Disrespect and abuse during facility-based childbirth in southern Mozambique: a cross-sectional study

Maternal mortality refers to deaths caused by complications from pregnancy or delivery. From 1990 to 2015, during the Millennium Development Goals (MDGs) era, the global maternal mortality ratio declined by 44% – from 385 deaths to 216 deaths per 100,000 live births, based on UN inter-agency estimates. Despite the fact that every region has advanced, the maternal mortality ratio is still very high in sub-Saharan Africa compared to the rest of the world [1]. Maternal mortality reduction remains a priority under in the new Sustainable Development Goals (SDGs). By 2030, the global community wants to reduce the global maternal mortality ratio (MMR) to fewer than 70 maternal deaths per 100,000 live births.

Global efforts during the MDGs era have largely focused on increasing antenatal care (ANC) coverage and facility-based childbirth as a key mechanism to reduce maternal mortality [2]. These efforts met with some success. There was much less emphasis on quality of care, although individual studies suggest that poor quality is limiting health gains [3, 4]. Improving quality of care, along with women’s experiences of care, has been highlighted as a key strategy to further reduce preventable maternal mortality and morbidity and achieve the health-related SDG targets by the World Health Organization (WHO) [5]. In 2016, WHO published new guidelines for improving quality of care for mothers and newborns in health facilities, which included an increased focus on respect and preservation of dignity. Experience of care is as important as clinical care provision in achieving the desired person-centred outcomes in the WHO framework for improving quality of care for pregnant women during childbirth [5]. Recent evidence suggests that many women experience mistreatment and are abandoned during childbirth in health facilities across the world, but the magnitude of the problem is unknown [6‐10]. An often cited framework for describing interpersonal aspects of care during labor and delivery are the seven domains of disrespect and abuse (D&A) defined in Bowser and Hill’s landscape evidence review, published in 2010: physical abuse; non-consented care; non-confidential care; non-dignified care; discrimination; abandonment of care; and detention in facilities [11]. Afterwards The White Ribbon Alliance spread the Respectful Maternity Care Charter: The Universal Rights of Childbearing Women, a statement grounded in the Universal Declaration of Human Rights [12].

The mistreatment of women during childbirth often occurs at the level of the interaction between women and healthcare providers but deficiencies in the health care system (e.g. lack of adequate personal and poor infrastructure) also contribute to its occurrence [13‐15]. The occurrence of disrespect and abuse in maternity care services may undermine women’s trust in the health system and deter them from seeking facility-based care for delivery [16]. Disrespect and abuse during childbirth is more and more being recognized as an indicator of poor quality of care and cited as a key barrier in achieving better maternal health outcomes [17].

Mozambique, with a maternal mortality ratio of 489 maternal deaths per 100 000 livebirths in 2015 and only 54% of births attended by a skilled birth attendant, is one of the priority countries for improving maternal health [18]. Several actions have been taken and progress is ongoing but slow. Recognising the importance of quality of care, since 2007 the MoH (Ministry of Health) of Mozambique has made humanization and patient friendly care during ANC and delivery one of its priorities [11]. Over time, the culture of promoting Respectful Maternity Care (RMC) has become more widespread in Mozambique and the MoH has transformed a selection of maternity wards into centers of quality and humanized Maternal and Newborn Health (MNH) care provision under the “Iniciativa Maternidade Modelo” (Model Maternity Initiative). Respectful maternity care is one of the essential packages of the model and includes respect for beliefs, traditions, and culture; the right to information and privacy; choice of a companion; freedom of movement and position; skin-to-skin contact and early breastfeeding; appropriate use of technology and effective lifesaving interventions; and prevention of violence and disrespectful care [11]. By 2017 the initiative was implemented in all hospitals (central, provincial and district) within the country and almost half of the health centres [unpublished report JHPiego & MoH]. However, no evaluation has been conducted so far from the perspective of users after introducing this model. Studies examining the prevalence of disrespect and abuse in maternity care in Mozambique are scarce, especially in comparison to other countries in the region like Tanzania, South-Africa and Kenya [12, 15, 19, 20]. Recognizing that poor experiences for women might lead to less deliveries in the facilities and affect the quality of care by several pathways, this study aims to assess the experience of women giving birth in hospital in different settings in Maputo City and Province, Mozambique.

In total 932 women gave birth during the study period and 628 women were approached for an interview. The main reasons that some women were not invited to participate were their bad health condition or that they went home very soon after birth (< 24 h). Of the 628 women that were invited for the study, 572 participated. The main reason for not participating when invited was being < 18 years old (n = 36); other main reasons were not interested or not feeling well. During data cleaning 52 data entries had to be removed because of poor quality and/or incompleteness, resulting in a final sample of 520 women (see Fig. 1). The final dataset did not contain missing data. Sociodemographic characteristics of the participants can be found in Table 1.

The prevalence of disrespect and abuse in our study was similar to the prevalence in other countries in the region: 23.51% in the central hospital and 79.82% in the district hospitals. Studies from Ethiopia, Kenya and Tanzania report D&A prevalence rates between 20 and 70% [9, 12, 17, 20]. However, it may be problematic to focus only on overall prevalence of D&A as an outcome, as this covers a wide range of forms of D&A that are very different in nature (e.g. injections without permission versus slapping and beating). In this study we found that more severe forms of abuse such as detention in the facility (for failure of paying) and physical violence (such as slapping) are almost non-existent in the study sites in Mozambique, while studies conducted in other countries often report much higher figures. For example, a systematic review of D&A in Ethiopia estimated a prevalence of 13% for physical abuse and 3.2% for detention in the facility [28]. The implementation of the “Iniciativa Maternidade Modelo” might have contributed to this positive result in Maputo City and Maputo Province and further efforts should focus on reducing abandonment (when the patient is being left alone) and disrespectful treatment (being shouted/scolded at), which continue to be prevalent.

The occurrence of D&A in maternity care services is often considered as a marker for quality of care: it might affect quality of care in both terms of discouraging women to deliver in facilities but also directly through inadequate monitoring during childbirth (eg. infrequent fetal monitoring during labour and delivery, or absence of a skilled provider for resuscitation of the newborn or to intervene in case of bleeding of the mother) [14]. Several participants in our study reported they delivered alone in the health facility, which imposes a serious risk on both mother and child. This might also indicate that the number of women delivering without a skilled birth attendant is probably under reported in the region. Mozambique is struggling with a weak health system, characterized by poor health infrastructure, shortage of providers and insufficient supervision [29]. Certain forms of D&A (abandonment and lack of privacy) we found to be common might be triggered or worsened by resource scarcity within the health system. The inadequate health system resources (lack of separate rooms, insufficient skilled providers) are probably a major contributing factor to certain forms of D&A and prevention should be oriented at this level.

Stigmatization and emotional abuse of women by providers (discrimination of primigravidas due to being unexperienced, slanderous remarks, lack of privacy regarding age) are also a prevalent problem in maternity care in Mozambique, according to our results. Discrimination and stigmatization of certain subgroups in health care settings have been studied mostly in high income countries. The problem has much less been studied in low income countries and has had a strong focus on minority groups and HIV stigmatization [30, 31]. The role of medical education (e.g. training to shape the attitudes of providers) in prevention of discrimination in health care settings may be well recognized, especially in high-income countries, but it is inadequately explored in the context of D&A [10, 32]. On a global level, countries with strong colonial roots often have a health system culture where providers morally instruct and educate their patients [14], which might contribute to the occurrence of D&A in Mozambique. This is in line with research that suggest that nurses’ and midwives’ inferiority in medical hierarchy and lack of power within their own professional and organizational structures might contribute to their need to dominate and control even more disempowered patients [13, 33]. When designing interventions to prevent D&A, a participatory approach with providers will be needed to explore the roots of their abusive behaviors towards women and identify ways to overcome them.

The overall prevalence of D&A in the district hospitals was much higher compared to the central hospital (79.8% vs 23.5%). Furthermore, we could demonstrate that women in the district hospitals more often experience a combination of different forms of disrespect and abuse compared to the central hospital. In our study the lower D&A prevalence in the central hospital compared to the district hospitals might be related to the fact that providers work under better circumstances in the central hospital. The central hospital is a teaching hospital with more supervision and control mechanisms than the district hospitals (e.g. extensive maternal death audits and academic meetings), and in general the maternity care system in Maputo City is better resourced than the rest of the country [34].

There is no consensus in the literature on the role sociodemographic and institutional factors play in the actual prevalence or reporting of D&A [9, 13, 19, 35]. Moreover, the influence of these factors might be very context specific [10]. This was confirmed in our study: sociodemographic factors played a different role in the central hospital compared to the district hospitals. In our study women with a secondary degree experience and/or report more forms of D&A [12]. This relationship might be related to the fact that these women expect higher standards of care and more easily recognize abusive behavior [12, 36]. Echoing the results of other studies, women in our study who had a delivery with complications reported more D&A [37]. Age was a protective factor against D&A in the central hospital. Several qualitative studies report that especially young and unexperienced women experience D&A due to power dynamics and low status [38, 39]. But they might also less easily recognize and report unacceptable behavior of providers, which might explain the contradicting findings in the literature. Nevertheless, more qualitative data from both women and providers will be needed to explore contributing factors regarding D&A in the Mozambican health system and specific context.

Labor companionship is a key component of providing respectful maternity care and has been included as one of the WHO standards for improving the quality of maternal and newborn care in health facilities [40]. Despite the benefits of a companion of choice throughout labor, implementation of this approach is not universal [41]. In Mozambique all maternities are officially obliged to allow birth companions since the introduction of the Model Maternity Initiative in 2017. However, in practice there are different rules depending on the provider (e.g. only women are allowed, no traditional birth attendants, only during the day, not able to switch) [experience in the field]. In most facilities in Mozambique it is strictly forbidden to allow male partners as birth companions during labor and delivery. This rule is partly linked with an overall lack of privacy on maternity wards (e.g. women deliver in beds next to each other in one room), which is perceived as more problematic when men are allowed to be present. However, as public facilities are improving more maternities now have separate rooms, and also in very small facilities privacy can often be guaranteed due to low numbers of births. Recognizing that the Respectful Maternity Care Charter and MoH policy state that women have the right to choose their own birth companion it is then contradictory to only allow female birth companions [42]. Also the World Health Organization recommends in their intrapartum guidelines that a parturient woman should be encouraged to have a supportive companion she trusts and can feel at ease with in labor and birth [5, 43].

Our study found that a majority of women were in favor of involving their male partner as birth companion and many also believe their partners would be in favor. The desire of women to involve their male partner should be taken into consideration by maternities and might be a motive to reconsider current restrictions, where privacy can be guaranteed. Another argument for allowing men on maternity wards is that research suggests that disrespectful care would be less frequent if partners were present [44‐46]. Birth companions in general are a protective factor against D&A [13, 19], and there is some evidence that bringing in the male partner might further protect the women against experiencing D&A. A study from Tanzania showed that male partners of women who experience abuse during labor or delivery find it easier to request better care or lodge a complaint than the women themselves [46]. Qualitative studies on experiences of men who have attended the births of their children in Malawi also showed that with a supportive environment and positive attitude of the midwives, it is possible to involve male partners during childbirth and for this to be a positive experience for both men and women [47, 48]. Further research is needed to explore the feasibility of allowing men in the delivery room in Mozambique and to examine potential strategies that create the ideal conditions for men to be present during labor and birth as the birth companion. It would also be interesting to examine whether involving men in maternity care might have an impact on the prevalence of disrespect and abuse during childbirth.

Offering modern contraception services as part of care provided during childbirth increases postpartum contraceptive use and is likely to reduce both unintended pregnancies and pregnancies that are too closely spaced [49]. It is recommended by the WHO standards for improving the quality of maternal and newborn care in health facilities [40] but very often neglected in studies examining quality of childbirth care [50]. Our study showed that both in the district hospitals and the central hospital the number of women receiving counselling about family planning was very low (17 and 9% respectively). For women with limited access to health care in facilities, delivery at a facility affords a unique opportunity to address their fertility intentions and need for contraception: it does not require a return visit that may be prohibitively expensive or inconvenient. Previous studies have shown that in the year following childbirth, many women want to postpone or avoid further births, but do not use a contraceptive method [51]. Offering family planning counselling before women leave the hospital might be an important and unique opportunity to protect women from an unplanned pregnancy, as only a minority of women (40-44%) return to the health facility for a postnatal care visit in Mozambique [18, 52]. Evidence has shown that discussing family planning before discharge from the maternity ward is an effective intervention to increase the uptake of family planning methods postpartum [53‐56]. Mozambique has a comprehensive strategy to reduce the unmet need for family planning including guidelines for integrating family planning counselling and provision of contraceptives across the health service including during the intrapartum period [57‐60]. However, increased attention is required to translate this policy into practice in order to improve uptake of family planning services in the post-partum period.

Currently there is a lack of standardized definitions, instruments, and study methods to quantify D&A in childbirth facilities, which affects the generalizability and comparability of results [7]. A validated instrument, taking into account the severity of each form of abuse, is needed if we want to continue to compare overall prevalence of D&A across different countries and/or regions. Furthermore some reported forms of D&A might not actually constitute mistreatment: for example, giving an injection without permission or stitching a first degree tear without anesthesia might be justified under certain medical conditions. A recent qualitative evidence synthesis also showed that RMC is a broader concept than merely the absence of mistreatment, although the two are intertwined [61]. While qualitative studies show that provider’s and women’s views on respectful maternity care are widely consistent globally, further research is needed to assess the validity and responsiveness of quantitative indicators to measure RMC [61].

Previous studies have shown that the factors that contribute to D&A in maternity care services and potential prevention measures are very context specific, which was confirmed in our study. We acknowledge that our study results cannot be generalized to other settings and further studies in different contexts in Mozambique are needed. Nevertheless we were able to show that D&A is a prominent problem in the country despite some significant quality improvement programs in maternity care over the last years.

We did not include minors, which is a limitation of our study. Data suggest that D&A and especially discrimination happen more often with adolescents, and our study found that increased age was protective against D&A. A follow-up study focusing on the occurrence of D&A in this specific group is recommended. In addition, our study was conducted inside the health institution, where social desirability bias can underestimate the occurrence of D&A. A community-based study might give women more freedom to express their feelings and report their experiences without fear, and eliminate this social desirability bias [62].

The overall prevalence of disrespect and abuse in our study was similar to the prevalence in other countries in the region but the more severe forms of abuse such as detention in the facility (for failure of paying) and physical violence (such as slapping) are almost non-existent. Occurrence of disrespect and abuse was much higher in the district hospitals. The majority of women were in favor of involving their male partner as birth companion and further research is needed to explore the feasibility of allowing men in the delivery room. Both in the district hospitals and the central hospital the number of women receiving counselling about family planning was very low. Investing in intrapartum counselling for family planning is currently a missed opportunity for improving the uptake of contraception in the country.