Background

As clinical indicators of maternal health improve worldwide, increased emphasis has been placed upon quality of care to improve maternal health outcomes, reduce maternal mortality, and ensure the fulfilment of women’s sexual rights and reproductive rights [1]. There has been a growing body of evidence pointing to the prevalence of disrespect and abuse (D&A) during facility-based childbirth. D&A is a violation of women’s rights to respectful maternity care (RMC), has a deterrent effect on the utilization of facilities for childbirth, and negatively impacts maternal and newborn health [2,3,4,5,6].

D&A is a complex phenomenon that includes a degree of subjectivity, as women, families, healthcare workers, and community stakeholders might define it differently. Furthermore, D&A is deeply imbedded within clinical practices and health systems [7]. This complicates intervention efforts, which must target the various contextual elements and practices that precipitate D&A [8]. Bowser and Hill described seven categories of D&A during childbirth: physical abuse, non-consented care, non-dignified care, discrimination, abandonment, and detainment in facilities [9]. In 2014, Freedman et al. defined D&A as the “interactions or facility conditions that local consensus deems to be humiliating or undignified, and those interactions or conditions that are experienced as or intended to be humiliating or undignified,” a definition which captures individual instances of D&A but also the systemic factors contributing to disrespectful or abusive environments [10].

In Latin America, D&A is recognized as a component of violencia obstétrica (obstetric violence), a concept which has gained traction throughout the region. Obstetric violence (OV) was first codified into law in Venezuela in 2007 and has subsequently been decried by activist movements, legislative advances, and scholarly discourse in other Latin American countries [11]. OV frames disrespectful and abusive practices within the context of structural inequities and criticizes the power hierarchies imbedded in medical obstetrics. Though these analyses are critical to promoting RMC, the present article will use the term D&A instead of OV to focus the scope of discussion.

Despite the robust body of evidence for the prevalence of D&A worldwide [9, 12,13,14], less emphasis is placed upon identifying the drivers of the phenomenon. Country-specific analyses and solutions are needed, as successful interventions will vary greatly by local culture and context. Additionally, few existing studies of D&A pay special attention to identifying—and addressing—D&A as experienced by vulnerable patient populations.

This report is part of a larger study funded by the Fulbright Commission in Colombia, “Maternal Health Care: The Indigenous Migrant Perspective in Medellín, Colombia,” aimed at addressing these gaps to identify the manifestations and drivers of D&A as experienced by internally displaced indigenous women in Colombia. Previous studies have identified D&A within Colombian women’s experiences of hospital-based childbirth (though no true measurements of prevalence have been obtained) [15,16,17]. However, these studies focus mainly on making this phenomenon visible. There is a lack of data identifying the concrete drivers of D&A, which may be stunting intervention efforts.

Previous studies have indicated that migrants and refugees are particularly vulnerable to disrespectful and abusive treatment during childbirth [5, 18]. Others note that indigenous women experience language and cultural barriers that prevent the provision of RMC and facilitate D&A [19]. Colombia has faced one of the world’s most severe forced migration crises due to conflict and violence. Despite the 2016 peace agreement signed between the country’s largest armed group, the Fuerzas Armadas Revolucionarias de Colombia (FARC) and the Colombian government, there were 139,000 new displacements in 2019, bringing the total number of internally displaced persons in the country to over 5.5 million [20]. Internally displaced women of indigenous descent are an especially vulnerable group in Colombia: they are more likely to be of low socioeconomic status, are at increased risk of post-displacement trauma given historical attachment to their land and face unique cultural barriers to receiving health care. Even those indigenous women who migrate to urban regions for economic or educational reasons, rather than being forcibly displaced due to conflict, face linguistic and cultural barriers [21, 22].

The present study is the first of our knowledge to identify drivers and potential interventions to eliminate D&A against an indigenous community in Colombia. The indigenous group Embera (which in the Colombian department of Antioquia includes the subgroups Chamí, Eyabida, and Dobida) represent the largest proportion of indigenous persons (nearly 44%) in Medellín, the capital of Antioquia and the setting of this project [23, 24].

Ratcliffe’s 2013 framework, based on Bronfenbrenner’s ecological systems model, illustrates interactions between individual and systems-level factors that impact D&A [25, 26]. Under this model, D&A is affected by four interrelated subsystems: the individual and community, the healthcare professional, the healthcare facility, and national systems. Each level presents a potential target for intervention but does not exist in isolation from the other subsystems. This paper uses this framework to describe the perceived drivers of D&A against Embera women during childbirth in Colombia, and potential interventions to mitigate mistreatment.

Methods

Qualitative data in the form of semi-structured, face-to-face interviews were collected between November 2019 and March 2020. Participants included 10 indigenous Embera women who had migrated to Medellín, Colombia, either forcibly due to the armed conflict or for economic or educational reasons and had given birth in a health facility or at home, 5 community stakeholders (including indigenous leaders and experts in indigenous health), and 6 healthcare workers (physicians and nurses). See Tables 1 and 2 for a detailed description of the participants. Participants were recruited by the research team with the help of community leaders at the Organización Indígena de Antioquia (OIA) and scholars at the Universidad de Antioquia Facultad Nacional de Salud Pública (FNSP) through the snowball effect. Throughout the interview period, data were preliminarily analyzed for emergent themes and a saturation matrix was used to determine the point at which thematic saturation was reached and thus, determine the point for interview completion [27, 28]. To mitigate the potential for selection bias, participants were recruited only on the basis that they be indigenous women with an institutionalized childbirth experience, not based on their prior identification as having experienced D&A [29]. Interviews lasted roughly 45 min and were facilitated in Spanish by the principal investigator. Interviews with the women were also accompanied by a female Colombian indigenous psychologist, who served as a trauma-informed cultural liaison. All the women interviewed spoke conversational Spanish, but the psychologist aided in interpreting certain phrases. A semi-structured interview guide with open-ended, non-leading questions was used to mitigate confirmation bias [30].

Table 1 Participant characteristics (healthcare workers and community stakeholders)
Full size table
Table 2 Participant characteristics (indigenous women)
Full size table

Interviews with indigenous women discussed (1) cultural practices and ancestral knowledge surrounding maternity care and childbirth, (2) experiences of D&A during childbirth, (3) births in an indigenous community versus those in a health facility (if applicable), (4) perceived drivers of D&A, (5) indigenous women’s reasons for avoiding skilled maternity care and (6) potential interventions to improve childbirth experiences in the hospital setting. Interviews with community stakeholders and healthcare workers discussed (1) the cultural and political context of maternity care in Colombia, indigenous versus biomedical health systems, and providing intercultural care, (2) healthcare workers’ experiences witnessing D&A, (3) perceived drivers of D&A against indigenous women and (4) potential interventions to mitigate mistreatment. Interviews explored participants’ experiences without imposing medical definitions of phenomena such as D&A to mitigate chronological bias. The specific manifestations of D&A experienced by the women in this study and deterrents of skilled maternal care utilization are described elsewhere.

Audio recordings were transcribed and coded thematically in Spanish using NVivo Qualitative research software [31]. A qualitative content analysis approach was taken to identify the factors contributing to D&A toward indigenous women and potential solutions [32, 33]. A hierarchically organized codebook was developed deductively based on the Ratcliffe model and inductively based on other themes that emerged. All authors reviewed the coded transcripts and discussed ambiguities or discrepancies. Codes were organized according to emergent themes and analyzed using the Ratcliffe framework. Following data analysis, a meeting was held with study participants to validate the results obtained.

Results

The results here are presented in two sections: first, the perceived drivers (factors contributing to the incidence) of D&A and secondly, proposed solutions or areas where interventions might be applied. Illustrative quotes are found in Table 3.

Table 3 Perceived drivers of D&A
Full size table

Drivers of disrespect and abuse during childbirth

Individual or community-level drivers

Normalization of D&A

Participants identified the normalization of D&A as contributing to its incidence. Women described mistreatment during childbirth as somewhat expected or routine. Healthcare workers also described the normalization of D&A (especially verbal abuse) in the medical field—one obstetrician-gynecologist (OB/GYN), for example, recounted that his proposed intervention to combat D&A was immediately rejected for not capturing a “real” phenomenon.

Lack of autonomy and empowerment

Women, healthcare workers, and stakeholders alike agreed that very few indigenous women are adequately empowered to identify D&A or demand RMC. Women were unaware of their rights to RMC and noted that demanding better treatment was rare among indigenous communities who are unfamiliar with health systems and facilities. Several participants initially denied having experienced D&A, but later indicated having experienced discomfort, humiliation, or shame during childbirth. Participants also said that women’s rights, especially during childbirth, are seldom discussed in indigenous communities.

Even for those women who did identify mistreatment while in a health facility, fear and pain during labor prevented them from focusing on their quality of care. Furthermore, if they did wish to speak against D&A, language barriers prevented them from doing so. Multiple women reported that there were no accountability mechanisms, such as incident reports, that they could use to denounce D&A. Healthcare workers contradicted these claims—accountability mechanisms within health facilities do exist, but indigenous women were unaware of them or unable to access them.

Lacking antenatal care

Healthcare workers noted that indigenous women are more likely than non-indigenous Colombian women to arrive at a hospital in labor without having received any antenatal care (ANC). Lack of ANC contributed to poor rapport between woman and clinician and led women to feel overwhelmed and confused by the hospital birth process. Furthermore, as births without ANC are higher-risk, healthcare workers often became frustrated with women for not seeking care, chastising them for not “taking care of themselves” or even blaming them for poor outcomes.

Clinician-level drivers

Clinician prejudice

Women felt that clinicians’ biases against women or indigenous communities affected the care they received. They described being “looked at differently” and shamed for their culture or beliefs. This prejudice manifested through verbal abuse (disrespectful or derogatory comments) and the exclusion of traditional medicine (for example, prohibiting a woman from undergoing a sacred ritual or bringing the placenta home after birth, or barring traditional midwives from attending births). Healthcare workers agreed that indigenous women were subject to increased verbal abuse but attributed the exclusion of indigenous culture to facility and health system constraints. Participants also noted that even non-indigenous women were subject to D&A due to gender discrimination.

Linguistic or cultural barriers to communication

One of the most frequently cited drivers of D&A mentioned was barriers to communication. Some indigenous women in Medellín have limited Spanish proficiency, but virtually no clinicians speak indigenous languages. Furthermore, hospitals were never equipped with interpreters. Healthcare workers are thus unable to explain the birthing process to a woman, answer her questions, or ensure that care is delivered with cultural humility. Additionally, clinicians indicated that being unable to communicate with patients led to “dehumanization” of the patient.

In addition to language, cultural differences pose other communication barriers. Most indigenous women have little experience with healthcare systems. Medical jargon is particularly disconcerting. The hospital birth process is markedly different from births in indigenous communities, such that women are often unsure how to or to whom direct their questions.

Lack of understanding of indigenous culture

Participants described how childbirth is a meaningful event in indigenous communities and an important vehicle to preserving culture and ancestral knowledge. Indigenous women prefer to give birth in a kneeling or squatting position, be accompanied by a traditional midwife or healer, partake in sacred rituals and baths, and be surrounded by members of their communities.

Women complained that clinicians did not understand indigenous culture and lacked the context needed to provide intercultural maternity care. Healthcare workers and community stakeholders echoed these sentiments, indicating that neither medical education nor workplace-specific training adequately equipped clinicians with the knowledge and tools needed to provide patient-centered, culturally respectful obstetric care.

Clinician training and medical culture

Healthcare workers described disrespectful or abusive behaviors as routinely practiced in clinical settings. They highlighted an underemphasis on RMC in both medical training and clinical practice, noting that clinicians are neither trained in the principles of RMC nor how to implement it for diverse patients. In fact, the mention of D&A is often taboo and can alienate clinicians. Several participants also discussed the hegemonic role played by clinicians during childbirth, whose word is seen as so superior that it can never be questioned, especially by the patient.

Clinician burnout and demoralization

Maternal health facilities in Medellín are overcrowded, understaffed, and as described by many healthcare workers, “on the brink of collapse” due in part to recent closures of several maternity wards and an increase in patient volume. These conditions lead clinicians to feel burnt out, frustrated, and demoralized. Moreover, they have little time to provide individualized, culturally respectful care to their patients.

Facility-level drivers

Inadequate infrastructure, space, and human resources

Women described maternity wards, which often contain 8–10 beds for women in labor in a single room, separated only by thin curtains, as uncomfortable and lacking privacy. Several women who denied experiencing D&A themselves felt fearful or apprehensive upon hearing insults directed towards other women. Additionally, this lack of space makes it impossible for a woman’s family or friends to attend a birth. Indigenous women are particularly impacted by these restrictions, as births in their communities are typically attended by close family and a traditional midwife.

Additionally, clinicians described insufficient infrastructure and lack of human resources, noting the logistical difficulties to providing RMC in understaffed hospitals with large patient volumes.

National health systems-level drivers

Lack of laws or policies

Colombia, unlike other Latin American countries, does not have a law against D&A. Without a clear national policy, redressing disrespectful or abusive behavior is challenging, if not impossible. Furthermore, the lack of clear policy contributes to the normalization of D&A and makes it more difficult to integrate such issues into medical education or facility standards.

Devaluing of RMC

In addition to RMC lacking from clinical curricula, participants described RMC as a low priority of the Colombian health system. Intercultural care is similarly devalued. For example, many health insurances do not reimburse care given by traditional healers or midwives, which has contributed to the decline of these practices. Healthcare workers also connected inadequate infrastructure to the need for a systems-level intervention building the capacity of individual facilities and clinicians to provide respectful, culturally informed care.

Solutions to D&A

Drivers of D&A are “trigger points” at which interventions might be applied [34]. Many study participants suggested potential solutions to D&A based upon these triggers, which are described below using the same ecological systems framework to underscore that interventions must be multifactorial to address the complex, interrelated drivers of D&A.

Individuals and communities

D&A must be made visible, within the medical community, indigenous culture, and society, for any intervention to be effective. Several women underscored the need for educational campaigns and rights-based discourse aimed at increasing their autonomy and empowerment to demand respectful, intercultural care. Additionally, interventions to increase ANC among indigenous women are needed, as it presents an opportunity to provide women with information that will increase readiness for childbirth, develop rapport with clinicians, and reduce risk of poor birth outcomes.

Clinicians-focused interventions

Participants called for a shift in medicine, beginning in medical education, to emphasize patient-centered care and understanding that RMC is complimentary, not contradictory, to evidence-based practice. Participants emphasized their desire for clinicians to listen to patients and recognize women as active participants in their births. Participants also called for healthcare workers to be trained in indigenous culture and beliefs (including working collaboratively with traditional midwives) to provide more appropriate care for this population.

Facility-focused interventions

Participants argued that increased intercultural care cannot be accomplished through clinician education alone. Rather, indigenous communities must be active participants in the design and execution of any intervention, especially as RMC varies with culture. Several women suggested intercultural trainings be held between hospital administrators, indigenous leaders, healthcare workers, traditional healers, and indigenous women to help bridge this gap.

Hospital infrastructure presents an opportunity for intervention against D&A. Ideally, each woman would have a private room with space for birth companions, healthcare workers would be well-compensated and not overworked, patient volume would be small enough to allow individualized care, and interpreters would always be present. Despite the financial and logistical barriers to accomplishing such an infrastructural overhaul, smaller interventions might effectively mitigate D&A. For example, healthcare workers identified the need for facility norms for intercultural births so that all agree on alternative birth positions, allowing women to take home their placentas, herbal medicines, and, space permitting, birth companions. Healthcare workers felt uneasy about some of these interventions, suggesting that studies examine the safety of these practices before integrating them into clinical practice.

Healthcare workers also called for campaigns within hospitals to spur discourse surrounding RMC, as it currently is a controversial topic many do not wish to acknowledge. Additionally, community stakeholders noted that cultural liaisons and interpreters are often denied entry to hospitals and requested increased collaboration between facilities and indigenous health advocates.

National health systems-level

Participants desired the inclusion of rights-based language, accountability mechanisms against D&A, increased funding for RMC (and maternity care in general), and improved integration of indigenous and biomedical health systems into national health policy. Many proposed interventions, especially at the facility level, require funding increases to facilities in general and to programs that specifically target D&A. Participants also called for the strengthening of indigenous health systems to promote the preservation of indigenous culture and ancestral knowledge.

Discussion

The articles of the United Nations’ 1979 Convention on the Elimination of All Forms of Discrimination against Women established the rights of women to be free from discrimination, physical or psychological violence, abuse, aggression, or coercion in all fields, including sexuality and reproduction [35]. A 2011 document affirming the universal rights of childbearing women by the White Ribbon Alliance for Safe Motherhood and a 2015 statement by the World Health Organization furthered the international discourse condemning D&A by explicitly affirming maternal health rights [2, 3]. The Colombian judicial system has agreed that the protection of sexual rights and reproductive rights includes the safeguarding of access to reproductive health services [36]. Despite the growing body of evidence detailing D&A worldwide and these calls for its eradication, local policies are not often enacted, and specific and meaningful action has not been taken. Notably, D&A experienced by internally displaced indigenous women in Colombia has not been evaluated.

The Political Constitution of Colombia of 1991 defined the nation as polyethnic and multicultural, indicating the obligation of the state and the rest of society to recognize, protect, and respect the different cultures and worldviews represented in the country. The vulnerable position of indigenous women in Colombia has also been recognized by the country’s Constitutional Court [37]. However, Colombia’s general health system (Sistema General de Seguridad Social en Salud, or General System of Social Security in Health) fails to fully recognize indigenous communities’ cultural beliefs, ancestral knowledge, and traditional medicine practices. It has not dismantled barriers to health care access by indigenous communities, such as those related to language and geographic barriers, among others [38]. Currently, the Indigenous System of Proprietary and Intercultural Health (Sistema Indígena de Salud Propio Intercultural, or SISPI) operates parallel to the general health system. A greater articulation between the two is greatly needed to address the particular needs of Colombia’s indigenous population [39].

This study’s findings illuminate a potential path towards eradicating D&A against indigenous women in Antioquia, Colombia. Drivers of D&A and points for intervention are identified, drawing from the perspectives of indigenous women, healthcare workers, and local stakeholders. Under the framework developed by Ratcliffe, the qualitative data indicates that D&A is driven by interrelated phenomena that fall under four connected subsystems: the individual and community, the clinician, the health facility, and the national health system.

Participants named individual and community-level factors as central to the incidence of D&A. Normalization of D&A during childbirth and indigenous women’s lack of autonomy and empowerment were described by each category of participants. Several emphasized that, while these phenomena exist across patient populations, indigenous women are particularly unaware of their rights within healthcare delivery spaces. Participants also described how indigenous women are “silent,” during childbirth itself and in denouncing D&A. Many highlighted that women’s rights during childbirth are seldom discussed in indigenous communities. This data agrees with findings from Colombia [37, 40] and Mexico [41] highlighting the invisibilization that often occurs surrounding mistreatment during childbirth and that D&A against indigenous women is difficult to identify by women themselves and by healthcare workers. These findings also explain data from a Mexican study wherein indigenous women are less likely to report experiencing OV during childbirth [42]. Rather than being less likely to experience D&A, it might be that indigenous women are less equipped than their nonindigenous counterparts to identify it. Indigenous women called for efforts to equip their communities with information about their rights within healthcare and mechanisms to lodge complaints against D&A.

Lack of ANC is another identified driver of D&A that contributes to poor rapport between women and their clinicians. ANC is known to be dependent upon geographical access, financial resources, and cultural beliefs [43]. Indigenous women are more likely to face barriers to accessing and receiving this care and are less likely to seek it from health facilities—many opt to receive care from traditional midwives and community healers instead and may only travel to the hospital should an emergency arise. Targeted interventions to promote ANC among indigenous women and incorporate indigenous healers into the general health system might help to provide more continuous and consistent care during the pregnancy in a way that is culturally respectful and accessible.

The second and third subsystems are the interrelated clinician-level and facility-level factors. The drivers associated with these subsystems include clinician prejudice, cultural, linguistic or cultural barriers to communication, lack of understanding of indigenous culture, clinician training and medical culture, burnout and demoralization, and inadequate infrastructure, space, and human resources. Participants demanded a review of medical curricula and urged the inclusion of a multicultural perspective promoting respect for traditional medicine and ancestral knowledge. Most health facilities in Colombia prohibit practices such as the presence of traditional midwives or healers alongside physicians during the childbirth, performing sacred rituals, and allowing the mother to bring the placenta home. Welcoming these practices will allow indigenous women to experience higher quality, more compassionate care. Additionally, hospitals must ensure that patients who do not speak Spanish have access to medical interpreters, rather than relying on friends or family to provide these services.

Hospital infrastructure was criticized by all categories of participants—overfull, understaffed facilities greatly complicate the delivery of RMC. Certain elements of the institutionalized birth experience, such as the prohibition of visits by family or friends or lack of private rooms, are a product of space and resource constraints. All women are treated the same in this regard, several clinicians explained—indigenous women are not subject to poorer conditions than non-indigenous patients. However, it is important to consider that indigenous women are less familiar with the hospital environment than their peers. Experiences deemed typical by the non-indigenous Colombian woman, such as crowded rooms, solitude during the birth process, or uncomfortable exams by male clinicians, are often viewed as D&A by indigenous women. The ultimate solution to these concerns would be greater allocation of funding and human resources to maternity wards. In particular, allowing birth companions for continuous support was cited by participants as being a major potential protective factor against D&A, and has been shown to improve maternal and infant outcomes [44]. Administrators might also consider other intermediate interventions, such as equipping clinicians with the tools (including interpreters) to provide reasonable accommodations to indigenous women and outreach efforts to familiarize indigenous women with the hospital environment during their pregnancies.

The final level is that of the health system, including policies regulating maternal care. Calls for greater promotion of RMC has led to some progress in this area—in 2017, the Ministry of Health and Social Protection called for departmental and local governments to ensure the safe, dignified, and adequate care for women during childbirth [45]. However, widespread interventions have not yet been designed or implemented and many feel that RMC remains a low priority in public policy. Our findings indicate that the first step to combatting D&A is to make it a socially recognized phenomenon in Colombia—among legislators, public health officials, clinicians, indigenous communities, and women themselves. D&A is symptomatic of a fractured health system which prevents clinicians from realizing every woman’s right to a quality, respectful, and dignified childbirth [7]. It is only by strengthening the agency of all women, especially those from vulnerable communities, the re-allocation of resources, and the creation of clear accountability mechanisms that mistreatment during childbirth will be eradicated.

Limitations

The indigenous women who participated in this study belonged only to Embera subgroups. Those who elected to participate in the study had a higher level of Spanish proficiency and a closer proximity to non-indigenous settings (such as academia or public organizations) than indigenous women who were unreachable for recruitment, indicating potential selection bias within this study. We mitigated this bias to the fullest extent possible by recruiting participants through a central indigenous organization (OIA) and selecting an adequate number of participants such that thematic saturation was reached [29]. However, the results described may not be transferable to every Embera woman in Antioquia, or to different indigenous communities in Colombia or other countries. Despite the specificity of our study population, the women interviewed demonstrated diversity in birth facility and time period. Furthermore, they indicated that their proximity to academic and public settings enhanced their abilities to identify D&A within their experiences. Additionally, the timespan between the time of childbirth and interview varied greatly among participants, with some births having occurred many years prior to this study. Though this range illustrates the lasting effects that D&A might have on women, it also introduces the potential for recall bias. We followed established guidelines to ensure content validity despite these potential biases, including interviewing clinicians and community stakeholders in addition to indigenous women, maintaining consistency in the interview facilitator and semi-structured interview guide, continuing interviews until data saturation was reached, and following standard qualitative data analysis guidelines [46]. A meeting was held with the participants after the completion of data analysis to validate the results obtained.

Further studies are needed to assess the prevalence of D&A among indigenous women in Colombia and evaluate the implementation of interventions such as those proposed here.

Conclusions

D&A is a fundamental violation of women’s rights to respectful maternity care and threatens global progress in improving maternal health outcomes. This is one of the first studies to evaluate indigenous women’s experiences of D&A, and the first to do so in Colombia. The results presented here indicate that indigenous women, like their non-indigenous peers, experience D&A during childbirth due to interrelated risk factors. However, they are also vulnerable to mistreatment due to added cultural and linguistic barriers and prejudice—factors which must be made central in public health agendas. Broad and meaningful action is urgently needed to ensure respectful, accessible, and dignified care during childbirth. Health systems must be bolstered to provide this care from a rights-based perspective, and interventions to mitigate D&A must be interdisciplinary and locally specific, considering the needs and wants of the very women they serve.