Background
Women’s experiences in maternity care are of paramount importance. Positive experiences have been associated with improved maternal and infant outcomes [
1], while negative experiences have been associated with a higher risk of adverse postpartum mental health outcomes, including depression, anxiety, and fear of childbirth [
2]. Women who have negative childbirth experiences may also postpone future pregnancies or choose not to conceive again [
3]. Additionally, they often prefer cesarean section in their subsequent deliveries [
4].
Improving maternal care begins with evaluating the current state of affairs from the perspective of women, who are the primary users of the system [
5]. Assessing women’s maternity care experiences is crucial for evaluating the quality of care, as the findings can significantly impact policy decisions regarding the allocation of limited maternal care resources [
6]. Recognizing this importance has led to the development of psychometric scales in this field [
7].
The process of shaping women’s maternity care experiences starts during pregnancy and evolves through labour, birth, and the postpartum period [
6]. However, existing scales primarily focus on assessing labour and birth experiences, with fewer tools available for evaluating pregnancy and postpartum experiences [
8]. Recent efforts have aimed to develop scales that cover the entire maternity care continuum, such as the Pregnancy and Maternity Care Patients’ Experiences Questionnaire (PreMaPEQ) [
9], Women’s Perception of Their Entire Maternity-care Experience [
10], and Women’s Experience of Maternity Care (ESEM) [
11].
While these scales address various aspects of maternity care, challenges hinder their effective use [
7]. Some scales have an excessive number of items [
9] or lack test-retest reliability [
12]. Additionally, key statistical indices such as exploratory factor analysis (EFA) or confirmatory factor analysis (CFA) indices are not reported for some of them [
9‐
11]. Another significant criticism is the inadequate reporting of content validity indicators [
9‐
11]. It is crucial to consider women’s perspectives, needs, and expectations when assessing their experiences [
8]. The lack of information on the content validity of the scales indicates a deficiency in women’s involvement in their design [
7]. A recent systematic review of the psychometric properties of patient-reported measures of maternity care has also emphasized the importance of including women’s participation in the design of new scales, particularly for identifying relevant, understandable, and comprehensive items. The review also highlights the necessity of establishing the content validity of existing tools in other target populations [
8].
In England, Redshaw and colleagues (2019) developed and validated three concise, related scales to assess women’s experiences of maternity care (EMC) during pregnancy (EMC-PR), labour and birth (EMC-LB), and early postnatal (EMC-PN). These scales were developed using psychological and social theories, and findings of qualitative interviews with women and previous surveys to effectively address the key needs of women. The simple and short design of the scales makes them practical for use, with initial testing demonstrating suitable factor scores and internal consistency [
5]. Each of the three EMC scales has its own identified factor structure, allowing for individual use at various stages or collectively during the postpartum period [
5]. However, the original version of the EMC scales lacks reported quantitative content and face validity indicators.
In Iran, similar to other world regions, numerous tools have been validated to assess different aspects of care during childbirth. The use of these scales has highlighted a significant disparity between the care provided during this period and the desired standards in the country [
13‐
15]. There is a lack of information regarding women’s experiences during pregnancy and postpartum in the country. A study conducted in Zahedan, Iran, revealed that almost all women (97%) received prenatal care, but half of them received inadequate prenatal care (initiated after the 4th month or with less than 4 visits received) [
16]. Another study carried out in six provinces of Iran indicated that in high-risk provinces (determined by maternal mortality rates), the proportion of inadequate/intermediate prenatal care (care less than 6 visits) was significantly higher than in other areas (37% vs. 21%) [
17]. These studies mainly focused on the quantity of care.
To the best of our knowledge, a validated scale for assessing Iranian women’s experiences of maternity care from pregnancy through the postpartum period has not been established in Iran. If the validity of the EMC scales is confirmed in the Iranian context, their use could play a pivotal role in identifying and enhancing strengths, addressing weaknesses, and ultimately improving the quality of maternity care in our society.
Discussion
The results of the current study demonstrated the EMC scales as robust, valid, and reliable tools for assessing the maternity care experiences of Iranian women. The EMC scales used in this study demonstrated desirable content and face validity. CFA confirmed the structures derived from EFA, and the convergent validity of EMC-LB with CEQ-2 was established. Additionally, the study confirmed the divergent validity of the EMC scales by showing no relationship with women’s age. Furthermore, the scales were found to be reliable regarding internal consistency and test-retest reliability.
In our study, the KMO value of the EMC scales was satisfactory, and the extracted structures accounted for 57–66% of the cumulative variance. In the original version of EMC, similar scale structures had been extracted and confirmed, but the KMO value and explained variance were not reported [
5]. In a study conducted in Norway, eight translated versions of EMC-LB were used, but no information was provided on the psychometric indices [
28]. KMO and explained variance were not reported for the Pregnancy and Maternity Care Patients’ Experiences Questionnaire [
9] and Clark et al.‘s scales [
10], which have a similar purpose. The other measure, ESEM [
11], consists of 30 items in three scales of pregnancy, intrapartum, and postnatal, with structures explaining 55%, 62%, and 70% of the cumulative variance, respectively. In comparison to our study, these percentages of explained variances are lower for pregnancy, almost equal for intrapartum, and slightly higher for postnatal. It appears that the five-factor structure of EMC-PR encompasses more diverse aspects than the one-factor structure of ESEM-pregnancy. For instance, ESEM-pregnancy lacks items related to sonography and continuity of care. In ESEM-intrapartum and EMC-LB, there are aspects of care that do not exist in the other, such as the absence of any items about pain relief and continuity of care in ESEM-intrapartum and the absence of physical aspects of care in EMC-LB. ESEM-postnatal includes items about participation in decision-making, education on common problems during the postnatal period, and physical aspects of care, but it has no item about the first weeks of care at home. The Pregnancy and Childbirth Questionnaire (PCQ) has a two-factor structure for pregnancy and a one-factor structure for childbirth, explaining 53% and 56% of the cumulative variance, respectively. However, this measure does not address the quality of postnatal care [
12].
Despite the subscale loading variance for the
Continuity and
Antenatal checks subscales in the EMC-PR scale falling below the recommended threshold of 5%, we decided to retain these subscales for several reasons: Firstly, keeping these subscales enabled us to maintain alignment with the original scale, preventing the removal of associated items. Secondly, these subscales play a critical role in capturing specific aspects of the construct under investigation. Their importance is underscored by evidence from Cochrane reviews, emphasizing the pivotal role of midwifery-led continuity of care [
29] and standard antenatal check-ups [
30] in enhancing maternal and neonatal outcomes. Additionally, the World Health Organization (WHO) advocates for both continuity of care and a minimum of eight antenatal contacts, incorporating effective interventions and tests, as essential components of antenatal care to promote a positive pregnancy experience [
31].
In the EMC-LB subscale, the item of
“I did not mind being looked after by midwives or doctors I had not met before” displayed a factor loading of 0.256, falling below the conventional threshold of 0.300. Nevertheless, we chose to keep this item in our analysis for two main reasons. Firstly, to maintain consistency with the original questionnaire, and secondly, due to the paramount importance of continuity of care in maternal and newborn health. A Cochrane systematic review highlighted the significant advantages associated with midwifery-led continuity of care, leading to enhanced woman satisfaction and improved outcomes for both women and newborns compared to other care models [
32]. Additionally, based on the evidence, the WHO recommends the implementation of this care model in settings with well-functioning midwifery programmes [
33]. However, only New Zealand has fully embraced it as a standard national practice, as documented in the literature [
34]. Therefore, the inclusion of such an item in the assessment of care quality holds particular relevance, especially within the societal context of Iran. Although Iran places a strong emphasis on enhancing childbirth satisfaction as part of its new pronatalist population policy [
35,
36], the adoption of this care model remains limited [
29,
37,
38].
Our study’s average scores of EMC-PR and EMC-PN are almost identical to those reported in a study conducted in England to develop the original version [
5]. However, our study reveals a lower score for EMC-LB, likely due to the absence of a woman-centred approach in childbirth services in Iran [
39‐
42]. A study within our research context recognized the necessity for effective interventions in childbirth care, including the presence of a companion, respectful care, effective communication, education, responsiveness to needs, participation in decision-making, reduction of unnecessary interventions, and provision of pain relief options [
43].
The strong correlation observed between EMC-LB and CEQ-2, a validated scale for assessing childbirth experience [
21], reinforces the validity of EMC-LB. This finding is consistent with the original EMC study, which also found a significant correlation between this measure and a question about the right to choose in maternity care. Measures assessing similar constructs are anticipated to exhibit a strong correlation [
44]. EMC-LB and CEQ share similar concepts, including pain relief, personal control, safety, confidence, information, communication, shared decision-making, and support. Notably, the variance explained by the EMC-LB scale with 12 items surpasses that of CEQ-2 with 23 items (57% versus 43%) [
21]. This could be attributed to EMC-LB’s emphasis on crucial concepts such as continuity of care [
45] and individualized care [
8], in addition to the shared concepts. The strong positive correlation between the EMC-LB subscale in care quality and CEQ professional support indicates the importance of receiving support from professional care providers in women’s experience of labour and birth care quality, as highlighted in other studies. A systematic review emphasizes the importance of having competent, reassuring, kind, and supportive clinical staff to facilitate a positive childbirth experience [
46].
In our study, Cronbach’s alpha for all EMC scales and subscales, except for the Antenatal checks’ subscale in EMC-PR, exceeded the minimum acceptable threshold. However, in the original version, four subscales related to EMC-PR, including Antenatal checks, had Cronbach’s alpha values below the acceptable threshold [
5]. The researchers did not provide an opinion on this matter. The lower internal consistency of the Antenatal checks’ subscale may be due to its limited number of items (only two). It is important to note that when assessing the internal consistency of a two-item scale, Pearson’s r is preferred over Cronbach’s alpha, with an r value greater than 0.15 considered satisfactory [
47]. In this instance, Pearson’s r value was 0.23. Additionally, the dual nature of the “I would have liked more antenatal checks and scans” item may contribute to its inconsistency with the “I felt I had the right number of antenatal checks with the midwife/doctor” item. A recent systematic review has highlighted a shift in the culture of prenatal care management in many low- and middle-income countries, where ultrasound technology has replaced important components of clinical exams [
48]. In Iran, the indiscriminate use of medical technology, especially in urban areas, is widespread. For instance, a study in the city of Urmia-Iran revealed an average of 5.9 ultrasound scans per woman during pregnancy, with most women overestimating the diagnostic power of ultrasound and expressing only a few negative feelings about it [
49].
In the present study, all EMC scales and their subscales had high test-retest reliability, except for the “EMC-PR-information giving” subscale, which had an ICC of 0.69. It is possible that receiving prenatal care from multiple caregivers may have interfered with responses to the items in this subscale. Some assessments of the reliability of maternity care experience scales [
5,
10,
12] have relied only on Cronbach’s alpha, while according to the COnsensus-based Standards for the selection of the health status Measurement INstruments (COSMIN), assessing ICC is of high importance [
50]. Test-retest reliability has not been mentioned for the original scale [
5]. Moreover, half of the 16 Pregnancy and Maternity Care Patients’ Experiences Questionnaire scales also had ICC values less than 0.8, which could be due to the response burden resulting from the large number of items (145 items) in these scales [
9].
The validity indicators, especially content and face validity, have been inadequately dealt with in psychometric evaluations of maternity care experience scales, as indicated by two recent systematic reviews [
7,
8]. It is argued that when claiming that a scale is designed for measuring women-centred maternity care experiences, it should incorporate women’s participation in decision-making for relevant, understandable, and comprehensive items [
8]. Content validity is the most important characteristic of a patient-reported experience measure. It ensures that the scale’s content precisely reflects the phenomenon that the scale’s user intends to measure [
51]. In this study, we attempted to investigate more psychometric indicators and present a more comprehensive report of the relevant results.
Maternity care managers and providers, especially midwives, are expected to use these validated short scales to assess the quality of current care. Furthermore, researchers are encouraged to employ these tools to explore the effects of different care models in various studies.
Strengths and limitations
This study holds significant importance for several reasons. Firstly, it assessed the EMC scales’ psychometric properties for the first time in a cultural context outside of England. We reported comprehensive validity and reliability indices in this paper. Also, we executed exploratory and confirmatory factor analyses on two distinct random samples of the participants. The study was conducted among randomly selected women covered by the health centers of Tabriz, the capital city of the fifth-largest province in Iran. Given that the centers cover the majority of postpartum women, the results might be applicable to all eligible women in the city. Most of the fertility indicators in this city closely align with the mean of the indicators in urban areas of the country. Therefore, the results may be generalizable to all urban areas of the country.
This study’s results may not apply to women who have had multiple pregnancies, premature births, or those who have infants with disabilities, as these groups were not included in the study. Furthermore, due to the small number of participants with pregnancy complications, a history of illness, or those who underwent emergency cesarean sections, the results may not be generalizable to these groups and they may have different experiences.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.