Underpinned by a social constructivist paradigm, ¹⁵ this study employed a qualitative approach to explore how midwives, paramedics and medical officers (MOs) experience the phenomenon of upward referral of obstetric emergencies in their natural work settings. While the broader study followed a Grounded Theory design, ¹⁶ its use here is limited to its analytic strengths, guiding the systematic identification of antecedents and attributes without presenting the full emergent theory.

Despite a decrease in maternal deaths in healthcare facilities in the Eastern Cape, from 138 per 100 000 live births in 2017 to 128.9 per 100 000 in 2022, ¹⁷ districts with referral hospitals continue to experience challenges. Among these districts, OR Tambo is the most challenged district with the highest Maternal Mortality Rate (MMR), largely attributed to the presence of a highly specialised central hospital, within the district. ⁶

Within this district, a major sub-district which includes five Community Health Centres (CHCs), carries a disproportionately high MMR of 378.5 deaths per 100 000 live births. ¹⁸ Referral pathways in this area are often strained due to wide geographic coverage, limited transport infrastructure ¹⁹ and the urgent nature of obstetric emergencies. Accordingly, data collection for this study focuses on this high-burden sub-district to better understand and address the elevated maternal health challenges present.

Purposive and theoretical sampling methods were employed to ensure that the selected participants could provide rich and meaningful data essential for understanding the phenomenon explored in this study. ²⁰ Midwives, MOs and Emergency Medical Services (EMS) paramedics were purposively sampled for their key roles in managing and referring obstetric emergencies. In the sub-district that this study is focusing on, there are 128 midwives (CHCs and hospitals), 13 MOs and approximately 202 paramedics, with only 3 holding Emergency Care Technician or Emergency Care Practitioner (ECP) certifications. The theoretical sampling in this study did not rely on a predefined sample size; instead, it involved tracking the participants over time, guided by the theoretical concepts that emerged during data collection. To ensure the selected participants contributed relevant information, theoretical sampling was used to identify and address data gaps.

Data were collected from August 2024 to May 2025 using focus group discussions (FGDs), individual in-depth interviews (IDIs), document analysis and observations. This multi-method approach strengthened methodological rigour and enabled systematic, constant comparative analysis to identify key antecedents and attributes of the phenomenon. ²⁰ FGDs were conducted with midwives at 5 CHCs and hospitals. A total of thirteen FGDs were conducted comprising five at CHCs and eight at hospitals. Each FGD, facilitated in English and isiXhosa, lasted 35 minutes to an hour and involved three to five midwives, as recommended. ²¹ The FGDs were led by the primary researcher and co-facilitated by her supervisor. Additionally, the primary researcher conducted eight IDIs with MOs and three with paramedics, each lasting 35 to 45 minutes, conducted in English and isiXhosa. The 24 data collection sessions adhered to the suggested approach, ²² which emphasise that saturation should prioritise conceptual completeness and relational depth across categories, rather than rigid sample sizes. Data collection continued until theoretical saturation was achieved, indicated by no emergence of new categories or properties and confirmed relationships among existing concepts, as guided by the iterative principles of theoretical sampling. ^{22, 23} The FGDs and IDIs were audio recorded in distraction-free settings, at times convenient to participants. To further enrich the data, a triangulation approach was used, which included non-participant observations and document analysis, guided by an observation and a document analysis guide. These methods provided valuable insights into the upward referral process for obstetric emergencies, revealing both commonalities and differences in the perspectives of midwives and MOs.

Ethical principles were maintained throughout the study, beginning with obtaining ethical clearance from the University of Kwa-Zulu Natal (UKZN)’s Biomedical Research Ethics Committee BREC (reference number: BREC/00006633/2024), approved on the 22 ^nd of May 2024. Further approval was also secured from relevant authorities, including the Eastern Cape Department of Health, as well as key departmental heads, such as the District Manager and Sub-District Manager. Additionally, approval was obtained from the Chief Executive Officers and Operational Managers of the participating data collection sites. The researcher sought informed consent from participants, including permission for audio recording. Participants were reminded of their voluntary participation and right to withdraw without consequences. Strict confidentiality measures were implemented, ensuring all data remained anonymous and untraceable to individual participants.

Given the sensitive nature of obstetric emergencies, the researcher prioritised mitigating potential emotional or psychological distress, particularly when participants discussed adverse or life-threatening events, ²⁴ such as experiences of loss or trauma in the referral process. The researcher proactively provided contact details for the South African Depression and Anxiety Group (SADAG) to all participants. No instances of distress requiring referral were reported. All data is managed with stringent confidentiality measures where it is stored on a password-protected computer, accessible only to the researcher and authorised supervisors. In compliance with data protection protocols, remaining electronic files will be permanently deleted five years after study completion.

Data analysis followed a Straussian Grounded Theory approach ^{20, 25} using open, axial, and selective coding with constant comparative analysis. Interviews were transcribed verbatim by the researcher and research assistant and verified by the supervisor, co-supervisor, and co-authors to ensure accuracy. Analysis began with open coding, examining transcripts line by line to identify initial codes and patterns, with reflections captured through memoing. Axial coding explored relationships between categories, such as linking “clear communication” with “prompt decision-making,” while selective coding refined categories into a conceptual framework focused on antecedents and attributes, without presenting a full emergent theory. An independent coder used Atlas.ti to conduct inductive, line-by-line coding and organised the resulting codes deductively in alignment with the study’s objectives. Rigour was ensured through reflexive thematic analysis by maintaining a transparent coding trail, grounding interpretations in the data, and applying credibility, dependability, and confirmability criteria. Throughout, constant comparative analysis allowed iterative comparison of codes across interviews, observations, and documents. The researcher applied theoretical sensitivity, approaching data without preconceived notions, consulting GT experts, and co-constructing meaning with participants, consistent with the social constructivist paradigm. This systematic process ensured that findings were firmly grounded in participants’ experiences.

Trustworthiness was ensured following established criteria for credibility, dependability, confirmability and transferability ²⁶ within the Straussian Grounded Theory framework. ²⁰ Credibility was established through prolonged engagement with participants, iterative constant comparative analysis, and member checking, allowing emerging codes and categories to be validated and grounded in participants’ experiences. Dependability was supported by maintaining a transparent audit trail documenting all steps of data collection, coding, memoing, and category development, reflecting the systematic and iterative nature of GT. Confirmability was achieved through reflexive memoing, independent coding by the research team, consensus discussions on emerging categories, and investigator triangulation, ensuring that findings were derived from the data rather than researcher preconceptions. Transferability was addressed through rich contextual descriptions, detailed accounts of research objectives, methods, and the researcher’s role, enabling readers to understand how the identified antecedents and attributes may apply in similar settings. Multiple data sources and perspectives further strengthened triangulation, reinforcing the rigor and reflexivity central to GT methodology.

The study involved 59 participants each contributing valuable insights on aspects of the upward referral process for women with obstetric emergencies. The participants, consisted of midwives, MOs and paramedics. Participants ranged from different age groups with varying levels of professional experience, providing a diverse range of insights. All participant codes (e.g., FGD1_MW1) in Table 1 were assigned solely for organisational and analytic purposes and do not contain or derive from any personal identifiers. All data were de-identified in accordance with the Health Insurance Portability and Accountability Act (HIPAA) Safe Harbor method.

Antecedents are the conditions or factors that must exist before the upward referral process can occur effectively. ²⁷ According to the healthcare practitioners, these elements set the stage for upward referral and are often system-level or preparatory in nature. Table 2 presents the key antecedents identified as critical for ensuring effective upward referral of obstetric emergencies from CHCs to higher-level healthcare facilities.

3.2.1 Clear and timely communication

Data sources indicated that clear and timely communication is crucial for obstetric emergencies upward referral process to guarantee readiness of the receiving facility. Two primary categories were identified: (a) Communication of relevant information clearly and quickly and (b) Provision of accurate and concise patient details to the receiving facility.

3.2.1.1 Communication of relevant information clearly and quickly

Data sources mentioned that a seamless upward referral process of obstetrical emergencies is ensured by quick and clear communication, which prevents misunderstandings and unnecessary delays, as highlighted in the quotes below:

Although participants consistently emphasised the importance of clear, quick, and relevant communication, they also highlighted various barriers that hinder the timeliness of referrals. These include communication breakdowns, network issues, and unpaid telephone bills. These challenges represent contextual and structural constraints that disrupt the implementation of otherwise effective communication practices.

3.2.1.2 Providing accurate and concise patient details to the receiving facility

The provision of accurate and concise patient details emerged as equally important in ensuring effective upward referral of obstetric emergencies. The participants alluded that the referring facility needs to provide the receiving facility with accurate and concise details of the patient. Participants expressed that timely communication facilitates adequate preparation at the receiving facility, helping to prevent delays in emergency care. The extracts support this view.

MOs and midwives at receiving facilities further stressed the importance of knowing how the patient was managed at the referring hospital before transfer which is crucial for ensuring continuity of care.

However, the consistency and completeness of such details are often undermined by factors such as the absence of standardised communication protocols, inconsistent documentation systems and inadequate digital infrastructure, which compromise the reliability of information flow, even when staff are committed. This issue was confirmed by the primary researcher, who noted that patients are frequently referred with essential pre-referral information missing, incomplete, or inconsistently recorded. Table 3 below represents an extract from the document analysis illustrating this:

3.2.2 Adequate preparation and documentation

Participants indicated that proper preparation and documentation ensures that the receiving facility has all the necessary information to provide the best possible care upon the patient’s arrival. Under this facet, two key domains were noted: a) Thorough initial patient preparation and b) Ready availability of patient documents.

3.2.2.1 Thorough patient preparation

Participants stated that thorough patient preparation involved ensuring that the patient receives initial treatment, such as putting up an intravenous (IV) line, ensuring that stat doses of emergency drugs are given, etc. The extracts below reflect participants’ views:

Participants at the receiving hospitals indicated that patients are sometimes not adequately prepared. Gaps in patient preparation and documentation are primarily rooted in structural challenges, particularly workforce constraints that limit the capacity of CHC midwives, who are often overburdened and required to juggle multiple roles. As a result, even when the intent to stabilise patients is present, preparation may fall short due to these broader structural barriers. The following extracts attest:

3.2.2.2 Ready availability of patient documents

Data sources emphasised the importance of including the patient’s medical history and current condition to ensure that critical information is available upon arrival, thereby preventing unnecessary delays. Their views are reflected below:

The participants further noted that missing documentation and medical history lead to unnecessary delays and can worsen the patient’s condition in the healthcare facility, as reflected below:

3.2.3 Resource and human resource management

It emerged from the data that the seamless operation of an efficient upward referral process for women with obstetric emergencies and the provision of appropriate care is contingent upon the effective management of resources and human resources, within the context of the healthcare system’s infrastructure and capacity. Two areas emerged under this category: (a) Efficient distribution of resources and (b) Sufficient number of trained midwives available.

3.2.3.1 Efficient distribution of resources

Successful referrals depend on the effective use of human resources, transportation and medical equipment. Yet, systemic barriers, such as limited ambulance availability and inadequately equipped ambulances, under-resourced CHCs and unmet basic needs often disrupt the process. As explained below:

The primary researcher also observed that none of the CHCs had an ambulance on site, despite most being in remote locations. Additionally, participants emphasised the importance of proper equipment in ambulances to ensure that they are reliable for transfers:

Referral records from CHCs further illustrated the practical constraints facing the upward referral system. The table below highlights systemic inefficiencies, including inconsistent documentation, lack of functional communication tools (such as absence of work telephones) and delays in ambulance response times. Critically, many of these referrals involved obstetric emergencies such as severe preeclampsia, eclampsia, incomplete miscarriages and per vaginal bleeding, all of which require urgent and time-sensitive interventions. Table 4 below represents an extract from the document analysis illustrating this:

Participants reported that ambulances were initially equipped with essential tools, but systemic operational failures, including poor maintenance, led to damage or disuse, as mentioned below: “ Sometimes there are not even the necessary tools, such as advanced monitors and it jeopardises the patient condition. Even when we did have the big monitors, they end up broken” (IDI9_PARAM1).

3.2.3.2 Sufficient number of trained midwives available

In addition to resource allocation, having an adequate number of trained midwives is essential for effectively managing referrals of obstetric emergencies. Inadequate staffing and trained midwives may cause delays, which jeopardise the mother’s safety and makes the referral process harder. Participants highlighted the importance of adequate staffing:

The ability to deliver prompt and efficient care during referrals is directly correlated with staffing levels, as these statements demonstrate.

The primary researcher observed that staffing challenges in some facilities required a single midwife to cover multiple departments simultaneously, including the maternity ward, antenatal care, postnatal care, and family planning. As a result, midwives had to divide their attention and resources, shifting focus from managing emergencies to performing routine checks and providing counselling. This common operational reality highlights how institutional limitations, rather than deficits in clinical knowledge, constrain the midwife’s ability to ensure timely referral and safe patient preparation.

Attributes, as defined ²⁷ are the essential characteristics that constitute a concept. In the context of this study, attributes refer to the defining features of upward referral in obstetric emergencies. Table 5 below provides a summary as conceptualised from the perspectives of the data sources, illustrating how the phenomenon of ‘upward referral of obstetric emergencies’ was understood in this study.

3.3.1 Timely and effective decision-making

Effective prompt and timely decision-making in the process of upward referral of obstetric emergencies emerged as the cornerstone characteristic of ensuring safe and efficient transfers of patients needing advanced care. This attribute directly impacts the success of the referral process by guaranteeing that patients receive appropriate care without preventable delays.

3.3.1.1 Timely and decisive action

Timely and decisive action emerged as key for effective upward referral in obstetric emergencies. According to the data, when transferring patients to a higher level of care, prompt and efficient decision-making is crucial to ensuring they receive appropriate care as quickly as possible, as indicated below:

While participants emphasised the importance of timely action, delays in ambulance response and other systemic obstacles frequently disrupted this ideal. They described having to make instant decisions about how to manage the patient’s condition while rapidly assessing risks. Field observations likewise revealed how healthcare workers assessed patients under pressure and initiated referrals with urgency. This is illustrated in the participants examples below:

3.3.2 Effective collaboration and teamwork

Effective collaboration and teamwork are essential characteristics in the upward referral process in ensuring that all parties involved in the woman’s care work together seamlessly to obtain the best results. Two domains emerged: (a) Unified trust in teamwork and (b) Collaboration with the referral hospital’s team.

3.3.2.1 Trust in teamwork

Unified trust in teamwork emerged as a key element for effective upward referral of obstetric emergencies. This means that the team members trust each other’s abilities and are willing to support one another during emergencies, reflecting a shared confidence. The following participants excerpts support this claim:

The data sources further highlighted the significance of good working relationships and mutual trust among team members to ensure a smooth upward referral process for women with obstetric emergencies in the following excerpts:

3.3.2.2 Collaboration between referral and receiving hospital teams

Effective collaboration between the referral and receiving hospitals is yet another key characteristic in ensuring the effective upward referral of obstetric emergencies. The participants reflected on the beneficial collaborative relationships with specific hospitals, particularly in urgent cases:

Furthermore, the value of a collaborative relationship with MOs in critical cases was noted:

Although the data sources agreed on trust and collaboration as the cornerstones for effective upward referral, some alluded to the systemic factors which often hinder these dynamics. Fragmented healthcare systems and hierarchical structures were described as key intervening conditions emerged, which created silos and undermined collaboration. This is what the participants said:

3.3.3 Effective referral outcomes

Achieving favourable outcomes when referring obstetric emergencies to specialised care emerged as yet another characteristic for effective upward referral of obstetric emergencies. Under this attribute, two key categories emerged (a) Safe and stable transfer of the patient and (b) Ensuring timely arrival at the referral hospital. However, the extent to which these outcomes are realised is often influenced by systemic and contextual constraints that shape how effectively safety protocols and timeliness are enacted in real-world settings.

3.3.3.1 Safe and stable transfer of the patient

Successful transfer of obstetric emergencies is characterised by safe and stable transfer of patients which ensures that the patient is transferred in a condition that minimises risk and prevents further complications, as seen below:

3.3.3.2 Ensuring timely arrival at the referral hospital

Timely arrival at the referral hospital emerged as another feature of successful upward referral of obstetric emergencies. This outcome, however, is heavily shaped by external operational factors, particularly the availability and responsiveness of ambulance services. Below are verbatim extracts from participants:

3.3.4 Post feedback and continuous improvement

Post Feedback and continuous improvement emerged as another vital characteristic of effective upward referral of obstetric emergencies. The goal is to continually improve the process by discussing and receiving feedback on what went well and suggestions for improvement. Two sub-attributes were identified: (a) Post Feedback and review of referral cases for continuous improvement, and (b) Opportunities for regular engagement and learning.

3.3.4.1 Post feedback and review of referral cases for continuous improvement

The data sources highlighted that effective upward referral of obstetric emergencies is marked by regular feedback and continuous review of referral cases to drive ongoing improvement. This is reflected in the participant statements:

Some participants suggested that feedback can be provided during perinatal meetings, where cases are reviewed, and constructive input is shared.

However, structural limitations were reported to compromise the usefulness of these processes. Participants noted that perinatal meetings were often attended by OMs rather than those directly involved in patient care, creating a disconnect between the review process and frontline realities.

Although feedback mechanisms are widely recognised as critical for learning and quality improvement, hierarchical communication structures and the exclusion of frontline staff frequently limit their relevance and impact.

Participants identified the provision of relevant, accurate, and concise patient information and the prompt relay of vital details as a cornerstone of effective upward referral. Communication practices were viewed as essential to ensuring the preparedness of the receiving facility, avoiding misunderstandings, and preventing delays in emergency care. These findings align with broader evidence highlighting communication as a cornerstone of effective emergency referral systems. ²⁸ However, intervening conditions such as poor network connectivity and unpaid phone bills often disrupted these processes, particularly in rural areas with fragile telecommunication infrastructure. In contrast to the findings of the current study, where delays persisted despite referral efforts, studies in other contexts have shown improved outcomes. For example, in Uganda, telephonic communication between sending and receiving facilities during obstetric emergencies significantly reduced admission and treatment delays. ²⁹ Similarly, in Ghana, direct phone communication among providers played a critical role in enhancing the referral process for obstetric emergencies. ³⁰

The findings also show that effective referrals depend heavily on adequate patient preparation and complete documentation. Participants from all cadres repeatedly emphasised this antecedent. Maintaining continuity of care was thought to depend on stabilising patients before to transfer (e.g., giving IV fluids). This is consistent with previous research, ³¹ who show that maternal morbidity and mortality are directly impacted by insufficient pre-referral care in obstetric emergencies. However, in this study, structural limitations such as ongoing midwife shortages and excessive multitasking, often compromised the ability to adequately prepare patients. Similar findings were reported, showing that structural issues impair patient record quality and hinder continuity between referral levels. ³²

Additionally, thorough patient preparation and documentation were widely recognised by participants as essential steps in the referral process. However, these tasks were often hindered by systemic challenges such as chronic staff shortages, multitasking demands, and the absence of standardised tools. In several facilities, a single midwife was expected to simultaneously manage maternity, postnatal, and emergency units. These institutional constraints, rather than gaps in clinical knowledge or professional commitment, undermined the ability to fully stabilise patients or communicate clinical histories during handover.

Inadequate pre-referral stabilisation has been shown to increase maternal and neonatal morbidity and mortality, especially in resource-limited settings. ³² Participants’ reports align with existing evidence that excessive workloads and understaffing restrict midwives’ capacity to initiate critical pre-referral interventions. These gaps in preparation can delay treatment at the receiving facility, as clinicians must repeat or begin stabilisation efforts that should have been initiated at the primary care level. ³³

The availability of resources, particularly emergency transport and ambulances emerged as a critical antecedent influencing the promptness and safety of upward referrals. The main barriers to safe transfers, according to the participants, are unresponsive emergency services, inadequately prepared ambulances and ongoing ambulance shortages. These results support the systematic review by Banchani and Tenkorang, ³⁴ which identifies transportation obstacles as a major cause of maternal mortality in low-income nations. Rural ambulance delays are also considered a public health emergency, as noted by Ogunleye et al. ³⁵ Notably, this study also showed that, even in cases where ambulances were available, patient safety during transit was jeopardised by subpar equipment (such as outdated monitors) and poor vehicle maintenance, pointing to a more serious systemic problem than just availability.

Timely and decisive decision-making was consistently identified by participants as a defining attribute of effective upward referral. They described the urgency with which referral decisions must be made in obstetric emergencies to avoid life-threatening delays. In their view, hesitation or delayed action could compromise maternal and foetal outcomes. This aligns with findings from Curtin et al. ¹³ and Tiruneh et al., ³⁶ who report that delays in clinical decision-making are closely linked to adverse maternal outcomes. Ameh ³⁷ further emphasises the value of experience and training in enabling prompt action during emergencies. The study data suggest that the effectiveness of upward referral depends not only on systems and resources but also on the clinician’s ability to assess, decide, and act swiftly. As such, timely decision-making is both a defining attribute and a practical necessity in the upward referral of obstetric emergencies.

Effective referrals were also made possible by the team members’ mutual trust. Participants indicated that under high-stress situations, trust served as a mechanism to lessen hesitation and needless repeating of activities. These observations support previous observations ³⁸ that found that enhanced intra-team trust is associated with better communication and efficiency during medical emergencies. Similarly, Harris et al. ³⁹ emphasise that mutual trust and cohesive collaboration have a direct impact on maternal safety in obstetric facilities. For practitioners operating in an environment of urgency, uncertainty and shortage, trust was more than just interpersonal in the setting of this study; it was a coping strategy.

In addition to internal trust, participants highlighted the need for stronger collaboration between referring CHCs and receiving hospitals. Effective upward referral was often undermined by a lack of communication and continuity across facilities. Participants reported that collaborative relationships, such as joint case reviews and open communication channels, improved mutual understanding and streamlined patient transitions. This finding aligns with authors, ⁴⁰ who argue that interprofessional collaboration in obstetric care leads to safer and more consistent referrals, especially in LMICs. Yet participants noted that where such collaboration was weak or inconsistent, delays and mismanagement were more likely to occur, negatively impacting patient outcomes. When collaboration and teamwork were present, referral processes tended to run more smoothly. However, organisational barriers, particularly rigid hierarchies and poor inter-facility coordination, limited the extent to which healthcare workers could function as a unified team across the continuum of care.

Participants consistently linked patient safety to the stability of the patient during transit and the timeliness of arrival at the receiving facility. However, they highlighted several challenges compromising safety, particularly in the rural Eastern Cape, where long distances, poorly resourced facilities, and inadequate transport infrastructure increase the risk of adverse outcomes during transfer. These reflections align with findings, ¹¹ which report that geographical isolation and poor road conditions significantly delay timely referrals. Similarly, operational difficulties and remote locations hinder patient safety. ³⁶ As such, safeguarding patients during inter-facility transfers is not just a logistical necessity but a core attribute of effective referral systems in obstetric emergencies.

In agreement with previous studies, ⁴¹ participants emphasised the importance of receiving feedback post-referrals to optimise referral systems. However, participants in this study expressed concern that such feedback often excludes frontline healthcare workers. Instead, meetings were frequently attended by senior managers, limiting opportunities for those directly involved in patient care to engage in learning and improvement processes. This contrasts with Wibbelink et al., ⁴² who advocate for inclusive feedback systems involving all levels of healthcare professionals to enhance continuity, accountability, and quality of care. Similarly, Avoka et al., ⁴¹ highlight that feedback and quality improvement initiatives can significantly improve outcomes in obstetric emergencies and regular reviews of referral cases help reduce the risk of adverse events. ⁴³

Finally, feedback and learning were found to be inconsistent, despite participants’ clear desire for improvement. The exclusion of frontline staff from perinatal meetings not only limited the scope of review but diluted the practical insights needed to improve future referrals. As a result, opportunities for shared learning were lost, and similar challenges recurred.

The study demonstrates that effective upward referral is conceptualised by maternity staff as a process defined by preparation, communication and coordination across levels of care. Participants viewed effective referral as contingent on specific antecedents (such as stabilisation, documentation and communication) and core attributes (such as teamwork, readiness and continuity of care). This conceptualisation positions effective referral not merely as a transfer event, but as a relational and systemic process grounded in collaborative action.