Background

Increasing the rate of facility-based childbirth is one of the main pillars in reducing global maternal and neonatal morbidity and mortality [1]. However, many women across the globe experience mistreatment during facility-based childbirth [2,3,4,5,6,7,8,9]. Mistreatment is used as an indicator of the quality of care and affects the mother’s decision to seek and utilize maternal health services for childbirth [10]. Research recently has grown on this global and significant public health issue.

There is a lack of a standardized definition and methodology that addresses mistreatment. Several publications have proposed definitions and conceptual frameworks for understanding the concept [11,12,13]. In 2015, The World Health Organization (WHO) researchers; Bohren and colleagues presented a new phrase, “mistreatment of women during facility-based childbirth.” They conducted an extensive mixed-method systematic review to develop a standardized typology of what constitutes mistreatment during childbirth using 65 studies from 34 countries [13].

The WHO team identified seven typologies of mistreatment of women during childbirth: “physical abuse, verbal abuse, sexual abuse, stigma and discrimination, failure to meet professional standards of care, poor rapport between women and providers, and health system conditions and constraints” [13].

With the growing recognition of mistreatment of women during facility-based childbirth, the WHO saw a clear need to develop standardized, evidence-based measurement tools that can be applied at a global level to define, measure, and prevent mistreatment [14]. As a result, the WHO initiated a multi-country research study to develop and validate two tools (labor observation and community survey) to measure the mistreatment of women during childbirth in health facilities in four countries: Ghana, Guinea, Myanmar, and Nigeria [14]. It was a two-phased, mixed-methods study, with phase one aiming to develop and validate the two tools that would provide data on the burden of mistreatment comparable across settings and over time [14]. Phase two aimed to apply these tools and report the prevalence of mistreatment in the four countries [15].

Mistreatment of women throughout the birthing process is associated with several factors [16]. These factors include age, number of previous births, attending antenatal care, time of delivery (day or night), method of delivery (vaginal vs. cesarean), marital status, facility sector (governmental or private), education level, economic situation, type and sex of birth attendant during childbirth, birth companion, race, ethnicity, and immigration status [2, 5, 15, 17,18,19,20,21,22].

The exposure of women to mistreatment during childbirth can lead to severe and adverse health outcomes for the mother or baby [23,24,25]. These may include feelings of fear, sorrow, disrespect, insecurity for mothers and the expected baby, distrust and lack of confidence in the health care providers, a sense of weakness and powerlessness, and postpartum depression [23, 26, 27].

However, women are expected to encounter skilled health workers who are competent, respectful, and caring. In addition, they desired to receive timely information, effective communication, and physical and emotional support. Women emphasized that all the above expectations and desires are needed for positive health outcomes during facility-based childbirth [28, 29].

Although this problem is faced by women worldwide, there is limited data on their mistreatment during facility-based childbirth in Arab countries. Furthermore, it is unknown what terminologies, typologies, and methods are used to address this topic in this part of the world.

Review objectives

This systematic review aimed to identify the status of the conducted studies concerning mistreatment, along with the terminology, typologies, and methods used. As well as to try and estimate the prevalence of mistreatment of women during facility-based childbirth. This would provide an evidence-based strategy to alleviate this problem in Arab countries, improve women’s childbirth experiences in health facilities, and reduce overall maternal mortality regionally.

Research methodology

This systematic review was conducted following a protocol registered on PROSPERO with ID number CRD42020182806 [30].

Search strategy

The search was conducted using the accessible databases: PubMed, Embase, and CINAHL. The search was done in May 2020 and was limited to studies published in English and Arabic, with no restrictions on the publication year. The search included observational studies that reported the prevalence of women's mistreatment throughout the birthing process in Arab countries (i.e. cross-sectional and cohort). After doing a literature search [4, 10, 13, 31,32,33,34,35,36,37,38,39,40,41] to ensure that terms were not missed; keywords were determined and can be found in Additional file 1.

Study selection and inclusion criteria

The inclusion criteria included studies that satisfied the following:

  1. 1.

    Women of reproductive age [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49], giving or giving birth in a health facility, experienced mistreatment during childbirth.

  2. 2.

    Arab Countries (Jordan, Palestine, Syria, Lebanon, Morocco, Mauritania, Algeria, Tunisia, Libya, Sudan, Somalia, Egypt, Saudi Arabia, Yemen, Oman, Qatar, Bahrain, Kuwait, the Comoros Islands, Iraq, Djibouti, and the United Arab Emirates),

  3. 3.

    Cohort or cross-sectional study designs

  4. 4.

    Articles containing experiences of mistreatment reported by women or observed by trained professionals during labor and delivery in a health facility from health workers

  5. 5.

    Articles related to improving birth outcomes, effective communication between women and healthcare providers, positive/negative birth experience, or birth care.

Two reviewers independently screened the studies’ titles and abstracts for inclusion. The full-text screening was also done by two reviewers independently. Both screenings were done using Covidence portal for reviews. Database searches were downloaded to Endnote and then uploads to Covidence. Conflicts during the title and abstract and full-text screenings were resolved through consensus. If an agreement was not reached, a third reviewer was consulted who made the final decision. The reference lists of the included studies were then hand-searched for any other relevant studies. Articles found by reference list were screened in the same way as the other articles.

Data extraction

The data extraction sheet was built, piloted, and agreed upon by the research team.

The extracted information included the names of the authors, journals, the publication year, the study title, the aim of the study, the study design, country, inclusion/exclusion criteria, sample size, method of data collection, the primary outcomes of each study, outcome measurements, and the results (i.e. prevalence, mean score, number of women reporting the outcome of interest). The data extraction was also conducted by two reviewers independently, and conflicts were resolved by consensus.

Risk of bias (quality) assessment

The risk of bias was completed by two reviewers independently using the 10-item tool developed by Hoy et al. to assess the risk of bias in prevalence studies [42]. This tool was user-friendly and demonstrated high interrater agreement. It included assessment for the following domains: selection bias, non-response bias, measurement bias, and bias related to analysis [42]. Disagreements were resolved between the two reviewers through consensus.

Data synthesis

A flow chart was used to show the number of studies remaining at each stage of the selection process. A narrative and descriptive summary tables of the findings were provided for the included studies structured around the articles’ general data, the terminology, the measurement tools, and the types used. The analysis was conducted based on the evidence-based typology developed by Bohren et al. [13] as a guide to try to estimate the prevalence of mistreatment of women throughout the birthing process in health facilities in Arab countries. This evidence-based typology was used because the authors proposed an inclusive definition for mistreatment. They defined a broad scope of categories that consider different sources of mistreatment that may stem from both intentional or unintentional actions either by healthcare providers or from conditions within the health facilities and system [13]. The prevalence was calculated as the average prevalence of all the items included within the outcome of interest.

Results

Search findings

The database and hand search of reference lists resulted in 183 studies with nine duplicate studies, giving 174 studies. The title and abstract screening yielded 32 studies that were eligible for full-text extraction, resulting in 11 included studies [43,44,45,46,47,48,49,50,51,52,53]. The list of full text excluded with reasons is presented in Additional file 2.

Figure 1 presents the flow chart of the included studies.

Fig. 1
figure 1

Flowchart of study selection

Full size image

Characteristics of the included studies

More than half of the studies were conducted between 2010 and 2019. Six out of the 11 studies were published between 2005 and 2014, and the remaining five were published between 2015 and 2020. The majority of the 11 studies were published in journals specific to reproductive health, obstetrics and gynecology, birth, and nursing. All studies were cross-sectional studies. The 11 studies were conducted in seven out of the 22 Arab countries. These countries were Egypt, Jordan, Iraq, Lebanon, Morocco, Syria, and Yemen. The majority of the 11 studies were conducted in health facilities. Specifically, four were conducted in hospitals, two were in clinics, and one was in both settings. The rest were conducted in the community. The data collection method for ten of the 11 studies was an interviewer-administered questionnaire by trained fieldworkers, and one was a self-administered questionnaire and limited to literate women. Sample sizes were relatively large, with the smallest number of participants being 250 women.

The majority of the 11 studies had aims directly related to mistreatment. However, the aim of three of the studies was indirectly related: women’s preferred location of childbirth in case of future pregnancies, trends in postpartum care, and women’s preferences for the type of birth attendant and place of delivery. The characteristics of the included studies are shown in Table 1.

Table 1 Characteristics of the 11 Studies
Full size table

The women included in each study varied in terms of time and place of the interview. Participants were grouped into three main categories based on measuring the outcomes. These were women in postpartum wards before they left the hospital; second, women who had a healthy newborn up to 1 year after delivery in clinics; and third, women with childbirth experience without specifying the period.

Findings for the terminology and tools used in measuring mistreatment

There are currently multiple phrases being used to express mistreatment, such as “disrespectful care,” “disrespect and abuse,” and “obstetric violence.” None of the included studies yielded any of the terms. Indirect terms were used to reflect mistreatment. These terms fall under the following main categories: women’s satisfaction, perception of control/authority during childbirth, postpartum care, mothers’ experiences of care related to client-provider interaction, mother-infant proximity, other terminologies, and irrelevant terminologies.

Women’s satisfaction was reported as the primary outcome in six studies [43, 47, 50,51,52,53] and three as the secondary outcome [43, 44, 53]. Satisfaction was measured differently in each of these six studies. For example, “women’s satisfaction”, “dissatisfaction with intrapartum care,” “satisfaction with childbirth experience,” “overall maternal satisfaction with delivery services,” “mother’s satisfaction with delivery care,” and “overall satisfaction of women with the communication of midwives and physicians during labor and birth.” In addition, each of these studies operationalized the satisfaction definition differently. Consequently, each study used a different measurement tool for estimating “satisfaction.”

Women’s Perception of Control/Authority during Childbirth was used in four of the 11 studies as the primary outcome [47,48,49, 51]. These outcomes included “perceived control during childbirth experience,” “women’s perception of control,” “women’s perceived authority at birth,” and “women’s authority during childbirth.” The authors also did not utilize a clear conceptual framework for this term; however, they determined the operational definition differently in each of these studies.

Postpartum Care was used in one of the 11 studies [46]. The author used a clarified conceptual definition for the term “postpartum care” and operationalized it using the “demographic health survey (DHS)” that was modified to include information on postpartum and post-natal care for women delivering in health facilities.

One study measured mothers’ experiences of care related to client-provider interaction [53], and another measured mother-infant proximity [49], and both were parts of the questionnaire used in this study.

Other terminologies describe mistreatment used and measured in the studies that could not be classified in the different terms above. These were “privacy sensation during hospital stay” [52], “own choice of birth attendance” [49], “presence of birth support/companion” [49, 53], “not talked to any health professional about how they felt about what happened during labor and birth,” and “attendance of anyone that they did not want to be there” [50]. There were no specific indications for measuring these outcomes since they were extracted from tables within the articles.

There was diversity in the terms and measurement tools used. Each term has an operational definition, making the comparison between these articles difficult.

Considerable variation was observed among the tools used to measure mistreatment. Four studies reported developing their own tools; three were tested/piloted [44, 51, 53], and one was not tested/piloted [43]. Four studies were translated to Arabic [46, 47, 46,49,47], one study reported only adaptation [52], one study reported only testing[48], and one study adapted and tested/piloted a tool used in an Arab country [45].

For this review—despite having only three of the 11 studies reporting on validation [46, 50, 51]; it was considered that both adaption with testing/piloting or development with testing/piloting, as validation as well.

Findings for the types identified in measuring mistreatment

The evidence-based typology for mistreatment [13] was used to guide the analysis of this paper. The findings of the 11 studies were mainly classified within two typologies “poor rapport between women and providers” and “health system conditions and constraints.” Table 2 presents typologies with detailed information. Some outcomes were classified under more than one typology; for example, "women’s satisfaction” and “satisfaction with childbirth experience” could be classified under both the sixth typology and the seventh typology, because the scales used to measure both of these outcomes cover several dimensions. Also, Table 2 shows that “satisfaction with childbirth experience” was classified under the fifth category and the sixth and seventh categories because it has one item directly related to the physical examinations and procedures sub-category.

Table 2 Summary findings for the types identified, the findings of the studies and the overall range of prevalence
Full size table

Findings for estimating the prevalence of mistreatment

This paper aimed to present prevalence estimates of mistreatment of women throughout the birthing process in health facilities in Arab countries. However, this was not possible due to high heterogeneity in the 11 studies. Therefore, the prevalence for each outcome of interest within each sub-category (either reported or calculated) was used to inform the overall prevalence range for each sub-category, as shown in Table 2.

The overall prevalence of women satisfied with the communication throughout the birthing process ranged between 24.4% and 74.7%. The overall prevalence of women who had supportive care at any point throughout the birthing process ranged between 4.14% and 16%. The overall prevalence of autonomy in women throughout the birthing process ranged between 35 and 36%. The overall prevalence of women who were satisfied with resources of the health system birthing process ranged between 55.5% and 89%. Finally, the overall prevalence of women who were satisfied with the health system's policies throughout the birthing process ranged between 54.4% and 78.5%.

Risk bias assessment

The findings mentioned above were all based on the risk of bias judgments that are shown in Table 3. Six out of the 11 studies have a low overall risk of bias, while the remaining five showed a moderate overall risk of bias. The majority of the studies were judged to be high risk for the following items on the Hoy checklist [42]: bias for the target population, the sampling frame, and the random selection. As well as all studies were judged to be low risk of bias on the checklist for the case definition, the standard mode of data collection (i.e. data were collected from all the subjects using same mode), and the numerators and denominators (i.e. appropriate calculations for the parameters of interest).

Table 3 Risk bias assessment as per Hoy's criteria
Full size table

Discussion

Estimating the prevalence of mistreatment of women

It was difficult to estimate the prevalence of mistreatment of women throughout the birthing process in health facilities in Arab countries because the tools and definitions were not consistent. Consequently, this review reported a broad range of prevalence for each sub-category. A similar finding was reported by Sando and colleagues’ systematic review, which demonstrated that the lack of standardized definitions, instruments, and study methods used in measuring disrespect and abuse affected the comparability between results and introduced systemic errors in reported prevalence estimates [16].

Moreover, using a single methodology and definition in the WHO study to estimate the prevalence of violence against women across ten countries and cultures significantly reduced the challenges faced by earlier conducted research. It also enabled comparability with the newly completed international research initiatives using the same methodology and definition. This resulted in gaining a more comprehensive picture of violence against women worldwide [54].

The terminology used in measuring mistreatment of women

The terms used in the identified studies were not synonyms for the mistreatment of women. However, they were considered a proxy for mistreatment since they did not directly measure it. The terms found were more related to childbirth experience, the attitude of health personnel, and physician–patient relations. General terms such as “obstetric violence” and “disrespect” were not found in quantitative studies in Arab countries. A qualitative study conducted in 2017 used the terms “disrespect and abuse” and “mistreatment.” [55]. This indicates that such terms are newly introduced in Arab countries and are still being explored. However, they have yet to develop tools to measure them.

The tools used in measuring mistreatment of women

The tools used to measure mistreatment varied widely in terms of the content, length, purpose, validation, and outcomes. This resulted in incomparable results between the included studies. As well as, indicating the essential need for utilizing a standardized tool in quantitative studies that are adequately adapted and validated. This finding agrees with the result of a systematic review that emphasized the necessity of providing a standard tool to accurately estimate the prevalence of women’s childbirth experiences and enable comparability [56].

Several studies that measured disrespect and abuse based on Browser and Hill categories employed different definitions of a standardized operational definition. This resulted in differences in prevalence estimates, thus facing the same problem of having incomparable results and preventing the possibility of conducting a meta-analysis [5, 57, 58]. This finding agrees with Nilver’s systematic review that even when researchers and clinicians use different instruments to measure the same construct of interest, it will be difficult to compare and statistically report the results of systematic reviews [59]. This emphasized that standardized terminology should be accompanied by the standardized tool.

The typology used in measuring mistreatment of women

The types of mistreatment found did not exceed two main typologies out of the seven from the evidence-based typology- which are “poor rapport between women and providers” and “health system conditions and constraints.” It was apparent that the types used were found to be marginal in comparison with the ones mentioned in the evidence-based typology that the WHO developed. A theory behind why other types were not used was that researchers in Arab countries did not consider the remaining typologies in their research about childbirth experience as a priority to improve the quality of care. The result of not considering other types yielded tools that were not comprehensive enough to include additional domains of mistreatment or measure childbirth experience.

Other typologies were found in the qualitative studies [55, 60,61,62,63,64]. These studies included additional types such as: “physical abuse” [55, 61], “verbal abuse” [55, 60, 61, 60,63,61], “stigma and discrimination” [55, 60], and “failure to meet professional standards of care” (i.e., “neglect and abandonment” [55, 61, 63, 64], “lack of informed consent process” [62], and “physical examinations and procedures”) [62].

It seems that when women were allowed to speak their minds about their childbirth experience, they reflected on these four main typologies due to the negative impact it had on their childbirth experience [23]. Consequently, these typologies should be explicitly explored in future quantitative research in Arab countries. Also, the tools used in the quantitative studies should be comprehensive for all the typologies mentioned in the evidence-based typology. They should as well cover all dimensions of the childbirth experience.

Moreover, it was hard to find the typology of “sexual abuse” in health facilities in studies conducted in Arab countries because obtaining information about “sexual abuse” or “sexual violence” in Arab societies is not easy. Talking about sex, in general, is a sensitive topic in the Arab world, and cultural values reinforce silence and inaction around such events [65, 66].

Methodological issues within the included studies

The timing for interviewing women to ask about mistreatment varied between studies as directly after the following childbirth or later up to a year after delivery. The timing of the interview is crucial and essential. The literature suggested that recall is more accurate in the postpartum period than immediately following childbirth when women are physically exhausted and have no time to mentally process the events that occurred during childbirth [16, 67]. Additionally, exit surveys on the facility's grounds may induce courtesy bias or reluctance to disappoint researchers by reporting negative experiences, especially when the interviewers were perceived to be affiliated with the same facility [5, 6]. Women may fear their opinions affecting their future use of services at the same facility, even if conducted privately [6, 68]. After a long time after birth, interviewing might over or underestimate the prevalence due to induced recall bias [8].

There were no published articles before the year 2005. Women’s experience during facility-based childbirth is a new and neglected area of research in Arab countries [69]. Since then, several studies in Egypt, Lebanon, Syria, and occupied Palestinian territory have been conducted to understand the experience of facility-based childbirth from a women’s point of view [69, 70].

Strengths and methodological considerations of the current review

This systematic review was the first to be conducted among Arab countries. It showed the contribution of Arab countries to the research of mistreatment of women during facility-based childbirth. It also showed that this topic is still new and has not been tackled quantitatively yet. It is recommended that future researchers increase their research in this area using standardized definitions, tools, and study methods to accurately measure the mistreatment of women to enable comparison across time and between countries. It also prepares future researchers to face some challenges when using the WHO standardized tool because some terms and typologies have not been used in Arabic countries yet.

The hand searching yielded 17 new articles after excluding all reports, qualitative and randomized control trial designs, and doctorate dissertations. Moreover, the purpose behind the hand searching was to ensure that no relevant articles were missed during the search. In this paper, grey literature and unpublished reports were not screened because the focus was on published articles only.

This current review focused on the article's content rather than the quality of the research published to be inclusive.

Recommendations

This review recommends that future quantitative studies in Arab countries use standardized and comprehensive terminology for the mistreatment of women. Also, a standardized tool that covers all aspects of mistreatment. In addition to a standardized timing, data collection method, data collection setting, and inclusion criteria to enhance comparability between results and allow pooling when estimating the prevalence. Still, this tool needs to be validated and adapted to suit Arab culture and context.

It is also recommended to update this review specifically to include qualitative studies related to mistreatment of women in Arab countries to provide sufficient information for a more holistic understanding of the mistreatment of women and further investigate what other terms and types would result.

Finally, it is recommended to use the evidence-based typology developed by WHO researchers since it has a clear definition for the mistreatment of women during facility-based childbirth and a standardized tool and methodology.

Conclusion

The quantitative studies in Arab countries did not directly tackle the mistreatment of women throughout the birthing process in health facilities, and the resulting terms were a proxy for the word mistreatment. However, estimating the prevalence for these proxy terms was hard to obtain due to the heterogeneity of the terms used, tools used to operationalize the terms, inclusion/exclusion criteria, and the methodological characteristics of the conducted studies.