Respectful maternity care and intersectional identities in low- and middle-income countries: a systematic review

Abstract

Background

Disrespect and abuse (D&A) during facility-based births remains widespread in low- and middle-income countries (LMICs). This disproportionately impacts women from marginalized social strata. Respectful Maternity Care (RMC), as articulated in the 2014 World Health Organization (WHO) statement, constitutes both a fundamental human right and a quality care imperative. However, women’s experiences of RMC are not uniform and are influenced by intersecting social identities such as caste, class, ethnicity, age, disability, marital status, language, literacy, and migrant status.

Objectives

This review aimed to (1) quantify and compare the prevalence of disrespect and abuse (D&A) across social subgroups in LMICs; (2) identify social and systemic determinants such as poverty, education, migration status, and provider bias, that influence women’s experiences of respectful or disrespectful maternity care; (3) synthesize qualitative insights to understand how intersecting identities shape women’s childbirth experiences; and (4) develop a conceptual model explaining how identity factors and health‑system level drivers interact to produce inequities.

Methods

We conducted a systematic review following PRISMA guidelines. We synthesized literature on intersectional experiences of RMC and D&A in LMICs. Searches spanned five databases and key grey literature sources. The data extraction was finalized in April 2025. Eligible studies were published between 2014 (coinciding with the release of the WHO RMC statement) and 2025. Eligible studies included peer-reviewed primary research and high-quality systematic or scoping reviews that provided identity-disaggregated evidence on respectful or disrespectful maternity care in facility-based childbirth settings in LMICs. A thematic narrative synthesis integrated quantitative and qualitative findings. .

Results

Seventeen studies across 15 LMICs were included. Women with intersecting vulnerabilities, especially those who were young, impoverished, unmarried, disabled, lower caste/ethnic minority backgrounds, or migrants, were consistently more susceptible to experience neglect, non-consented care, verbal abuse, or exclusion from decision-making processes. These patterns indicate systemic injustices rather than individual occurrences. Key explanatory mechanisms included social stigma, provider biases about who is “deserving” of care, and communication obstacles. Disability, migration status, socio-economic status, and religion were under-explored in the literature despite being linked to discrimination.

Conclusion

This review highlights that intersectional inequities substantially influence women’s experiences of respectful maternity care in LMICs. However, causal inference is limited by study heterogeneity and uneven representation of identity factors.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-025-08566-0.

Background

Every woman has the right to dignified, respectful maternity care during childbirth [1, 2]. Respectful Maternity Care (RMC) entails care that upholds a woman’s dignity, privacy, autonomy, confidentiality, and freedom from harm or mistreatment, and that facilitates informed decision-making and continuous support [2]. Yet this ideal remains elusive across many low- and middle-income countries (LMICs). Disrespect and abuse (D&A) during facility-based childbirth could range from verbal and physical abuse to non-consented procedures, discrimination, and neglect. These are alarmingly common and, in some contexts, normalized [1, 3, 4]. In 2014, the World Health Organization (WHO) released a statement that unequivocally framed such mistreatment as a violation of women’s rights and a barrier to high-quality maternal care [2].

Growing evidence shows that the risk of receiving disrespectful care is not evenly distributed. However, women’s experiences of RMC or D&A are strongly shaped by their social positions and the intersecting systems of power in which they are embedded [5]. Intersectionality theory, first articulated by Crenshaw [6], holds that multiple identities such as gender, ethnicity, caste, class, age, marital status, and disability shape people’s experiences of privilege or oppression [7]. In the context of childbirth, a woman’s likelihood of being treated with dignity is influenced by how her various identities position her within societal hierarchies [3, 8, 9].

Initial evidence indicates that women who are younger, unmarried, poorer, less educated, or from marginalized ethnic, caste, or religious groups disproportionately experience abuse when giving birth in facility settings. The global review by Bohren et al. [3], documented how women from low-income backgrounds, migrants, and ethnic minorities were often subject to discriminatory slurs, neglect, or coerced procedures. Adolescent or unmarried mothers faced moralizing scorn and reduced clinical attention. A study in Tunisia showed that single mothers were actively humiliated and denied basic care on moralistic grounds [8]. Similarly, a large multicenter study in Nepal revealed that younger women and those from the Dalit caste had significantly higher mistreatment scores than women from more privileged social strata [10]. Across settings, structural discrimination, which includes casteism, racism, ageism, ableism, and class-based prejudice, appears to shape both provider behavior and facility norms [1, 11–13].

These abuses are the result of individual attitudes that are deeply embedded in health systems. Studies from South Asia, sub-Saharan Africa, and Latin America consistently show how social stratification influences who receives attentive care, who is left unattended, and who is subjected to punitive or dehumanizing treatment [5, 14, 15]. Disrespectful maternity care has been linked to lower facility utilization for skilled births, especially among marginalized groups who associate hospitals with shame or abuse [16, 17]. This erodes trust in health systems and contributes to persistent inequalities in maternal and newborn health outcomes.

At the same time, research on intersectional dimensions of RMC remains patchy. Some identities, such as religion, disability, and migration status, are often excluded or superficially treated in studies [18, 19]. Even in studies where socioeconomic or ethnic variables are included, results are usually not disaggregated or analysed across intersectionality domains. For example, Christe and Padmanaban [20] found high RMC satisfaction among postpartum women in India but did not stratify responses by caste or poverty, despite these relevant factors. Similarly, a U.S. CDC survey found that Black, Hispanic, and multiracial women were significantly more likely to experience mistreatment during childbirth, but many such studies are rarely contextualized in LMICs [21].

To design interventions that close these equity gaps, we must better understand how multiple disadvantages interact to shape maternity care experiences, intersectional frameworks, such as the “Cycle to Respectful Care” proposed by Green et al. [9], offer a system‑wide transformation rooted in community voices and anti‑discriminatory practice while others, such as Lusambili et al. [22], emphasize local accountability structures such as community scorecards and social audits to uncover hidden disparities and reduce abuse.

This systematic review aims to consolidate and critically appraise the evidence on how intersecting social identities shape women’s experiences of RMC or D&A during facility-based childbirth in LMICs. Specifically, it examines how individual and overlapping identity factors such as caste, religion, socio-economic status, migration status, language, marital status, disability, and literacy alter the likelihood of receiving respectful maternity care. The review has four objectives: (1) to quantify and compare the prevalence of disrespect and abuse across social subgroups and types of mistreatment (physical, verbal, non‑consented care, neglect, discrimination, etc.); (2) to identify social and systemic determinants such as poverty, education, migration status, discrimination and provider biasthat influence women’s experiences of respectful or disrespectful maternity care; (3) to synthesize qualitative insights on how intersecting identities shape women’s childbirth experiences; and (4) to develop a conceptual model explaining how identity factors and health‑system level drivers interact to produce inequities. Utilizing this multifaceted lens, we aim to transcend mere average prevalence estimates and spotlight the social mechanisms that influence which women receive respect during childbirth and which do not. This review can inform more equitable, rights-based, dignified, and culturally responsive approaches to maternal care in LMICs.

Methods

Protocol and reporting

The review was conducted and reported following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The review was registered in PROSPERO with registration number CRD420251067084. The protocol outlines the search strategy, inclusion criteria, and analysis plan, including an intersectionality framework for data synthesis (See supplementary material 1).

Eligibility criteria

Study types

Only peer-reviewed studies providing primary data were included. However, two high-quality syntheses (Bohren et al. [5]; Ansari & Yeravdekar [23]) were retained because they offered intersectionality-specific evidence and contextualized drivers of disrespect and abuse that were not captured in individual studies. Given the focus, we also included systematic reviews when they provided disaggregated analyses of identity‑related factors in RMC or D&A. Opinion pieces or purely theoretical papers were excluded.

Population

Studies must focus on childbirth in healthcare facilities (hospitals, clinics, maternity centers) in LMICs (as per World Bank classification and definition at the time of the study). We included studies from all WHO regions as long as the country was low- or middle-income during the study period. Both urban and rural settings were included. We excluded studies from high-income countries and studies focused on home births or community settings without facility interaction. Studies involving home‑to‑facility transfers were included if the birthing experience occurred predominantly in the facility, and we considered provider perspectives only when linked to identity‑driven mistreatment.

Exposure/factors

The study must report on at least one of the relevant social identity factors (caste, ethnicity, religion, socio-economic status, educational level or literacy, marital status, migration/refugee status, language, disability, or related proxies like age as an intersecting factor). We included studies that stratified results by these factors or explored these themes qualitatively. For this review, “social identities” refer to socially constructed categories such as caste, ethnicity, religion, socio‑economic status, literacy, marital status, migration/refugee status, language, disability, and related proxies. Studies had to disaggregate or analyze outcomes by at least one identity axis rather than merely mention it. When studies reported combined identity groups (e.g., “migrant women of low socio‑economic status”) without disaggregated results, we extracted identity‑specific findings where possible. Otherwise, such studies were noted qualitatively but not included in axis‑specific quantitative analyses. Facility‑level characteristics (e.g., type of facility, provider qualification, timing of delivery, length of stay) were recorded for descriptive purposes but were not treated as social identity exposures.

Outcomes

The outcomes of interest were measures of respectful or disrespectful care during childbirth. This included: observed or self-reported instances of mistreatment (physical abuse, verbal abuse, non-consented care, neglect, discrimination, etc.), composite scores or indices of RMC (e.g., the Person-Centered Maternity Care scale scores), women’s satisfaction or perceived respect, and any negative experiences attributable to provider behaviour or facility environment during labour, birth, or immediate postpartum. Measures of “satisfaction” were included only when explicitly linked to respectful or disrespectful care. Studies using validated instruments (such as the Person‑Centered Maternity Care scale) and non‑validated but clearly defined tools were eligible, provided they measured respectful or disrespectful care domains. We also included qualitative accounts of respectful or abusive treatment as recounted by women. Studies focusing only on clinical outcomes (without addressing the experience of care) were excluded unless they explicitly linked outcomes to experiences of mistreatment.

Articles must be published in English to be eligible. The search spanned from January 2010 to May 2025. However, given that RMC as a concept gained prominence in the past decade, 16 of the 17 included studies were published between 2014 and 2025. We limited the search to English‑language publications due to resource constraints and because most relevant literature in LMICs is published in English.

Information sources and search strategy

We performed a comprehensive literature search across multiple databases and platforms: PubMed/MEDLINE, Dimensions, and Google Scholar. These were chosen to capture both global and region-specific literature.

The search strategy used combinations of keywords and controlled vocabulary (where applicable) around three themes: (1) maternal care setting (“childbirth”, “maternity care”, “labor and delivery”, etc.), (2) respectful or disrespectful treatment (“respectful care”, “disrespect”, “abuse”, “mistreatment”, “obstetric violence”), and (3) social identity factors (“caste”, “ethnic*”, “religion”, “socioeconomic”, “poverty”, “education”, “literacy”, “adolescent”, “unmarried”, “single mother”, “disability”, “migrant”, “refugee”, “language”, etc.) (Supplementary File 1). These were combined using Boolean logic. Google Scholar searches were used to identify any recent papers or reports not indexed in the databases, using similar terms and snowballing from key article citations. All searches were conducted in May 2025. The results from each source were exported to a reference manager, and duplicates were removed before screening. The search strategy was developed in consultation with a medical librarian and informed by existing scoping reviews. Although our search encompassed grey literature to minimize publication bias, none of the identified grey sources met the inclusion criteria. Links to the search history and strategies for each database are provided in Supplementary File 2.

Selection process

Three reviewers (VAA, AA, CO) independently screened the titles and abstracts of all retrieved records for potential eligibility. We employed broad inclusion at this stage. Next, full-texts of all potentially relevant studies were obtained and assessed against the inclusion criteria. Any disagreements or uncertainties in inclusion were resolved through discussion or consultation with a fourth reviewer (QESA). We maintained a PRISMA flow diagram to document the selection process (Fig. 1). Screening was managed using Rayyan^® software to enhance transparency. Common reasons for exclusion at full-text screening included: not reporting outcomes on respectful/disrespectful care (described general quality of care without specifics on respect or abuse), not disaggregating by any social factors or any specific focus on them, or being purely commentary without new data. In total, 17 studies met all criteria and were included in this review.

Fig. 1.

PRISMA flow diagram of study selection

Data extraction and quality assessment

For each included study, we extracted key details: author, year, country, study design, sample characteristics (including any specific population focus such as “rural women” or “adolescents”), how RMC or D&A was measured, which identity factors were analyzed, and the relevant findings. We tabulated these data (see Table 1 in Results). For qualitative studies, we noted the main themes related to identity (for example, stigma against certain groups or differential treatment narratives). Quality appraisal was performed for each study using design-appropriate tools (e.g., Critical Appraisal Skills Programme (CASP) for qualitative, Newcastle-Ottawa for observational studies, etc.), but no study was excluded based on quality. Instead, quality considerations (such as risk of bias in measurement or sampling) are noted in the narrative synthesis of results. Overall, the evidence base is largely observational and qualitative.

Table 2.

Characteristics of the included studies

S/N	Author (Year)	Title	Country	Design & Sample	Population	RMC/D&A Outcome Measures	Identity Factors Examined	Summary of Key Findings
1	Amroussia et al. [8]	“Is the doctor God to punish me?!” An intersectional examination of disrespectful and abusive care during childbirth against single mothers in Tunisia	Tunisia (Middle East/North Africa)	Qualitative (phenomenological); n = 16 single mothers interviewed	Unmarried young mothers accessing public maternity care in Tunis	Women’s narratives of treatment during labor and birth (focus on respect, dignity, challenges)	Marital status, socio-economic status (all participants were low SES); and gender norms	Single women faced intense stigma and judgment from providers due to bearing a child out of wedlock. Participants described verbal abuse and moralizing scolding from midwives, being left unattended, and feeling “humiliated.” Intersection of gender + class: most were poor, young women, compounding their vulnerable position. Providers’ attitudes reflected societal prejudice, leading to denial of respectful care.
2	Bohren et al. [3]	The Mistreatment of Women during Childbirth in Health Facilities Globally: A Mixed-Methods Systematic Review	Multi-country (global South; 65 studies from Africa, Asia, Latin America)	Systematic review (mixed methods synthesis of 65 studies; includes > 3,500 women’s narratives)	Women in facilities (primarily public hospitals) across diverse LMICs	Catalogued types of D&A (physical, verbal, non-consent, neglect, discrimination, etc.) and qualitative themes of mistreatment	Socio-economic status, ethnicity, migration status, age/marital status (adolescents)	Found evidence of all seven types of D&A across regions. Marginalized women suffered unique abuses: Low-SES, migrant, and ethnic minority women were often called slurs or derogatory names during childbirth. Adolescents/unmarried mothers were shamed with accusatory remarks about their sexual activity. Language barriers contributed to poor care for migrant/refugee women. Concludes that underlying power imbalances and prejudices lead to disproportionate mistreatment of vulnerable groups.
3	Bohren et al. (2019) [24]	How women are treated during facility-based childbirth in four countries: a cross-sectional study with labour observations and community surveys	Nigeria, Ghana, Guinea, Myanmar	Mixed methods: Direct observations of 2,016 births + surveys of 2,672 women postpartum	Women giving birth in hospitals (urban/peri-urban) in the 4 LMICs; diverse ethnic groups	Observed frequency of different mistreatment events; Survey-reported experiences; quantitative analysis of predictors	Age, Education (as proxy for literacy/SES); other factors: parity, facility type	One in three women experienced some form of mistreatment. Key determinants of abuse (survey data) were young age and low education: women 15–19 years old and those with no schooling had the highest odds of mistreatment. For example, a teenage mother with no education had ~ 3.6 times higher odds of verbal abuse than an older, educated woman. Younger, less-educated women were particularly likely to be shouted at, scolded, or ignored by staff. This indicates substantial inequity in care quality linked to social status.
4	Gurung et al. [10]	Mistreatment during childbirth and postnatal period reported by women in Nepal — a multicentric prevalence study	Nepal (South Asia)	Quantitative cohort study; n = 62,926 women surveyed at discharge from 11 hospitals	Women delivering in diverse hospitals (urban and rural) across Nepal, very large sample	Structured questionnaire on mistreatment experiences; “Mistreatment index” computed via PCA from multiple items	Ethnicity/Caste, Age, education, parity, newborn factors	Reported that younger and multiparous women were more likely to experience non-consented care and neglect. Parity was inversely related to autonomy—primiparas were less confident in requesting consent. Educational level significantly predicted respectful treatment (p < 0.05).
5	Ishola et al. [13]	Disrespect and abuse of women during childbirth in Nigeria: A systematic review	Nigeria (sub-Saharan Africa)	Systematic review (14 studies, incl. surveys and interviews across Nigeria)	Women in maternity facilities in Nigeria (various regions)	Aggregate prevalence of D&A types; narrative synthesis of contributing factors	Socio-economic status, Education/empowerment, also health system factors	Identified non-dignified care (e.g., rude attitudes) as the most frequent abuse, while physical abuse was less frequent. Key drivers of D&A in Nigeria were low socio-economic status and lack of education/empowerment among women. Poor women were more likely to experience neglect or demands for unofficial fees. Inadequate provider training and weak accountability exacerbated abuses. The review underscores that disadvantaged women face both greater exposure to mistreatment and fewer avenues to voice complaints.
6	Kaur et al. [1]	Respectful Maternity Care during Childbirth among Women in a Rural Area of Northern India	India (Haryana, South Asia)	Quantitative cross-sectional survey; n = 485 women interviewed at the community level postpartum	Predominantly rural, low-income women who delivered in institutions (mix of public and private)	Person-Centered Maternity Care (PCMC) scale to quantify RMC; analyzed domain scores and overall “received RMC” vs. not	Caste (social group), facility type; also checked age, education, religion, parity	Women with secondary or higher education had higher RMC scores. Parity was marginally associated—multiparas reported more confidence and less fear of reprimand. Education positively influenced ability to demand consented care, aligning with empowerment theory.
7	Wassihun et al. [25]	Prevalence of disrespect and abuse of women during childbirth and associated factors in Bahir Dar, Ethiopia	Ethiopia (East Africa)	Quantitative cross-sectional; n = 410 postpartum women surveyed (multistage sampling)	Women who recently delivered in health facilities (public hospitals and health centers) in an urban Ethiopian town	Structured questionnaire on seven types of D&A experienced; logistic regression for factors	Economic status, education, etc., also health facility type, length of stay, ANC visits	Found 67.1% of women experienced ≥ 1 type of D&A during childbirth. The strongest factor: Low family income – women from poor households (< median income) had ~ 1.7 times higher odds of D&A than those better-off. Inadequate antenatal care (≤ 3 visits) and giving birth in a public hospital (vs. private) were also associated with higher D&A. Education was not a significant independent factor after adjusting (possibly due to correlation with income). Illustrates how poverty and reliance on under-resourced facilities increase mistreatment risk.
8	Prins et al. [26]	Migrant women’s experience of antenatal care in an urban and rural setting in north and North-West Thailand: A cross sectional survey	Thailand	Cross-sectional survey; N = 222	Migrant women from Myanmar attending ANC in Sarapee (urban) and SMRU (rural) clinics	Satisfaction (REPRO-Q domains), perceived disrespect, ability to refuse treatment, autonomy, dignity, communication, confidentiality, promptness	Migration status (documented vs. undocumented), language, and insurance status	Disaggregated findings showed that undocumented migrant women and those lacking Thai health-insurance coverage were more likely to report language barriers, delayed service, and perceived disrespect (p < 0.01). Legal status and insurance coverage together explained 21% of satisfaction variance.
9	Okafor et al. [27]	Disrespect and abuse during facility-based childbirth in a low-income country	Nigeria	Cross-sectional survey; N = 446 postpartum women	Postnatal women who gave birth at ESUTH-Parklane in Enugu	Seven categories: physical abuse, non-consented care, non-confidential care, non-dignified care, discrimination, neglect, detention	Age, marital status, education, parity, religion, ethnicity	98% reported ≥ 1 type of abuse; most common were non-consented care (54.5%) and physical abuse (35.7%); no significant association found with sociodemographic variables; pervasive normalization of disrespect during childbirth noted.
10	Ngarmbatedjimal et al. [28]	Refugee women and providers perceptions of person-centered maternity care	Chad	Qualitative study with 22 postpartum women, 5 Chadian midwives, and 12 refugee providers	Sudanese refugee women in Mile and Kounoungou camps; midwives and TBAs	PCMC domains: dignity, communication, privacy, social support, facility environment	Refugee status, age, parity, language, adolescent status, marital status, provider type	Care was largely respectful, but resource gaps led to neglect and scolding. Providers valued PCMC; language barriers and lack of space affected care.
11	Rajkumari et al. [29]	Assessment of Respectful Maternity Care during Childbirth Experiences among Mothers in Manipur	India (Manipur)	Cross-sectional study of 231 women using PCMC-based questionnaire	Mothers with children < 2 years who had institutional deliveries	Seven PCMC domains including dignity, support, environment, transparency	Place of delivery (tertiary vs. primary), employment, education, parity	Although employment, education, and parity were collected, none showed statistically significant association with RMC domains (p > 0.05). Authors noted this could reflect sample homogeneity or facility-level influences overriding individual identities.
12	Noori et al. [12]	Women with Disabilities’ Experiences with Respectful Maternity Care in Nepal: a Qualitative Study	Nepal	Qualitative study: 12 semi-structured interviews with women with disabilities, 7 with healthcare providers, 2 FGDs with 11 Female Community Health Volunteers	Women aged 21–42 with visual or physical disabilities, from various caste and socioeconomic backgrounds in the Kathmandu outskirts	RMC Charter domains: dignity/respect, freedom from harm, privacy/confidentiality, information/consent, non-discrimination, timely care, autonomy	Type of disability, education, employment, caste, economic status, parity	Explored intersectional identities among women with visual and physical disabilities. Education and employment influenced awareness of rights, while caste and economic status affected facility choice. Barriers included inaccessible infrastructure and provider assumptions of incompetence.
13	Freedman et al. [30]	Eye of the beholder? Observation versus self-report in the measurement of disrespect and abuse during facility-based childbirth	Tanzania	Mixed-methods study with 232 women at baseline and 237 at endline; observations and exit interviews at 2 hospitals	Women in active labor, aged 15+, observed during childbirth and interviewed post-discharge	14 indicators of D&A, including non-dignified care (shouting, scolding), physical abuse (slapping), neglect, non-consented care, privacy violations, and bribery/informal payments	Parity, education, age, marital status, income status, prior abuse, complications	Found that parity, education, age, marital status, and income shaped self-reporting differences. Younger, poorer, or less-educated women under-reported abuse due to normalization and fear. Observers recorded 70% D&A prevalence versus 10% self-report, revealing hidden bias linked to socio-demographic power gradients.
14	Dolatabadi et al. [11]	Disrespect and abuse during childbirth and associated factors among women: a cross-sectional study	Iran	Cross-sectional study of 300 women within 5 days postpartum in Tehran	Women who recently had vaginal births at comprehensive health centers	7 dimensions of D&A measured via validated questionnaire (e.g., physical abuse, neglect, privacy violation, non-consented care)	Ethnicity, pregnancy intention, caregiver type, companion presence, instrumental delivery	68.3% of women experienced D&A; Fars ethnicity, unwanted pregnancy, midwife-led care, and instrumental delivery were significantly associated with higher odds of mistreatment.
15	Bradley et al. [14]	Midwives’ perspectives on (dis)respectful intrapartum care during facility-based delivery in sub-Saharan Africa: a qualitative systematic review and meta-synthesis	Sub-Saharan Africa	Qualitative systematic review of 11 studies across 6 countries (e.g., South Africa, Ghana, Mozambique)	Facility-based midwives and student midwives working in labor wards	Thematic synthesis using a conceptual framework categorizing micro-, meso-, and macro-level drivers of (dis)respectful care (e.g., power, control, resource constraints, status in health hierarchy)	Caste/status, gender norms, power dynamics, race/ethnicity, provider hierarchy	Midwives described control over women, systemic blame culture, poor working conditions, and low professional status as drivers of disrespect; few acknowledged structural inequalities affecting women’s dignity during childbirth.
16	Ansari & Yeravdekar [23]	Respectful maternity care during childbirth in India: A systematic review and meta-analysis	India	Systematic review and meta-analysis of 7 studies (2016–2019); 5 cross-sectional, 2 mixed methods; total n = 4959	Women who delivered in facilities or the community	Pooled D&A prevalence: 71.31%. Forms of D&A: non-consent (49.8%), verbal abuse (25.8%), threats (23.3%), physical abuse (17%), and discrimination (14.8%); assessed across settings (hospital/community) and via interviews, FGDs, and observations.	Age, caste, socioeconomic status, parity, autonomy, birth complications, facility type, provider qualification, delivery timing	High prevalence of D&A reported across both public and private sectors. Community-based studies had a higher prevalence (77.3%) than hospital-based studies (65.4%). Drivers include sociocultural norms, systemic gaps, and environmental constraints. Targeted RMC interventions needed.
17	Bohren et al. [5]	Towards a better tomorrow: addressing intersectional gender power relations to eradicate inequities in maternal health	Multi-country (Global focus incl. LAC, Sub-Saharan Africa, SE Asia)	Mixed methods: Conceptual framing + Secondary analysis (ANC quality indicator in 8 LAC countries) + Scoping review (59 studies, 31 countries)	Women and birthing people (cisgender, trans, non-binary, intersex), including marginalized subgroups (e.g., Indigenous, rural, poor)	Explored equity in antenatal care quality (ANCq8+) using DHS/MICS; scoping review covered interventions targeting access, quality, and discrimination	Ethnicity, socioeconomic deprivation, gender identity, age, caste, disability, refugee/migration status, education, rural/urban residence	Intersectionality reveals compounded disadvantages (e.g., being Indigenous and poor) not captured by unidimensional indicators. Equity-informed interventions focused mainly on access and poverty; none fully used intersectionality in design or evaluation. Recommendations made to embed intersectionality in research, policy, and clinical practice.

Two reviewers independently extracted data using a standardized form and appraised methodological quality using the CASP for qualitative studies and the Newcastle–Ottawa Scale for observational studies. Disagreements were resolved through discussion and, if necessary, consultation with a third reviewer. For mixed‑methods studies, qualitative and quantitative components were appraised separately. Relevance was defined as any finding relating to respectful or disrespectful care. We extracted all such findings and coded them using the identity axis. The PROSPERO protocol (Registration ID CRD420251067084) Supplementary File 1, complete PRISMA 2020 checklist (Supplementary File 3), detailed inclusion/exclusion criteria (Supplementary File 4), and critical appraisal outputs (Supplementary File 5) accompany this manuscript.

Data synthesis

We employed a thematic narrative synthesis approach by integrating quantitative and qualitative evidence. Given heterogeneity in outcomes and measures, a meta-analysis was not feasible for most factors (and was not the primary aim). We combined quantitative and qualitative findings through narrative integration and side‑by‑side comparison rather than statistical meta‑analysis. Statistical significance and effect sizes were considered when reported in primary studies. Team members reflected on their positionality to enhance reflexivity and sought to mitigate interpretive bias throughout the analysis. Instead, quantitative findings were summarized to show patterns (e.g., which factors consistently emerge as significant predictors of D&A). Qualitative findings were synthesized using an inductive approach to identify common themes regarding intersectional mistreatment (such as “social stigma and judgment” or “power and silence”). These were used to explain the mechanisms behind the quantitative patterns. Finally, we developed a conceptual model (Fig. 2) to illustrate how intersecting identities and structural factors converge to influence respectful maternity care. We iteratively built this model based on recurring concepts in the data (e.g., provider biases, health system constraints, and women’s social positions).

Fig. 2.

Conceptual model illustrating the influence of intersecting identities and structural factors on respectful maternity care

Results

Study selection

The comprehensive search identified 1,234 records from database sources, with an additional 15 records from reference lists and grey literature. After de-duplication, 1,200 unique titles and abstracts were screened. Of these, 1,100 were excluded for not meeting relevance criteria. Full-text reviews were conducted on 100 articles, excluding 75 (primarily due to lack of intersectional analysis or LMIC focus). Seventeen studies met the inclusion criteria and were retained for analysis (Fig. 1).

Characteristics of included studies

The 17 included studies comprise seven quantitative, four qualitative, and six mixed-method or multi-component designs, primarily covering 15 LMICs across sub-Saharan Africa, South Asia, Southeast Asia, the Middle East, and North Africa (Tables 1 and 2). Across countries, India contributed three studies (including one systematic review), Nigeria contributed three, Nepal contributed two, and the remaining studies originated from a distinct LMIC. Several studies had large sample sizes (e.g., Gurung et al. [10], n = 62,926; Bohren et al. [24], n = 4,688), whereas others relied on in-depth interviews with smaller participant groups (e.g., Amroussia et al. [8], n = 16). Countries with multiple studies include India [1, 23, 29] and Nigeria [13, 24, 27]. Various tools were used to measure outcomes, including the Person-Centered Maternity Care (PCMC) scale, structured mistreatment checklists, and qualitative narrative accounts. All studies examined at least one relevant identity axis: socio-economic status, education, caste/ethnicity, age, marital status, migration, disability, or religion. These axes were central to determining experiences of respectful or disrespectful care.

Table 1.

Study aims and identity factors of the included studies

S/N	Study (author, year, country, design & sample)	Identity factors analyzed	Main aim
1.	Amroussia et al. (2017, Tunisia, qualitative interviews with 16 single mothers) [8]	marital status; socio-economic status	explore disrespectful care experienced by unmarried low-income mothers
2.	Bohren et al. (2015, global, mixed-methods systematic review) [3]	socio-economic status; ethnicity; migration status; age/marital status	synthesize evidence on mistreatment of women during childbirth worldwide
3.	Bohren et al. (2019, Nigeria/Ghana/Guinea/Myanmar, mixed-methods observational study) [24]	age; education (proxy for SES)	measure prevalence and predictors of mistreatment in four countries
4.	Gurung et al. (2022, Nepal, quantitative cohort of 62,926 women) [10]	caste/ethnicity; age; education	estimate prevalence of mistreatment and identify caste- and age-related risk factors
5.	Ishola et al. (2017, Nigeria, systematic review) [13]	socio-economic status; education	synthesize evidence on disrespect and abuse during childbirth in Nigeria
6.	Kaur et al. (2025, India, cross-sectional survey of 485 women) [1]	caste; religion	assess respectful maternity care and associated social factors in rural Haryana
7.	Wassihun et al. (2018, Ethiopia, survey of 410 postpartum women) [25]	socio-economic status; education	estimate prevalence of disrespect and abuse and associated factors
8.	Prins et al. (2025, Thailand, cross-sectional survey of 222 migrant women) [26]	migration status; language; documentation	examine migrant women’s satisfaction and respectful care in urban and rural clinics
9.	Okafor et al. (2015, Nigeria, cross-sectional survey of 446 women) [27]	age; marital status; education; parity; religion; ethnicity	determine prevalence of disrespect and abuse during facility-based childbirth
10.	Ngarmbatedjimal et al. (2024, Chad, qualitative study with 22 postpartum women and midwives) [28]	refugee status; language; age; marital status	explore refugee women’s and providers’ perceptions of person-centered maternity care
11.	Rajkumari et al. (2021, India, cross-sectional survey of 231 women) [29]	education; parity; employment	assess respectful maternity care experiences among mothers in Manipur
12.	Noori et al. (2023, Nepal, qualitative study with 12 women with disabilities) [12]	disability type; socio-economic status; caste; education	explore experiences of respectful maternity care among women with disabilities
13.	Freedman et al. (2018, Tanzania, mixed-methods study observing 232 births) [30]	age; parity; education; marital status; income	compare observed versus self-reported disrespect and abuse
14.	Dolatabadi et al. (2025, Iran, cross-sectional survey of 300 women) [11]	ethnicity; pregnancy intention	measure prevalence of disrespect and abuse and associated factors
15.	Bradley et al. (2019, sub-Saharan Africa, qualitative systematic review of 11 studies) [14]	gender norms; caste/status; power dynamics	synthesize midwives’ perspectives on drivers of respectful and disrespectful intrapartum care
16.	Ansari & Yeravdekar (2020, India, systematic review and meta-analysis of 7 studies) [23]	age; caste; socio-s status; parity	estimate pooled prevalence of disrespect and abuse and associated factors in India
17.	Bohren et al. (2024, multi-country, mixed-methods study/scoping review) [5]	ethnicity; socio-economic deprivation; gender identity; age; caste; disability; migration status; education	analyze intersectional gender power relations and inequities in maternal health

Findings by identity axis

Of the 17 included studies, nine analyzed socio-economic status or income [5, 10, 13, 23–27, 30], eight analyzed education or literacy [1, 10, 11, 13, 23, 24, 26, 29], seven examined caste or ethnicity [1, 5, 10, 11, 23, 24, 26], seven considered age and/or marital status [8, 13, 23–28], four addressed migration or language [26, 28–30], three focused on disability [5, 12, 23], and one explored religion [1]. Only two studies explicitly modelled interactions across multiple identity axes. Women of lower socio-economic status consistently reported neglectful care. In Ethiopia, Wassihun et al. [25] found that women belonging to the lowest income quintile had significantly higher odds of encountering at least one form of mistreatment during childbirth (AOR 1.74). Ishola et al. [13] similarly observed that impoverished Nigerian women frequently experienced maltreatment and were pressured to pay informal fees. Financial insecurity and weak accountability mechanisms reduced women’s ability to challenge or avoid mistreatment. In Ethiopia, women in the poorest quintile had 74% higher odds of mistreatment compared with those above the median income.

Wealth/income

Across contexts, income consistently predicted exposure to mistreatment. In Ethiopia, Wassihun et al. [25] found that women from the poorest households had 1.74 times higher odds of experiencing ≥ 1 form of disrespect. Similar associations were reported by Ishola et al. [13] in Nigeria and Rajkumari et al. [29] in India. Poverty, compounded by low education, constrained women’s ability to refuse non-consented procedures and to access private facilities.

Education was identified as a significant correlate of quality care. Bohren et al. [24] indicated that women with no formal education were significantly more likely to experience verbal abuse, with the risk increasing in younger demographics. Gurung et al. [10] observed that illiterate and younger women in Nepal received less consented care and were often disregarded. Literacy was frequently tied to empowerment, influencing women’s navigation of their healthcare encounters. Illiterate women had substantially lower consented‑care scores than women with secondary or higher education.

Caste and ethnicity influenced experiences profoundly. Kaur et al. [1], found that women from certain castes/tribes in Haryana, India, had significantly lower odds of receiving respectful care, even when controlling for education and income while those who can pay for deliveries in private facilities received higher RMC than those who use government facilities. Gurung et al. [10] found similar disparities in Nepal, with Dalit women receiving the lowest quality of care. In Iran, Dolatabadi et al. [11] noted that minority ethnic women experienced significantly more mistreatment compared to Fars ethnic majority women. Religion appeared less frequently as a discriminatory factor. Kaur et al. [1], found no significant difference in care by religious group in India. However, in some contexts, religion intersects with caste or ethnicity, potentially masking effects.

Age and marital status shaped provider responses. Younger women (15–19 years) experienced higher rates of verbal and physical abuse [24]. In Tunisia, unmarried mothers faced moral judgment and were denied support during labor [8]. The intersection of youth and non-marriage status intensified vulnerabilities to the denial of respectful care. Women aged 15–19 had approximately 3.6 times higher odds of verbal abuse than women in their 30s.

Migration status and language barriers were consistently linked with disrespectful or non‑consented care. Prins et al. [26], found that Burmese migrant women in Thailand experienced more disrespectful care, particularly in urban facilities. Ngarmbatedjimal et al. [28] reported that Sudanese refugee women in Chad felt excluded from decision-making, with communication barriers such as language leading to non-consented procedures. Migrant women using urban clinics reported lower autonomy and confidentiality compared with those using rural refugee clinics.

Disability was underexplored, yet available data highlight significant barriers. Noori et al. [12], found that Nepali women with disabilities reported frequent neglect, infrastructural inaccessibility, and provider assumptions about incompetence. Despite providers’ intentions, the lack of inclusive infrastructure and training undermined autonomy and quality of care.

Intersectional Disadvantage: Multiple studies underscored that compounded identities (e.g., being young, poor, and from a lower caste) exponentially increased the risk of mistreatment. Gurung et al. [10] showed that Dalit adolescents had higher mistreatment scores. Amroussia et al. [8] demonstrated that young, unmarried, low-income mothers in Tunisia were especially stigmatized. While few studies statistically examined multi-axis interactions, qualitative findings consistently indicated layered disadvantage.

Cross-Cutting Mechanisms: Three themes emerged in explaining how identity influenced mistreatment: a) Stigma and Labeling: Providers projected moral and social judgments, especially toward unmarried mothers and lower-caste women [8, 10]. Deservingness/Differential expectations: Providers appeared to prioritize care based on perceived social value. Wealthy or educated women were attended to first or treated with greater deference [13]. These attitudes result in neglect or even punitive care (like withholding pain relief or shouting). c) Communication Barriers: Language differences, illiteracy, and provider assumptions led to non-consented care and diminished autonomy [3, 28]. These findings suggest that respectful maternity care in LMICs is structured by social hierarchies. Women with intersecting vulnerabilities experience care that is less respectful, less responsive, and more abusive. These disparities are not anomalies but patterned outcomes of systemic and interpersonal inequities within facility-based childbirth care. Figure 2 illustrates the conceptual model summarizing how social identities intersect to shape women’s experiences of respectful or disrespectful maternity care. The model depicts three levels, individual, provider, and system. The framework shows that intersecting factors such as poverty, caste, and disability converge through mechanisms of stigma, provider bias, and structural resource inequity to influence observed outcomes.

Discussion

This review set out to examine how intersecting social identities shape women’s experiences of respectful or disrespectful maternity care (RMC/D&A) during facility-based childbirth in LMICs. The results provide compelling evidence that RMC has yet to be universally experienced and that disparities are structured along lines of social power, including caste, ethnicity, socioeconomic status, age, marital status, migration status, literacy, and disability. Across diverse LMIC contexts, women at marginalized intersections routinely face neglect, verbal abuse, consent violations, and exclusion from decision-making, while women occupying more privileged social positions are more likely to receive courteous and supportive care [3, 8, 10, 24].

Our conceptual model (Fig. 2) offers an explanatory framework for these patterns. It illustrates how individual identity axes (such as caste, disability, age, migration status or literacy) interact with structural drivers such as provider biases, resource constraints, and facility norms, in shaping respectful or disrespectful care. By mapping these interactions, the model underscores that vulnerabilities are multiplicative vis-a-vis layered identities intensify exposure to stigmatization, neglect, and abuse. It also identifies leverage points for intervention, including anti‑bias training, community oversight, inclusive infrastructure and accountability mechanisms. Thus, the model translates intersectional theory into actionable strategies for improving respectful maternity care.

These dynamics are ethical violations that deter care-seeking. Multiple studies in this review suggest that women who perceive themselves to be at risk of being disrespected delay facility visits, opt for home birth, or avoid prenatal care altogether [15, 16]. The fear of being shamed or mistreated creates a vicious cycle as vulnerable women disengage from services, reinforcing their risk of complications and poor outcomes.

Some findings warrant deeper interpretation. For example, the presence of midwives in rural Thailand was associated with higher satisfaction and more respectful treatment than doctor-led care in urban settings [26]. This may reflect closer cultural congruence and less hierarchical communication styles. Similarly, reports from refugee camps in Chad show that even in resource-poor settings, respectful care is possible when providers are empathetic, supported, and community-embedded [28].

But why does mistreatment persist, even in settings where resources are available? One explanation is the failure to recognize bias as a clinical issue. Disrespectful care is often framed as a provider attitude problem, not as a systemic failure. Yet, as Altman et al. [18] and Manning and Schaaf [31] highlight how intersectional disrespect is built into institutional routines, i.e., who is listened to, who is asked for consent, who is allowed to labour with a companion. Until health systems integrate measures of equity into quality benchmarks, the problem will remain hidden.

Addressing these patterns requires multi-layered responses. Health systems must go beyond individual training and implement structural safeguards such as anonymous reporting systems, community-led audits, and subgroup-specific quality monitoring. Some promising interventions are emerging. For instance, Lusambili et al. [22] describe community oversight mechanisms in rural Kenya, while Green et al. [9] propose a “Cycle to Respectful Care” framework rooted in anti-racist and intersectional praxis. These approaches emphasize co-creation with affected communities and continuous feedback loops. In parallel, international rights frameworks should be more forcefully operationalized. Gameiro [32] conducted a legal analysis of Angola and Mozambique and found that failure to ensure RMC constitutes a breach of global human rights norms. Incorporating such legal accountability into maternal health programming, especially for LMICs, can drive more robust protections for vulnerable women in accessing respectful care.

This review has several strengths. It integrates data from 17 peer-reviewed studies across diverse LMICs and supplements findings with a global literature base spanning high-, middle-, and low-income contexts. Its intersectional lens unravels within-country inequalities that transcend across-country averages. Some identity categories such as refugee status, or intellectual disability, are grossly underexplored in LMIC research. There exists a dependence on self-reported instances of mistreatment, which may be underreported due to normalcy or apprehension. Ultimately, numerous investigations were insufficiently powered to analyze multi-factor interactions, hence constraining our capacity to statistically identify intersectional compounding effects.

This review has several limitations. Most included studies were cross-sectional, limiting causal interpretation. Evidence on disability, religion, and intersecting identities remains sparse. Publication bias toward facility-based samples and heterogeneity of RMC measures constrain generalizability. The inclusion of two review papers was justified conceptually but may introduce interpretive bias. Few studies were sufficiently powered to analyse interactions across multiple identity axes. Consequently, our synthesis relied heavily on qualitative interpretation of layered disadvantage. The search strategy was confined to three major databases and English-language publications due to resource constraints; therefore, relevant studies published in other languages or indexed elsewhere may have been missed. Publication bias may be present, as mistreatment and social marginalization may be underreported or unpublished, particularly in under-resourced settings. Heterogeneity in study designs and outcome measures precluded formal meta-analysis and limited comparability. Finally, certain identities such as refugee status, and intellectual disability were either absent or poorly represented in the included studies, pointing to critical gaps in the evidence base. Moreover, the qualitative narratives elucidate vivid, emotional, and detailed framings that are anchored in lived experience. From the single mother in Tunisia who said, “the doctor is God to punish me” [8] to the Indigenous woman in the U.S. who reported being “treated like an animal” [33], the human toll and its direct effect to facility-based births are undeniable.

Policy implications

The findings of this review highlight several actionable policy implications. First, governments and health systems should institutionalize intersectionality in maternal health policy by routinely collecting and analysing data disaggregated by social identities (caste, ethnicity, socio‑economic status, age, migration status, disability and religion) and using these data to monitor equity gaps. Second, health worker training should incorporate anti‑bias curricula and respectful maternity care standards tailored to address specific forms of discrimination (e.g., casteism, ableism, xenophobia). Third, facilities should implement accountability mechanisms such as anonymous reporting systems, community‑led scorecards, and regular audits that involve women from marginalized groups. Fourth, investments in infrastructure must ensure physical accessibility and privacy for women with disabilities, while language interpretation services should be provided for migrants and refugees. Fifth, policy frameworks should align with human rights obligations by explicitly prohibiting mistreatment and linking respectful care to accreditation and funding. Finally, multi‑sectoral collaborations with community organizations, legal advocates, and policymakers are needed to design and evaluate interventions that address the structural determinants of disrespect and abuse. These actions can transform respectful maternity care from a discretionary practice into an enforceable standard.

In conclusion, respectful maternity care in LMICs is inextricably linked to overarching issues of justice, equity, and power dynamics. Health institutions embody societal hierarchies, which are perpetuated in the treatment of women during one of the most vulnerable periods of their life, child birth. Mitigating these disparities necessitates not merely professional reform but a comprehensive health systems transformation that prioritizes the voices, rights, and dignity of the most vulnerable populations to ensure equitable access to respectful and dignified care across public and private health facilities in LMICs.

Abbreviations

RMC

Respectful Maternity Care

D&A

Disrespect and Abuse

LMICs

Low- and Middle-Income Countries

WHO

World Health Organization

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO

International Prospective Register of Systematic Reviews

PCMC

Person-Centered Maternity Care

AOR

Adjusted Odds Ratio

CASP

Critical Appraisal Skills Programme

CDC

Centers for Disease Control and Prevention

HIV

Human Immunodeficiency Virus

Authors' contributions

VAA conceived the study. VAA, CO and AA designed the search strategy. VAA, CO and AA conducted the search, study selection, and data extraction. VAA, CO, AA, and QESA wrote the first draft and critically revised the manuscript. The authors read and approved the final version of the manuscript.

Funding

This manuscript received no grants from any public, commercial, or not-for-profit funding agency.

Data availability

All data generated or analysed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.