A critical review: developing a birth integrity framework for epidemiological studies through meta-ethnography

The first three steps are equivalent to other review methodologies, and can be described according to the PRISMA guidelines and flowchart for reporting systematic reviews of qualitative and quantitative evidence [25]: (1) Identifying and defining the topic and purpose of the review, (2) Deciding what is relevant to the initial interest (3) Reading the studies. The search streategy was developed according to the PICO format [26] using Boolean operator. For P (Population), we subdivided the search terms into birth (e.g. “birth”, “delivery”, “labour”) and setting (e.g. “maternity ward”, “obstetric care”, “birth clinic”) and connected them by OR. We connected I (Intervention) and C (Comparison) by OR as well, listing search terms for the violations of maternal rights or deficient care (e.g. “birth violence”, “discrimination”, “disrespect”) and the protection of maternal rights or respectful maternal care provision (e.g. respectful maternity care, self-determination, autonomy, informed consent). Electronic searches were conducted in PubMed, PsychInfo, CINAHL and Embase. Studies that we considered eligible for inclusion researched aspects directly relevant to the quantitative measurement of maternal care conditions or provision, birth experiences or the perception of birth. We excluded studies assessing maternity care’s general quality, access to and utilization of maternity health care services or pregnancy care, and studies on violence outside the context of labour and birth. Setting-wise, we included studies on facility-based childbirth (e.g. hospitals, obstetric clinics, birthing centres) and excluded studies on assisted and non-assisted home birth. Concerning the population, we included people during childbirth and early postpartum phase and attendees of the birthing context (e.g. health care professionals, doulas, partners). We excluded studies on people using reproductive health services other than those revolving around childbirth (e.g. fertility treatment, abortion). Primary quantitative studies of different types were included (e.g. cross-sectional, cohort-studies), whereas qualitative and mixed-methods studies and secondary research, reviews, editorials, commentaries or conceptual articles were excluded. The scope was limited to studies published between 2010–2020 since the thematic framework proposed by Bowser and Hill in 2010 [8] led to a significant increase in research.

After having included and read all eligible studies, we extracted all relevant information into a pre-structured data extraction table. The table comprised general study information (e.g., country, population, study aim), the definitional scope of the frameworks, concepts or terms, and the items used for measurement. For more information on search strategy, data extraction, and a comprehensive overview of the studies’ characteristics, see: Additional files 1, 3, and 4.

Determining how the studies are related (4), Translation of the studies into one another (5), and Synthesizing translations (‘in-line-argumentation’) (6). For this phase, we distinguished between two levels of analysis: the conceptual and measurement levels. The ‘data’ for the conceptual level refers to the lens through which the studies approach their research (e.g., terminologies, concepts, framework). We first grouped studies that showed similarities in their approaches into conceptual clusters. Then, essential terms, frameworks, or underlying perspectives were derived to create a shared definition for each conceptual cluster. Subsequently, we compared the metaphors and concepts of one article within clusters with those in others. We elaborated each clusters’ primary focus, distinguishing from each other the different approaches while at the same time indicating affinities between the research strands. Then, we performed an in-line argumentation that prepared ground for further conceptual development, including a differentiation between measurement components. At the measurement level, the ‘data’ is constituted of the individual items of each study’s data collection instrument. All items of each study were tagged with an underlying theme by using content analysis (e.g., “Some health workers shouted at me because I haven’t done what I was told to do” →  verbal abuse). Then, we compared each article’s items-themes with those in others to condense and merge (‘translate’) themes into one another. Themes were grouped by similarity, and we again used content analysis to create specific, but not redundant, overarching categories and subthemes.

Express the synthesis (7). Syntheses can be expressed by designing a framework, model, hypothesis, or theory that is supported by findings from the previous steps. Driving from the conceptual level analysis, we will introduce the concept of birth integrity. Based on this, and on the results of the measurement level, we present a six-field framework in which all categories identified at the measurement level were placed.

To meet our first research objective, we identified and delimited from each other the existing frameworks and concepts through which research maternity care provision and birth experiences were researched. We arrived at 6 main conceptual clusters, namely: Disrespect and Abuse/ Mistreatment during facility-based childbirth (D&A/ MISC) [27‐56], Respectful maternity care (RMC) [57‐63], Obstetric violence (OV) [64‐70], Person-centered care (PCC) [71‐75], Childbirth experiences (CE) [76‐82], Maternal satisfaction (MS) [83‐98]. We grouped concepts that were marginally used in a separate cluster named “Other” [99‐107]. The results are presented in Table 1. The key concepts that were built therefore reflect the ‘shared quintessence’ of each cluster. An overview of the process of extracting key concepts can be found in Additional file 5.

The cluster D&A/MISC shows a strong focus on care interactions between healthcare professionals and the parturient and specifically emphasizes experiences of poor maternal care. In contrast to D&A/MISC, RMC takes a right-based and ethical perspective in claiming universal and fundamental rights for every woman in childbirth. While many of the rights correspond to D&A/MISC in building the desirable opposite, it is important to note that RMC highlights specific rights in childbirth that reach beyond the mere absence of abusive or disrespectful acts. OV shows high similarity to D&A/MISC and additionally points to the wider societal dimensions of structural and gender-based violence and addresses structural, cultural, and healthcare-related drivers that favor violative conditions, processes, and behaviors in obstetric care. PCC shows affinities with RMC and highlight that maternity care should be respectful of and responsive to the parturients’ preferences and needs, ensuring that the parturients’ values guide all clinical decisions. (CE) takes the individual’s evaluation of their birth as their starting point, claiming that the subjective experience of childbirth and the quality of relationship with the healthcare professional impact postpartum health and wellbeing. A similar focus is visible in the studies in the MS cluster: here, the impetus lies within the parturient’s expectations towards and feelings during birth, the satisfaction with the care provided, and how this impacts post-partum health.

With our second objective, we aimed to depict how the quality of maternity care provision and birth experiences have been operationalized and measured in quantitative studies. In all, we identified 72 themes and subsequently organized these by defining 14 overarching categories. We briefly introduce the categories and their related subthemes. An extensive overview of all underlying analytical steps (items, content analysis, juxtaposition of themes) and a presentation on the final themes organized by research strand (e.g., D&A, OV, RMC) is available in Additional file 6.

For the meta-ethnographic synthesis, we first derive from the conceptual analysis four principles that will guide the conceptual synthesis:

We additionally see three components that need to be reflected in the operationalization of studies on maternity care provision and maternal experiences of birth:

We suggest birth integrity as an umbrella concept that includes the four principles and helps to distinguish between the three components outlined before (Fig. 2:

Building upon the conceptual model of birth integrity, we make use of a multilevel framework that has initially been developed by sociologist Ritzer [108] for the social analysis of societies and distinguishes between a macro-and microlevel and a manifest and latent side. Manifest refers to anything objectifiable (e.g., directly observable, or assessable like language, laws, and interaction), whereas latent comprises of anything underlying or not directly assessable like culture, discourses, perceptions, or beliefs. We have adapted this framework by adding a third level of analysis (mesolevel) and applied it to our topic. The macro-manifest side e.g., comprises of structural factors such as the health systems (policies, coverage, legislation), whereas the macro-latent side includes a societies’ dominant discourse. On meso-level, the manifest side describes aspects related to institutions or organizations structures and processes, and the meso-latent level covers organizational cultures (shared norms, patterns of action). The individuals and their interactions with the environment reflect the micro-manifest level. Last, on the micro-latent level, the individuals’ expectations, and patterns of thought and action are organized.

Elsewhere, we theoretically underpin the multilevel birth integrity framework by outlining the concept within medicalization and risk theory, intersectionality, and embodiment theory [109]. By doing so, we open opportunities for further development in two directions: on the one hand, the multilevel birth integrity framework reflects the status quo of birth integrity-related measurements. Synchronizing the identified categories (and subtopics) with theoretical considerations demonstrate the gaps of current measurements. On the other hand, the evidence given by the categories identified in the studies’ measures (e.g., on non-consented care or non-evidence-based practices) can be taken up from theoretical (and health policy) positions and incorporated into existing discourses and policies.

In each of the six fields, we can place at least one category that reflects determinant of birth integrity.

On the macro-level, we assign in the manifest expression the category ‘Health service capacity’ and in the latent expression the category ‘Societal discriminatory norms’. We have matched most categories on the meso-level. On the meso-manifest side, we locate the categories: ‘Facility’, ‘Organization of care and management of the facility’, ‘Professional care’ and certain aspects of ‘Information, explanation, consent’. On the latent side, we assign the categories, ‘Dignified care’, ‘Birth accompaniment’, ‘Confidential care’, ‘Supportive care’ and ‘Information, explanation, consent’. On the macro-manifest side, we place the categories ‘Personal rights, ethics, and equity’ and on the macro-latent side category ‘Attitudes and expectations towards birth’.

Besides these categories identified as determinants, we characterized one category each for the subjective outcome of birth integrity (Childbirth perception and feelings) and the objective outcome (Health-consequence of violative birth experiences). Table 2 gives an overview of the multilevel birth integrity framework including the categories and subthemes of the macro-to-micro level in manifest and latent expression.

Current epidemiological research reflects and integrates a multitude of birth integrity determinants into quantitative study designs. Most measures refer to the meso- and microlevel as they put the institution and the individuals involved in childbirth in focus (health facility, healthcare professionals, person who gives birth). We propose to consider to a greater extent the macro-level driver(s) that potentially determine how maternal care is executed. This implies a stronger focus on structural forms of disrespect towards women that have manifested in a lack of resources (human, equipment, infrastructure) in maternity care settings [110]. Thereby we have in mind the financing of obstetrics (e.g., remuneration system, insurance coverage), policies and laws concerning obstetric and midwifery care (e.g., on staff sub-limits, ensuring sufficient staff and bed availability to prevent overcrowding, laws addressing obstetrical violence), the availability of maternity clinics close to residence (in remote areas), free choice of birth location (e.g., midwifery-led clinics or labor rooms for low-risk-pregnancies, specified resourced facilities for high-risk pregnancies), the existence of routes to report experiences or observations of violations against fundamental rights in childbirth, and abusive events during obstetric or midwifery care [111]. We also consider it relevant to integrate macro-level indices of discrimination as potential determinants of birth integrity, e.g., gender equality indices or racial equity indices [112].

A society’s gender relations, practices, and norms have cultivated dominant narratives that are mirrored in expectations, ideas, semantics, and actions around childbirth (e.g., on femininity and motherhood). Research on birth integrity should include measures on a society’s dominant gender roles, on narratives of childbirth medicalization, or reflect on how childbirth is negotiated as ‘risky’ at the macro-latent level. By considering medicalization and risk discourses, light is shed on different knowledge and normative patterns that highly determine how practices around birth are organized (e.g. ‘biomedical’/ ‘technocratic’ vs. ‘women-centered’/ ‘humanized' paradigm of birth [113]).

Additional to the manifold determinants identified at the meso-manifest level, we propose to examine in more detail how a facilities’ amenities impact options for coping with labor progress or pains, e.g., availability of warm-watered tubs or space for motion to find relaxing poses or favor the progress of birth [114, 115]. Technical equipment (e.g., wireless electronic fetal monitor) can enable movement. Similarly, posters that illustrate body postures beneficial to labor progress may positively affect the course of birth. One of the most discussed and obvious support while birthing is the continuous, 1-to-1 support, usually provided by a midwife or doula [116, 117]. (Epidemiological) research on birth integrity should in more detail evaluate how continuous support impacts birth integrity.

At the meso-latent level, we identified a strong emphasis on how the person in labor and birth is approached in terms of communicative skills (e.g., effective communication, respectful tone). As language matters, a rather passive/ passivating or active/activating wording implicitly creates certain expectations and perceptions about one´s birthing capabilities and a reflection of (dis)respecting a childbearing individual’s autonomy. For example, while the phrase ‘a baby is delivered by the doctor’ terminologically puts the parturient into a waiting and passive position, the term ‘giving birth’ connotes the actively involved childbearing person [118].

At the micro-manifest level, more emphasis could be directed to birth-related discrimination. We think it is useful to integrate intersectional theory [119] into measurement endeavors, as it highlights how intersecting lines of difference along with race, gender, sexuality, ability, religion or class create privileged and marginalized positions that result in advantages and disadvantages. Taking an intersectional lens on care provision (e.g., upon frequency and duration of care attendance during birth) or more birth-related outcomes (e.g. scrutinizing racial disparities in maternal and newborn mortality [120]) might be fruitful for research on the determinants of birth integrity.

For (violated and preserved) birth integrity, we found in the studies validated scales that capture different aspects of birth perception, e.g., on agentry and control [121] or satisfaction [91, 122]. Yet, we consider it highly relevant, and will address in our future research, to develop and validate a specific scale that measures whether a person’s birth integrity felt violated or protected during birth. Components of this measure could range from feeling embarrassed, scared, anxious, sad, out of control, ignored, vulnerable, dissociated, dehumanized, or traumatized to feeling seen, supported, and respected, deemed important, taken seriously, braced, empowered, confident, energetic, comfortable, and sheltered.

Supposing that preserved birth integrity builds a vital resilience for maternal health, simultaneously implicates that a violation of birth integrity may entail poor health outcomes in the short- or long run. As mentioned earlier, we did not integrate all the possible adverse health effects of birth integrity violations in our search. Nevertheless, we identified a few measures linking negative childbirth experiences to postpartum mental health conditions. These initial linkages between birth integrity violations and existing knowledge on traumatic birth experiences need to be systematically deepened. Research on traumatic birth experiences has so far focused on medically complicated deliveries or undesired health outcomes. We consider birth integrity violations as a critical vulnerability factor for developing birth-related post-traumatic stress disorders, postpartum depressions, or challenges in bonding and attachment of mother and child [123]. Mental or physical distress, a highly interventionist birth, or an unexpected course of childbirth can similarly cause trauma. Additionally, we see the necessity to research the physical and social consequences possibly deriving from birth integrity violations, like avoidance of subsequent pregnancies or impairments of sexual functions. Recently, a specific scale on postpartum, birth-related PTSD has been developed [124]. While there exist studies that demonstrate a significant association between birth experiences and postpartum depression [125], there is, to our knowledge, no scale that has operationalized a linkage between one’s birth perception (birth integrity) and postpartum depression. Understanding better if a violation of birth integrity can cause postpartum or birth-experience-related depression marks an important step towards recognizing, addressing, and improving mental postpartum health. Similarly, future research needs to better understand if and how violated birth integrity affects the bonding and attachment between the mother/parent and the child, the partnership, or negatively impacts the body functions.

The findings of this critical review (especially on the determinants of birth integrity) are closely related to the WHO recommendations on intrapartum care for positive childbirth experience [12]. However, both the WHO recommendations and the review results show that there is a lack of recognition that gender and power relations underlie the determinants of birth integrity (e.g., medical paternalism), birth integrity itself (e.g., gendered shame[126] or a normalization of violent acts against birthing bodies[127]), and are intertwined with the negative (mental) health consequences of motherhood myths (e.g., maternal guilt [128]). Conceptualizing research on birth integrity through a feminist research lens and methodology would not only make visible, analyze and tackle (gendered) health inequalities, but also contribute to a less hierarchical and more inclusive research process [129‐131]. To do justice to the complexity and multidimensionality of the birth integrity framework in methodological terms, mixed-methods appear to be an appropriate approach, allowing a combination of positivistic and interpretivist concepts as well as triangulation of quantitative and qualitative data [132]. One way to prioritize feminist ethics at all stages of research is to adopt a participatory action approach [133].

This review constitutes a critical appraisal of quantitative epidemiological research conducted in the field of maternal care conditions, maternal care provision, childbirth experiences, and perceptions of birth. We developed a new multilevel birth integrity framework that not only presents an overview on what is currently included in the studies’ concepts and measurement (determinants, birth perception, consequence of birth integrity violations) but additionally identifies gaps that may be addressed in future studies. Gaining a fuller picture of the reasons that evoke suboptimal treatment and birth integrity violations becomes relevant when planning interventions or health policies.

Despite a thoughtful and transparent review process, some measures might have been missed. Also, we did not appraise the studies’ quality since we were not focused on the study findings. After reading all the full texts, we decided to exclude research on Quality of Care since these studies mainly assessed how maternity care was rated through rating scales (e.g., “How would you rate the care you received? – “Excellent”, “good”, etc.). To identify determinants of birth integrity, we found these kinds of survey questions less insightful. Considering the high volume of quality-of-care studies, we decided to focus on the remaining 82 studies more relevant to our review aims.