives (inservice training, supportive supervision, etc) that have reinforced their skills and competencies.32

Skilled birth attendants who received CRC training provided low-quality childbirth care during the immediate postpartum period (β=−0.67, 95% CI −1.10 to –0.23). This result is contradictory to a study done in Ethiopia,32 Malawi57 and Ghana,58 and suggests that possessing a CRC certificate may not always be a guarantee of high-quality care unless practitioners put knowledge and evidence into practice. The ability to enhance the degree of confidence in providing services as well as to enable health practitioners to adhere to good practices and standards may be achieved not only by training but should also be supported by coaching and mentoring.32

The number of delivery couches had a negative direct effect (β=−0.29, 95% CI −0.44, –0.13) on the provision of quality childbirth care during the immediate postpartum period. This finding is comparable with a study done in Kenya.4 The increment in the number of delivery couches may result in a significant workload for the staff in different facilities and indeed in different units within the same facility by increasing the number of deliveries. Other studies showed that when the number of deliveries increases, the quality of care is negatively affected.32 59–62

Along the continuum of care, the quality of care at initial assessment has a positive and statistically significant association with the quality of intrapartum care (β=1.22, 95% CI 0.82, 1.62). This result is in line with a study done in Kenya.4 Although the quality inputs for each stage of childbirth care may vary, it is important to note that these inputs are delivered by the same team of healthcare providers operating within the same working environments. This correlation between the quality of care at one phase and the subsequent phase is evident and holds relevance, necessitating its consideration in regression models.63

One of the limitations of this study is the potential presence of observer bias among data collectors, as well as a Hawthorne effect. However, given the extended duration of observation, it becomes challenging for healthcare personnel to sustain artificial standards of behaviour throughout the entirety of the observation period. Consequently, any observed changes in behaviour are more likely to persist during the initial stages of childbirth care.64 To mitigate this effect, measures were taken in this study. First, the initial observation from each skilled attendant was excluded to minimise any potential Hawthorne effect. Additionally, experienced data collectors were recruited and underwent rigorous training and standardisation of data collection instruments beforehand. Furthermore, continuous supervision was implemented throughout the data collection process to ensure consistency and accuracy.

The tool used in this study focuses on measuring process quality, specifically targeting routine care. However, it is challenging to ensure that the quality indicators employed strongly correlate with maternal health outcomes within our sample. Additionally, making inferences about the determinants of quality of care provided to specific subgroups or less common events, such as mothers requiring caesarean delivery, is difficult. To address this concern, we only included mothers who had spontaneous deliveries assisted with instruments, aiming to reduce mix-up of populations. Nevertheless, these limitations could potentially lead to associations being underestimated or overestimated. Therefore, it is important to consider these limitations when interpreting the findings. Despite the mentioned limitations, this study possesses several notable strengths. The assessment of quality of care was conducted using validated indicators that are specifically tailored to the local context, offering a comprehensive and detailed understanding of the state of quality of care during the childbirth process. Furthermore, this study uniquely examined the quality of care across the entire continuum, spanning from admission to the immediate postpartum period. This aspect is particularly significant as it represents the first study of its kind conducted in Ethiopia, providing valuable insights into the quality of care during childbirth in the country.

Conclusion

The study found improvement in the quality of childbirth care along the continuum compared with previous studies, but more workers are needed to alleviate the problem of poor-quality service. The quality of childbirth care at admission and during intrapartum was good in more than half of delivery care. However, there is poor quality of childbirth care in more than half of deliveries during the immediate postpartum period. Referral hospital, age of the mother, and primary and postsecondary educational status of the mother significantly impacted the quality of routine childbirth care at admission. Referral hospital, presence of maternal and newborn clinical guidelines, and age of the provider had significant direct effects; age and educational status of the mother had significant indirect effects mediated by the quality of childbirth care at admission; and the level of health facility had both direct and indirect significant effects on the quality of routine childbirth care during intrapartum (delivery) care. Age of the mother, residence of the mother and skilled birth attendant experience have a positive statistically significant impact on the quality of routine childbirth care during the immediate postpartum period. However, training for CRC and the number of delivery couches are negatively associated with the quality of childbirth care during the immediate postpartum period. The quality of care at the initial stage has a significant impact on the quality of care at a later stage of delivery care. Thus, strategies should be designed and implemented to address the poor quality of service.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the University of Gondar, Institute of Public Health Ethical Review Committee (reference number of IPH/2136/2014). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

We would like to extend our heartfelt gratitude to the University of Gondar for providing us ethical clearance to do this research. We would like to acknowledge the study participants, data collectors and the Gondar City Health Office for their time and contribution to this work.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors WAA conceived of the research topic, designed the methods and materials, involved in the data collection, conducted data analysis, drafted and finalised the manuscript, and is also the guarantor of this work. HFW, WAA, WMT, ZAG and DFT all contributed to data analysis, study design and supervision of data collection. All authors participated in manuscript revision for intellectual content and approval of the final version. All authors read and approved the final manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.