Labor support in low- and middle-income countries - transformative impact of birth companions and doulas on maternal and neonatal outcomes: a narrative review

Abstract

Background

Maternal and neonatal morbidity and mortality remain disproportionately high in low- and middle-income countries (LMICs), partly due to inadequate intrapartum support, psychosocial stress, and disrespectful maternity care. Continuous labor support from birth companions or doulas offers a low-cost, high-impact intervention with demonstrated benefits in high-income settings, yet LMIC-focused evidence remains fragmented. This review synthesizes current evidence on the effects of birth companionship and doula care on maternal, neonatal, psychosocial, and implementation outcomes in LMICs.

Methods

A narrative review was conducted using structured searches of PubMed, Scopus, Web of Science, Lens.org, and WHO documents. Eligible studies were published between 2015 and 2025, conducted in LMICs or relevant to LMIC contexts, and reported empirical findings on continuous intrapartum support from informal companions, trained lay companions, or professional doulas. Data were synthesized thematically across maternal, neonatal, psychosocial, and implementation domains. A structured narrative appraisal was used due to heterogeneity in study design and outcome measures.

Results

A total of 27 studies were included in the narrative synthesis. These comprised 8 primary studies in LMICs, 6 high-income country (HIC) studies involving marginalized populations, and 13 systematic reviews. Continuous labor support was consistently associated with shorter labor, higher rates of spontaneous vaginal birth, reduced use of pharmacologic analgesia and oxytocin augmentation, and lower maternal anxiety and stress markers. Neonatal benefits included higher Apgar scores, earlier breastfeeding initiation, and enhanced mother–infant bonding. Psychosocial improvements were mediated through emotional reassurance, advocacy, improved communication, and culturally responsive support. Implementation barriers identified across LMIC studies included provider resistance, facility space limitations, inadequate training, and sociocultural constraints, while facilitators included community engagement, structured companion training, and phased integration into maternity systems. Evidence quality varied widely, with few LMIC-specific randomized trials and heterogeneous operational definitions of “continuous support.”

Conclusions

Evidence demonstrates that birth companions and doulas improve maternal well-being, early neonatal outcomes, and respectful maternity care in LMIC settings. Effective scale-up requires context-adapted implementation strategies, provider orientation, policy alignment, and rigorous LMIC-based trials assessing clinical outcomes, cost-effectiveness, and equity impacts. Companion-inclusive intrapartum care represents a feasible, low-cost strategy for strengthening quality and experience of maternity care in resource-limited health systems.

Background

Despite global progress, many low- and middle-income countries (LMICs) continue to record unacceptably high rates of maternal and neonatal morbidity and mortality, with a substantial proportion attributable to preventable causes and suboptimal intrapartum care. Traditional forms of community-based labor support were historically central to childbirth in many settings but have been marginalized by medicalized, institutional care models that often do not address psychosocial needs or respect women’s autonomy [1, 2].

Systematic reviews and randomized controlled trials (RCTs) from high-income countries (HICs) have established that continuous labor support reduces cesarean section rates, shortens labor duration, and increases women’s satisfaction with birth experiences. A Cochrane review by Bohren et al. demonstrated that continuous support reduced cesarean births by 25% (RR 0.75, 95% CI 0.64–0.88) and increased spontaneous vaginal birth rates (RR 1.08, 95% CI 1.04–1.12) [3]. Despite this robust evidence, translation into LMIC policy and practice remains limited. A recent scoping review found that only 15–30% of facilities in sub-Saharan Africa permit companion of choice during labor [4], suggesting substantial implementation gaps. Furthermore, systematic reviews synthesizing LMIC-specific evidence are scarce, with most existing reviews aggregating predominantly HIC data, creating uncertainty for policymakers in resource-constrained settings.

According to the World Health Organization (WHO), women have generally been encouraged to give birth in health care facilities instead of the home to ensure access to health care professionals and timely interventions [5]. However, disrespectful care is common in underprivileged populations, which may pose additional barriers and lead to poor maternal outcomes [6]. Increasingly, the literature highlights continuous labor support whether from informal companions, trained lay companions, or professional doulas as a low-cost, high-impact strategy that can simultaneously improve clinical outcomes and the birthing experience [5]. Although systematic reviews and randomized trials largely conducted in high-income settings demonstrate benefits of continuous labor support, translation to LMIC policy and practice remains unbalanced [7]. The overall aim of this narrative review is to synthesize available evidence on birth companions and doula care in LMIC contexts. We specifically hope to examine the effects of birth companions and doulas on maternal clinical outcomes, to assess the impact of birth companions on neonatal outcomes, identify barriers and facilitators to implementing birth companionship programs in LMIC health systems, and provide evidence-based recommendations for policy integration and future research.

Methods

Study design

This narrative review employed a structured synthesis approach to examine the evidence on birth companions and doula care in LMICs. A narrative review was particularly suitable for this complex topic where heterogeneous study designs and contexts required interpretive synthesis rather than statistical pooling. Secondly, the objectives encompass diverse clinical outcomes, psychosocial mechanisms, implementation barriers, and policy considerations requiring interpretive synthesis rather than meta-analysis. Third, narrative reviews allow examination of both ‘what works’ (effectiveness) and ‘how it works’ (mechanisms), which is essential for translating evidence into context-specific implementation strategies. While lacking the quantitative precision of systematic reviews, this approach provides comprehensive understanding of a complex, context-dependent intervention.

Research question

The PICO framework was used to develop the research question as follows:

• Population (P): Women giving birth in LMICs as defined by World Bank income classification (low-income and lower-middle-income economies), or marginalized populations in HICs facing similar barriers (language barriers, cultural disconnection from health systems, discrimination);

• Intervention (I): continuous labor support during the intrapartum period, including:

  • Informal birth companions: Family members, partners, or friends chosen by the woman, without formal training
  • Trained lay companions: Community members with structured training in labor support techniques
  • Professional doulas: Individuals with formal certification in labor support through standardized curricula

• Comparison (C): Standard maternity care without continuous labor support (where applicable in comparative studies)

• Outcomes (O):

  • Maternal clinical outcomes: labor duration, mode of delivery, use of interventions (oxytocin, analgesia, cesarean section)
  • Neonatal outcomes: Apgar scores, breastfeeding initiation, neonatal intensive care admission
  • Psychosocial outcomes: maternal satisfaction, anxiety, perceived respect, person-centered care
  • Implementation outcomes: barriers and facilitators to program adoption, feasibility, acceptability, costs

Search strategy

A comprehensive literature search was conducted across four electronic databases including PubMed, Scopus, Web of Science, and Lens.org between November 15–22, 2024. Two reviewers (PA and JK) independently designed and executed the searches. For each database, we combined index terms and keywords using Boolean operators (AND, OR) to capture three concept domains: (1) population (LMIC contexts), (2) intervention (birth companions, doulas, labor support), and (3) outcomes (maternal health, neonatal outcomes). Search results were exported to EndNote 20 for duplicate removal. Complete search strings are provided in Table 1.

Table 1.

Database Search Strategies

Database	Complete Search String	Results
PubMed	((“labor support” OR “labour support” OR doula* OR “birth companion*” OR “labor companion*” OR “continuous support” OR “intrapartum support”) AND (“low income” OR “middle income” OR LMIC* OR “developing country*” OR “resource limited” OR Africa OR Asia OR “Latin America”) AND (“maternal outcome*” OR “neonatal outcome*” OR “birth outcome*” OR “cesarean” OR mortality OR morbidity)) AND (2014:2025[pdat])	104
Web of Science	TS=((“labor support” OR “labour support” OR doula* OR “birth companion*” OR “continuous support”) AND (“low-income” OR “middle-income” OR “low income” OR “middle income” OR LMIC* OR “developing country” OR “developing countries” OR “resource-limited” OR “resource limited”) AND (“maternal outcome” OR “maternal outcomes” OR “neonatal outcome” OR “neonatal outcomes” OR “birth outcome” OR “birth outcomes” OR cesarean OR caesarean OR mortality OR morbidity)) AND PY=(2014–2025)	99
Scopus	TITLE-ABS-KEY ( “labor support” OR “labour support” OR doula OR doulas OR “birth companion” OR “birth companions” OR “labor companion” OR “labour companion” OR “continuous support” OR “intrapartum support” ) AND TITLE-ABS-KEY ( “low-income” OR “middle-income” OR “low income” OR “middle income” OR lmic OR lmics OR “developing country” OR “developing countries” OR “developing nation” OR “resource-limited” OR “resource limited” OR “low resource” ) AND PUBYEAR > 2013 AND PUBYEAR < 2026 AND PUBYEAR > 2014 AND PUBYEAR < 2026	111
Lens.org	(“labor support” OR doula OR “birth companion”) AND (“low income country” OR “middle income country” OR LMIC) AND (“maternal outcome” OR “neonatal outcome” OR mortality)	386

MeSH terms and field codes preserved as executed in each database interface

Inclusion criteria

Peer reviewed and grey literature written in the English language and published between January 2015, and August 2025, were included. This timeframe was selected to capture evidence published after the 2015 WHO intrapartum care guidelines [5], which explicitly recommended companion of choice during labor. RCTs, systematic reviews, meta-analyses, observational studies (cohort, case-control, cross-sectional), qualitative studies, mixed-methods studies, and implementation reports were included. Only studies conducted in LMICs as defined by World Bank classification, studies from HICs explicitly focused on migrant, refugee, or marginalized populations experiencing barriers similar to those documented in LMICs (e.g., language barriers, cultural disconnection from health systems, experiences of discrimination), were included. Studies that focused on any form of continuous labor support including informal companions, trained lay supporters, or professional doulas were included. Studies had to discuss maternal clinical outcomes, neonatal outcomes, psychosocial outcomes, implementation factors, or policy-related findings. Only studies focusing on the intrapartum period (during labor and birth) were included.

Exclusion criteria

The following articles were excluded: opinion pieces or commentaries without empirical data, studies conducted exclusively in high-income countries without LMIC relevance, studies focusing solely on antenatal or postnatal support without intrapartum components, and conference abstracts without full-text availability. Studies addressing only antenatal education or postnatal support were also excluded.

Data extraction and synthesis

Extracted information included population/context, interventions (informal companion, trained lay companion, or doula), outcome measures (maternal, neonatal, psychosocial), and implementation observations. Findings were synthesized thematically rather than pooled quantitatively, given heterogeneity in study designs, settings, and outcome definitions. The synthesis emphasized effect direction and consistency, contextual moderators, and gaps in implementation.

Quality appraisal

Study quality was independently assessed by two reviewers (PA and CL) using validated, design-appropriate tools:

• Randomized controlled trials: Cochrane Risk of Bias 2 (RoB 2) tool

• Observational studies: Newcastle-Ottawa Scale (NOS) or Joanna Briggs Institute (JBI) critical appraisal checklists

• Qualitative studies: Consolidated Criteria for Reporting Qualitative Research (COREQ) or Critical Appraisal Skills Programme (CASP) checklist.

• Systematic reviews: AMSTAR 2 (A MeaSurement Tool to Assess Systematic Reviews)

Disagreements were resolved through discussion or consultation with a third reviewer (EO). Quality ratings were used descriptively to contextualize findings rather than as exclusion criteria.

Results

The flow diagram of the 27 included studies is provided in Fig. 1. Of the included studies, 8 primary studies were conducted in LMICs, 6 were conducted in HICs and involved marginalized populations, and 13 were systematic reviews. Table 2 provides a summary of the studies conducted in LMICs. Tables 3 and 4 provide a summary of studies conducted in HICs and systematic reviews.

Fig. 1.

PRISMA Flow Diagram Demonstrating the Article Screening and Selection Process

Table 2.

Primary Studies from LMICs Included in Narrative Synthesis

Author, Year	Country	Study Design	Sample Size	Intervention Type	Outcomes Measured	Key Findings	Quality Assessment
Kiti et al. 2022 [8]	Kenya	Cross-sectional	865	Continuous support informal companions	Person-Centered Maternity Care (PCMC) score	Continuous support increases PCMC scores	Moderate (STROBE guidelines)
Getahun et al. 2020 [9]	Ethiopia	Cross-sectional	418	Birth companionship (informal)	Utilization of birth companionship	Utilization rate: 58.2%	Moderate (JBI checklist)
Doba et al. 2023 [10]	Ethiopia	Cross-sectional	406	Birth companionship (informal)	Utilization of birth companionship	Utilization rate: 19.5%	Low risk of bias (JBI checklist)
Mazimpaka et al. 2024 [11]	Rwanda	Cross-sectional	422	Birth companionship (informal)	Utilization of birth companionship	Utilization rate: 15.0%	Low risk of bias (JBI checklist)
Mirzania et al. 2024 [12]	Iran	Qualitative (in-depth interviews)	Not specified	Birth companionship (training not specified); focus on psychosocial support	Barriers and facilitators to implementation; perceived usefulness	Companionship appropriate for overcrowded settings; companions assist with massage, emotional support; High trustworthiness	High (Lincoln & Guba criteria; COREQ)
Khan et al. 2023 [13]	Pakistan	Feasibility & development study	98	Psychosocial support; defining companion’s role	Feasibility, acceptability of WHO adaptation	Cultural adaptations required; preference for female companion; Companions helpful in resource-limited settings	High (HCD methodology)
Kamath et al. 2022 [14]	India	RCT	768 (estimated)	Peanut ball device during active labor vs. standard care	Maternal & neonatal outcomes (labor duration, mode of delivery, cortisol, satisfaction)	N/A	High (CONSORT & SPIRIT guidelines)
Berhanu et al. 2023 [15]	Ethiopia, Kenya, Nigeria	Cluster RCT Protocol	744 per site	Trained birth companions supporting during labor, birth, postnatal period	Coverage of birth companionship; PCMC score; acceptability; facility readiness; costs	N/A	High (Cochrane RoB 2)

Kamath et al. [14] and Berhanu et al. [15] are trial protocols without outcome data and included for contextual information on intervention design and baseline utilization rates only. Quality was assessed using Cochrane RoB 2 for RCT protocols, NOS for observational studies, JBI checklists for cross-sectional studies, and COREQ for qualitative studies

Table 3.

High-Income Country (HIC) Studies

Author, Year	Country/Region	Study Design	Intervention / Type of Support	Outcomes Measured	Key Findings	Quality Assessment Tool & Score	Limitations
Mok et al. 2022 [16]	Hong Kong, China	Retrospective cohort	Continuous companion support during labor vs. no support	Mode of delivery; first-stage duration; analgesic use; neonatal outcomes; breastfeeding rate	No significant differences between mode of delivery, first-stage duration; C-section rates similar	No formal NOS/RoB tool reported	Retrospective design; comparison across pandemic periods; relatively small sample
Batram-Zantvoort et al. 2025 [17]	WHO European Region	Cross-sectional analysis	Presence of birth companion of choice	Quality of maternal & newborn care; dignity; communication; emotional support; abuse	Companion presence associated with better QMNC indicators	STROBE guidelines used; no NOS score	Self-reported data; pandemic-related bias; residual confounding likely
Purandara et al. 2024 [18]	Sweden		CBD support before, during, after birth; culturally-tailored informational and emotional support	Women’s experiences of CBD support	CBDs improved safety, preparedness, communication, and mitigated cultural/linguistic barriers	Mixed-Methods Design	Limited generalizability to migrant populations; incomplete reporting of limitations
Khaw et al. 2023 [19]	Australia	Qualitative study	Volunteer/bicultural doula support for migrant/refugee background clients	Doulas’ perspectives on culturally responsive care.	Bicultural doulas facilitated culturally safe maternity care	SRQR guidelines used	Focus limited to doula perspectives; majority cisgender clients
Elgzar et al. 2024 [20]	Saudi Arabia	Cross-sectional study	Utilization of non-pharmacological labor pain relief methods	Utilization patterns; predictors such as training, workload, ratio, experience, attitude	Significant predictors: provider ratio 1:6 (AOR 0.165), ratio 1:8 (AOR 0.155)	N/A	Cross-sectional design; self-reported behaviors; recall/reporting bias
Stjernholm et al. 2021 [21]	Sweden	RCT	Continuous midwife/nurse support during active labor vs. intermittent standard care (5–15 min)	Duration of active labor; delivery mode; analgesic use; maternal cortisol; satisfaction; neonatal morbidity	Continuous support shortened active labor; Higher vaginal delivery rate; Lower maternal cortisol correlated with shorter labor; Greater combined analgesia use	Low-to-moderate RoB due to unblinded allocation	No standardized training for support providers; intervention not uniform; sample underpowered for secondary outcomes

These studies from HICs were included due to focus on migrant, refugee, or marginalized populations experiencing barriers analogous to LMIC contexts. Quality was assessed using NOS for observational studies, STROBE guidelines for reporting quality, and Mixed-Methods Appraisal Tool (MMAT) where applicable

Table 4.

Summary of Systematic and Narrative Reviews Included

Author(s), Year	Country/Region	Study Details	Intervention / Type of Support	Outcomes Measured	Key Findings	Quality Assessment Tool & Score	Limitations
Gudeta et al., 2024 [1]	Sub-Saharan Africa (LMICs)	Systematic review & meta-analysis of 16 observational studies	Birth companionship (spouse/partner, friend, relative, community member, doula)	Utilization of birth companionship; associated factors	Utilization remained low across countries	JBI checklist; mixed quality (50–70%)	Cross-sectional design; substantial heterogeneity; English-only restriction
Leutenegger et al., 2022 [22]	Global	Systematic review of 4 controlled trials on antenatal education	Antenatal education (breathing/relaxation, partner coaching, multicomponent programs)	Spontaneous vaginal delivery, birth experience, labor duration, pain scores	Improved spontaneous vaginal delivery rates	Cochrane RoB; moderate–low quality	Heterogeneous content; blinding/reporting biases; evidence inconsistent
Sangy et al., 2023 [23]	21 LMICs	Mixed-methods systematic review	Midwife-led care	Barriers and facilitators to midwife-led care adoption	Common barriers: socio-cultural norms, physician resistance, poor referral systems, facility constraints	MMAT 2018; mostly high quality	Lack of independent quality appraisal; results heavily context-dependent
Evans et al., 2023 [2]	OECD Countries	Mixed-methods systematic review	Non-professional companions (family, partner, friend)	Experiences of women, companions, and midwives	Themes: feeling supported, overwhelmed, or “in the way.”	JBI appraisal; observational designs strong; others moderate–low	No trials for meta-analysis; heterogeneity; only OECD settings
Martin et al., 2023 [24]	High-income countries	Systematic review of economic evaluations of midwifery models of care	Midwife-led care vs. physician-led; doula-supported care	Cost, preterm birth, NICU admission, obstetric interventions	Midwife-led care cost-savings; doula models cost-effective	Economic evaluation checklist (scores 5.5–8/11).	Lack of standardized MMC definitions; methodological limitations
Alfirevic et al., 2020 [7]	High-/Upper-Middle-Income	Cochrane review of 7 RCTs	Home vs. inpatient induction using PGE or Foley catheter	Pain, hyperstimulation, C-section, neonatal infection, NICU admission, satisfaction	Slight ↑ satisfaction (one study); ↑ opioid use at home; no difference in C-section rates	GRADE + Cochrane RoB.	High imprecision, blinding risk, heterogeneity
Middleton et al., 2020 [25]	Global	Cochrane RCT review	Induction of labour at ≥ 37 weeks	Perinatal death, stillbirth, NICU, PPH, C-section	IOL reduced perinatal death & stillbirth; C-section unchanged.	GRADE; frequent downgrading.	Publication bias risk; heterogeneity; limited blinding.
Jayasundara et al., 2024 [26]	Global	Systematic review & meta-analysis of 35 RCTs	Labour companions (trained/untrained; familiar/unfamiliar)	Spontaneous vaginal delivery, analgesia, oxytocin, labour duration, C-section, instrumental birth, Apgar < 7, tocophobia.	Strongest effect for reducing poor Apgar scores and reducing tocophobia with familiar companions	Cochrane RoB; performance bias high	High heterogeneity; limited subgroups
Chang et al., 2022 [27]	Global	Meta-analysis of 8 trials	Non-pharmacologic pain, coping	Pain reduction	Mixed effectiveness with significant heterogeneity; no publication bias	Cochrane RoB; some detection/performance bias	Only 8 studies; English-only restriction
Bohren et al., 2014 [6]	Global	Mixed-methods systematic review	Not applicable.	Forms of mistreatment; typologies of abuse and disrespect	Developed an evidence-based classification of mistreatment	ENTREQ used; quality assessment of included studies reported	Specific limitations not clearly reported; reported heterogeneity
Bohren et al., 2017 [3]	Global	Cochrane Review	Continuous labour support (staff, chosen non-staff, or non-chosen non-staff)	Spontaneous vaginal delivery, C-section, negative birth experience, postpartum depression	Continuous support decreased C-section & negative birth experience	GRADE; evidence downgraded for design flaws, blinding bias, heterogeneity	Difficult planned subgroup analyses; high heterogeneity for pain outcomes
Bohren et al., 2023 [4]	Global	Scoping review on global policy and implementation of “companion of choice”	Companion of choice during labour and birth	Coverage and implementation of companion-of-choice policy	Showed wide policy–practice gap; implementation lag despite global recommendations	Not applicable	Poor global implementation; limited data on actual coverage
Oliveira et al., 2022 [28]	High/middle income	Synthesis integrated four systematic reviews/meta-analyses	Continuous support during labor and birth; Support provided by a doula, spouse/partner, healthcare professional, family member, friend, or student	Cesarean section rates; duration of labor; unnecessary clinical interventions (e.g., analgesia, oxytocin); low Apgar score at five minutes; maternal satisfaction/recovery; breastfeeding	Continuous support reduced cesarean rates, duration of labor, need for unnecessary clinical interventions, and lowered the probability of a low Apgar score	Quality of included reviews assessed via AMSTAR Two reviews were rated high quality, one moderate, and one low	Most data originated from high-income countries

Thematic synthesis of the supplied literature is presented under five headings: (1) maternal clinical outcomes; (2) neonatal outcomes; (3) psychosocial mechanisms and communication; (4) implementation barriers and facilitators; and (5) policy and system integration. A conceptual visual summary is provided in Fig. 2.

Fig. 2.

Impact of birth companions and doulas on maternal and fetal outcomes

Maternal clinical outcomes

Evidence from systematic reviews consistently demonstrated that continuous labor support was associated with reduced active labor duration and higher rates of spontaneous vaginal birth. A meta-analysis of Brazilian studies found that continuous support reduced cesarean section rates [28]. A global systematic review and meta-analysis by Jayasundara et al., synthesizing 35 RCTs, reported that labor companions improved multiple outcomes, with the strongest effect on reducing the fear of childbirth [26]. However, high heterogeneity was noted across outcomes.

Regarding interventions, systematic reviews suggested associations between continuous support and reduced analgesic use, though causality cannot be definitively established from predominantly observational LMIC studies. The qualitative study by Mirzania et al. in Iran noted that companions provided massage and emotional support, which may contribute to non-pharmacological pain management [12].

There were no primary LMIC studies that directly compared trained versus informal companions in maternal clinical outcome analyses. Most studies examined informal companions [8–11] or feasibility of trained companions [12, 13]. Protocol studies propose testing trained companions but have not yet reported outcomes [14, 15]. Among systematic reviews, Jayasundara et al. found familiar companions had the strongest effect on reducing tocophobia (ES = 1.73), while Bohren et al. reported best cesarean reduction with companions chosen by the woman (RR 0.58 vs. 0.75 overall) [3]. Kiti et al. found in a Kenyan cross-sectional study (n = 865) that continuous support by informal companions (partner, mother-in-law, mother, sister, friend) was associated with significantly higher person-centered maternity care (PCMC) scores (25.7 vs. 21.0; p < 0.001; adjusted β = 4.04, 95% CI 2.87–5.20) [8]. This suggests that companionship improves the experiential quality of care, even in observational designs.

In regard to psychological outcomes, supported women demonstrated lower anxiety, lower physiological stress markers (e.g., cortisol), higher satisfaction, and greater empowerment [20, 21]. One high-quality RCT from Sweden demonstrated that continuous support reduced labor duration and promoted vaginal delivery in primiparous women, with associated reductions in cortisol levels [21]. This physiological mechanism may explain observed clinical benefits. Doulas and trained companions enhance perceived control during labor and help women articulate preferences and consent for important components of respectful maternity care [13].

Neonatal outcomes

The impact of continuous labor support on neonatal outcomes is impressive. Multiple studies report improvement in Apgar scores and the initiation of early breastfeeding. Babies born to women with continuous labor support were noted to have higher Apgar scores than those who did not [16, 28]. Jayasundara et al. identified labor companion support as having the strongest effect on reducing Apgar scores (ES = 1.52), suggesting an improved immediate neonatal condition [26]. Increased rates of early breastfeeding initiation was also noted when mothers received continuous support [16, 28]. Early breastfeeding is a key survival intervention and may be facilitated by supportive postpartum assistance. However, a retrospective study in Hong Kong comparing women with versus without continuous companion support during the COVID-19 pandemic found no significant differences in neonatal outcomes or breastfeeding rates, though this was a high-income setting with already-optimized care [16].

Evidence for reductions in broader neonatal morbidity is promising but less conclusive. Although some studies show a lower incidence of low Apgar scores and physiological stress markers, long-term neonatal outcomes are under-studied in LMIC contexts [14, 21]. There was no demonstrated decrease in neonatal deaths in women receiving labor support as compared to those who did not [21]. Neonatal benefits observed in the compiled studies are plausible mechanistically (reduced maternal stress, fewer interventions) and supported by short-term outcomes.

Psychosocial mechanisms and communication

The literature emphasizes several interlocking psychosocial mechanisms. Emotional support reduces maternal anxiety and stress. Advocacy ensures clearer communication with clinical staff, and physical comfort strategies (positioning, massage, breathing guidance) improve coping and labor progress [17, 22, 29]. Bilingual or community-based companions bridge linguistic and cultural gaps, improving shared decision-making and perceived respect [18, 30]. Continuous support maps directly onto the respectful maternity care (RMC) framework enhancing dignity, autonomy, and effective communication, which in turn relates to uptake of facility-based births and satisfaction metrics [8, 19]. Doulas have been shown to counter negative factors that may preclude women from seeking maternal care [19]. The presence of doulas or labor companions enabled effective communication between women and health care providers, overall increasing a woman’s satisfaction and decreasing mistreatment and adverse birth outcomes [8].

Acceptance of birth companions and doulas depends on cultural norms, family dynamics, and women’s expectations. Traditional support may exist but is not always compatible with facility routines or provider expectations [23]. Provider resistance based on role confusion, concerns about interference, or perceived professional threats emerge as a recurrent barrier. Facility space, privacy concerns, workforce shortages, and funding constraints complicate integration into widespread practice [12, 23].

Training programs for companions and providers, clear role delineation, incremental pilot implementation, and use of community resources reduce friction. Case studies and pilot trials in Ethiopia, Kenya, Nigeria, and Tehran illustrate strategies [12, 15]. Adequate training of birth companions is paramount, and women should be informed about the role of birth companions [12]. Women seen by health care providers trained in respectful maternity care or trained birth companions had higher birth satisfaction scores [31].

Implementation barriers and facilitators

LMIC implementation studies revealed multilevel barriers and facilitators. The qualitative study in Iran identified that birth companionship was considered appropriate for overcrowded settings, with companions assisting through massage and emotional support [12]. However, the study noted cultural norms and facility routines as potential barriers. A feasibility study in Pakistan training 98 staff in psychosocial support found that cultural adaptations were essential, with preference for female companions in that context [13].

The mixed-methods systematic review by Sangy et al., examining midwife-led care implementation in 21 LMICs, identified common barriers: sociocultural norms, physician resistance, poor referral systems, and facility constraints (space, privacy, workforce shortages) [23]. The review rated 23 studies as high quality, 7 as moderate, and 1 as low, providing high confidence in these findings.

Cross-sectional studies from Ethiopia and Rwanda documented low baseline utilization rates (58.2%, 19.5%, and 15.0%, respectively), suggesting substantial implementation gaps despite policy recommendations [9–11].

Facilitators identified included: structured training programs for companions and providers, clear role delineation, incremental pilot implementation, and leveraging community resources. The Berhanu et al. protocol for cluster RCTs in Ethiopia, Kenya, and Nigeria proposes evaluating these strategies through standardized operating procedures and training videos [15].

Policy and system integration

Baseline data from Berhanu et al. indicated birth companionship coverage of only 5–20% across study sites in Ethiopia, Kenya, and Nigeria [15], despite WHO 2018 intrapartum care recommendations supporting companion of choice [5]. A systematic analysis of historical data of health systems changes in 78 LMICs documented slow policy adoption of respectful maternity care principles between 2008 and 2018 [32].

Discussion

Principal findings and interpretation

This narrative synthesis of the supplied literature finds consistent, plausible benefits of birth companionship and doula care for maternal and neonatal short-term outcomes and for experiential dimensions of childbirth particularly in the domains of reduced labor duration, lower intervention rates, improved maternal psychological well-being, higher Apgar scores, and earlier breastfeeding initiation [16, 21, 28]. Mechanistically, emotional support, advocacy, improved communication, and non-pharmacologic physical strategies appear to mediate these effects [17]. These mechanisms map directly onto the core tenets of respectful, person-centered maternity care [4] and aligns with the WHO recommendations for a suitable companion to be present during labor and childbirth [5].

The professionalization of doulas requires training frameworks balanced with local adaptation, guidance on scope of practice, and sustainable financing models. Tiered certification and community-based doula models offer scalable options. Training midwives and doulas to international standards is emphasized [23]. A further tension arises in professional dynamics. Some clinician groups perceive doulas as encroaching on clinical authority. Successful integration requires delineation of roles, so doulas complement rather than replace clinical care [12, 23].

Key limitations in the compiled body of literature include heterogeneity in defining and operationalizing “support” (informal companion vs. trained doula vs. lay supporter), variability in outcome metrics, and predominance of high-income setting data for many RCTs [7]. LMIC-specific randomized evidence remains limited. Where present, studies often vary in training intensity and scope, complicating transferability. Cost-effectiveness and equity impacts are understudied. The degree to which doula programs reduce disparities rather than primarily benefiting those who already have access is not yet clear [26]. Policy traction is growing, but operationalizing doulas into routine health services demands investments in workforce development, regulation, and financing strategies. Evidence on long-term cost-effectiveness in LMICs is sparse and mixed. Economic evaluations remain a priority [24].

Implementation and policy implications

Current policy uptake remains inadequate. Despite WHO 2018 recommendations for companion of choice during labor [5], our review identified birth companionship coverage of only 5–20% in recent LMIC studies [9–11, 15]. This implementation gap highlights the need for concerted policy and programmatic action.

For policymakers and program managers in LMICs, the pragmatic message is that birth companions and doula models are relatively low-cost, acceptable, and plausibly effective. Strategies to advance scale include: (1) pilot and phased rollouts that adapt to local culture and infrastructure; (2) embedding training within existing community health initiatives; (3) clear role definition and provider sensitization; (4) inclusion of companion presence in facility checklists and quality improvement indicators; and (5) exploring sustainable financing, public funding, community health budgets, or targeted reimbursements [15].

Policy operationalization also requires addressing doula professionalization. Training frameworks balanced with local adaptation, scope of practice guidance, and sustainable financing models are paramount. Evidence on long-term cost-effectiveness in LMICs is sparse. A systematic review of economic evaluations found that midwife-led and doula-supported care models were cost-saving in high-income settings, but LMIC-specific economic evaluations are urgently needed [24].

Research agenda and priorities

Future research should prioritize context-specific randomized and hybrid implementation trials in LMICs using standardized outcomes such as labor duration, mode of birth, neonatal morbidity, breastfeeding initiation, and patient-reported experiences alongside process evaluations. Economic studies are needed to assess short- and long-term cost-effectiveness, and equity analyses should determine whether interventions reduce disparities among marginalized groups. Standardizing core outcome sets and definitions of “continuous support” would enhance comparability, while provider-focused implementation studies should examine training, acceptability, and health system adaptations for sustainable integration of doulas and birth companions. Specifically, controlled LMIC trials should compare structured birth-companion education interventions with usual care or untrained companionship and measure standardized maternal, neonatal, psychosocial, and implementation outcomes.

Broader implications for respectful maternity care

Integrating companion-based support aligns with rights-based frameworks and WHO recommendations and can act as a lever for broader system improvements [4]. As part of quality improvement initiatives, continuous support models can simultaneously improve clinical metrics and person-centered outcomes, synergies that are particularly valuable in resource-constrained settings where low-cost, high-return interventions are prioritized.

Strengths and limitations

This narrative review offers several methodological strengths that enhance its contribution to the literature on birth companionship in LMICs. First, the explicit focus on LMIC contexts addresses a critical gap, as most existing syntheses aggregate predominantly HIC data. Second, the review employed systematic search strategies across four major databases with transparent, reproducible inclusion criteria. Third, quality appraisal using validated, design-appropriate tools (RoB 2, NOS, JBI, COREQ, AMSTAR 2) provides context for interpreting findings and is explicitly reported for each study. Fourth, the multi-domain synthesis spanning clinical outcomes, psychosocial mechanisms, and implementation considerations offers actionable insights for diverse stakeholders (clinicians, policymakers, program managers, researchers). Finally, the integration of primary studies with systematic reviews balances depth with breadth, situating LMIC-specific findings within the global evidence base.

This narrative review has several limitations. First, the search was restricted to four databases and English-language publications, potentially excluding relevant studies from non-Anglophone LMIC regions published in regional databases or other languages. Second, as a narrative rather than systematic review with meta-analysis, the synthesis involved interpretive judgment rather than statistical pooling, precluding precise effect size estimates specific to LMICs. Third, while we used validated quality assessment tools, heterogeneity in study designs, companion types, and outcome definitions prevented formal GRADE evidence profiling. Fourth, the search was conducted in November 2024, and rapidly evolving literature may have produced additional relevant studies since then. Fifth, we did not formally assess publication bias, though the inclusion of grey literature and protocols may partially mitigate this concern. Finally, our final sample included only 8 LMIC primary studies, limiting the strength of LMIC-specific conclusions.

Conclusions

The compiled evidence supports the transformative potential of birth companions and doula care for improving maternal and neonatal short-term outcomes, birth experiences, and person-centered care in LMICs. Implementation success hinges on culturally sensitive program design, provider engagement, pragmatic training and certification frameworks, and clear policy support. Priority research includes LMIC-specific controlled trials comparing structured companion education interventions with usual care or untrained companionship, alongside implementation trials, economic studies, and equity analyses. Ministries of health and facility leaders should consider piloting and integrating companion models into respectful maternity care and quality improvement programs, guided by implementation science principles and community partnerships.

Abbreviations

AMSTAR 2

A MeaSurement Tool to Assess Systematic Reviews

CASP

Critical Appraisal Skills Programme

CONSORT

Consolidated Standards of Reporting Trials

COREQ

Consolidated Criteria for Reporting Qualitative Research

HIC

High-income country

JBI

Joanna Briggs Institute

LMIC

Low- and middle-income country

MeSH

Medical Subject Headings

NOS

Newcastle-Ottawa Scale

PCMC

Person-Centered Maternity Care

RCT

Randomized controlled trial

RMC

Respectful maternity care

SPIRIT

Standard Protocol Items: Recommendations for Interventional Trials

STROBE

STrengthening the Reporting of OBservational studies in Epidemiology

WHO

World Health Organization

Authors’ contributions

PA: conceptualization, data curation, writing – original draft; JaK: data curation, writing – original draft, writing – review and editing; JoK: conceptualization, data curation, writing – original draft; EN: conceptualization, data curation, writing – original draft; AAA: conceptualization, data curation, writing – original draft; CRL: formal analysis, writing – review and editing, quality assessment; EO: conceptualization, writing – review and editing, quality assessment, supervision.

Funding

The authors received no funding from either the public or private sector for this research.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.