Competency gaps in emergency obstetric and neonatal care: Assessing knowledge,skills,and confidence in the use of magnesium sulfate among providers in Malawi

2.1. Study design

This study employed a purely qualitative design, incorporating both descriptive and exploratory approaches, and was guided by HCD principles. The aim was to investigate competency gaps among EmONC providers regarding the administration of parenteral anticonvulsants, specifically MgSO₄, in Malawi. A qualitative approach was deemed suitable as it facilitates a deep exploration of complex clinical practices, provider experiences, and contextual barriers affecting the application of MgSO₄ in real-world healthcare settings. Descriptive qualitative methods are impactful in health services research to understand the experiences and perspectives of healthcare providers within their natural work environments.^16,17 The exploratory aspect allowed for a flexible examination of emerging issues regarding provider competence, confidence, and systemic constraints in low-resource maternity settings.

To strengthen conceptual clarity, the study was informed by Miller’s Pyramid of Clinical Competence, which categorises clinical performance into four hierarchical levels: “knows”, “knows how”, “shows how”, and “does”. Using this framework, the study explored three interrelated domains: (1) knowledge (what providers know about MgSO₄ administration), (2) performance (how they apply this knowledge in real or simulated clinical situations), and (3) confidence (how assured and capable they feel when managing hypertensive emergencies). This framework provided a structured approach to examining provider-level factors that contribute to suboptimal MgSO₄ administration and maternal health outcomes.

Integrating HCD into the study complemented the qualitative inquiry by engaging healthcare providers to identify barriers and develop solutions to improve MgSO₄ administration. HCD is a participatory approach that values empathy, creativity, and collaboration with end-users when designing practical, contextually appropriate interventions. In this context, HCD empowered EmONC providers to do more than just identify challenges; it enabled them to develop solutions that could enhance clinical practice and health system responsiveness. Through participatory techniques such as journey mapping, systems mapping, and collaborative problem-solving exercises, the approach facilitated a deeper understanding of hidden performance barriers, including workflow inefficiencies, cognitive overload, emotional stress during obstetric emergencies, and infrastructural limitations often overlooked in traditional programme design. Past research has demonstrated that HCD practices can lead to user-driven innovations that are more acceptable, feasible, and sustainable within health systems, particularly in maternal and newborn health programmes in low-resource environments.^18,19 This integrated methodology aligned well with the study’s goal of not only identifying competency gaps in MgSO₄ administration but also creating context-specific interventions to improve EmONC service delivery.

The findings of this study are reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ). This guideline includes a checklist with 32-items. It is designed to improve the transparency and rigour of qualitative research. ²⁰ Adhering to these guidelines ensured a clear and comprehensive presentation of our methodology and results. It also enhanced the overall quality and trustworthiness of our qualitative analysis. The completed COREQ checklist is provided in the appendices.

2.2. Study setting

The study was conducted across three purposefully selected districts in Malawi, chosen to represent geographic diversity, variations in maternal health indicators, and disparities in health system infrastructures. Each district was represented by one district hospital providing Comprehensive Emergency Obstetric and Neonatal Care (CEmONC) and two Basic Emergency Obstetric and Neonatal Care (BEmONC) facilities within its catchment area. Facilities were selected based on predetermined criteria designed to capture a broad spectrum of service delivery contexts and provider experiences. The selection considered facility performance in administering magnesium sulfate, including both higher and lower-performing sites, to explore differences in clinical practice and identify contextual barriers affecting provider competency. The study included both rural and urban facilities to reflect geographic and infrastructural variability within the Malawian healthcare system. High-volume maternity facilities were prioritised depending on their frequent encounters with obstetric emergencies. This qualified their likelihood of administering parenteral anticonvulsants. Additionally, one facility was selected based on evidence of poor BEmONC performance or a high number of deliveries, facilitating the exploration of potential systemic challenges that could influence provider competence. This purposive selection strategy enabled the study to gather diverse provider experiences and health system challenges relevant to understanding competency gaps in MgSO₄ administration.

2.3. Research team and reflexivity

The study was conducted by a multidisciplinary team of experienced female researchers with expertise in midwifery, nursing, maternal and neonatal health, public health, and health systems research. The team was led by MK, a nurse-midwife and Associate Professor with a PhD in Nursing (Women’s Health) and specialization in reproductive health. The core team also included EC, a nurse-midwife and Associate Professor with a PhD in Nursing (Midwifery), AK, an Associate Professor and PhD-trained public health specialist with expertise in health systems strengthening and implementation research, and LK, an Associate Professor with a PhD in Nursing specializing in maternal and neonatal health.

Other members included CK, a lecturer with a Master’s degree in Public Health; GL, a lecturer with an MSc in Midwifery; and FK, a lecturer with a BSc in Nursing and Midwifery. All researchers had training and experience in qualitative research methods, including Good Clinical Practice (GCP) training and protocol-guided qualitative data collection techniques. The team conducted the study within their respective academic, clinical, and health systems roles in maternal and newborn health.

No prior relationships existed between the researchers and participants before data collection. Participants were informed about the study purpose, procedures, and the role of the research team through an information sheet provided prior to interviews. Reflexivity was maintained through adherence to the interview guide, regular team debriefing meetings, and critical reflection on how researchers’ professional backgrounds and clinical experience may have influenced data collection and interpretation of findings.

2.4. Participants and sampling

As highlighted in Figure 1, participants were purposively sampled through face-to-face recruitment and email invitations to ensure inclusion of healthcare professionals directly involved in Emergency Obstetric and Neonatal Care service delivery or oversight of maternity services. The sample comprised registered nurse-midwives, nurse-midwife technicians, clinical officers, medical officers, maternity in-charges, and members of the District Health Management Team (DHMT). These groups were selected due to their direct involvement in managing obstetric emergencies, supervising maternity services, or monitoring EmONC performance indicators, including the administration of magnesium sulfate.

OPEN IN VIEWER

Figure 1.

Study flow of participant recruitment and integrated HCD process).

Eligible participants included licensed healthcare professionals actively working in maternity or EmONC-designated facilities with direct experience in managing obstetric emergencies or supervising maternity care. To ensure adequate familiarity with routine clinical and institutional processes, participants were required to have been employed at their respective facilities for at least six months. Newly deployed staff with less than six months of service, interns, and personnel not directly involved in maternity care were excluded. Participation was voluntary, and all eligible participants provided written informed consent prior to enrolment.

A total of 48 participants were purposively recruited and engaged in an integrated Human-Centered Design (HCD) process conducted through structured one-day workshop sessions. Focus group discussions (FGDs) were embedded within these workshops as one of the methods for generating and refining insights, rather than constituting a separate data collection exercise or additional sample. The FGDs were conducted in professionally homogeneous groups of five to seven participants, which facilitated open and reflective dialogue among colleagues in similar clinical roles while contributing to iterative co-design discussions within the broader HCD process.

Participants were drawn from both Basic and Comprehensive EmONC facilities and represented a range of professional roles and levels of clinical experience, including 18 registered midwives, six nurse-midwife technicians, six facility maternity in-charges, six clinical officers, and 12 DHMT members. This diversity enabled the study to capture perspectives from frontline maternity care providers as well as health system managers overseeing service delivery.

2.5. Data collection

Data collection was conducted between February 2023 and June 2023 outside the clinical setting through a combination of FGDs and participatory one day HCD workshops. These complementary methods were chosen to generate rich experiential data about provider knowledge, skills, and confidence in administering MgSO₄, while simultaneously enabling participants to collaboratively identify barriers and propose practical solutions to enhance clinical practice. Eight HCD workshops were organised with a total of 48 participants drawn from both BEmONC and CEmONC facilities, as well as members of district-level health management teams. A variety of participatory design techniques were employed to facilitate reflection on clinical workflows and systemic challenges influencing MgSO₄ administration. Journey mapping exercises were utilised to visualise care processes involved in managing hypertensive emergencies and to pinpoint critical moments where competency gaps might arise. Systems mapping exercises provided a platform for participants to describe the institutional and infrastructural constraints affecting clinical practice. Ranking exercises were conducted to prioritise key barriers to MgSO₄ administration, while low-fidelity prototyping allowed participants to collaboratively generate and test ideas aimed at improving clinical workflows and decision-support tools.

In addition, eight focus group discussions were held to dig deeper into provider experiences. Three FGDs were conducted with staff from BEmONC facilities, three with staff from CEmONC facilities, and two with District Health Management Team members. Semi-structured discussion guides, informed by HCD principles, were used to stimulate reflections on clinical “pain points,” document common workarounds employed to navigate systemic challenges, and explore potential solutions for improving MgSO₄ administration. Discussions were held in English or Chichewa based on participant preferences, ensuring inclusivity and cultural appropriateness. All sessions were audio-recorded with consent and subsequently transcribed verbatim.

Data collection continued until thematic saturation was achieved, defined as the point at which no new themes or significant insights emerged from ongoing discussions. The research team conducted preliminary analyses concurrently with data collection, meeting after each round of FGDs and workshops to review emerging codes and themes. After the sixth focus group discussion, the absence of new themes prompted an additional two discussions to confirm that saturation had indeed been reached. This iterative approach ensured that the dataset comprehensively captured the range of provider experiences and perspectives pertinent to MgSO₄ administration in EmONC settings.

2.6. Data analysis

The data were analysed thematically using Braun and Clarke’s six-phase framework, which involved data familiarisation, generation of initial codes, searching for themes, reviewing themes, defining and naming themes, and producing the final report. This analysis occurred concurrently with data collection in an iterative process, allowing emerging insights to inform discussions and workshops, thereby refining probing questions and exploring concepts more deeply.

All audio recordings from focus group discussions were transcribed verbatim, with Chichewa transcripts translated into English prior to analysis. The transcripts were imported into NVivo version 14 for systematic organization, coding, and retrieval of qualitative data. A team-based analytical approach was used to minimize bias and ensure comprehensive interpretation. Two researchers (CZ and EP) independently coded the data to enhance analytical rigor. The researchers conducted initial open coding, identifying meaningful text segments related to providers’ knowledge, performance, confidence, and systemic factors affecting MgSO₄ administration. The researchers compared their coded outputs and developed a shared codebook, which was refined as more transcripts were reviewed and new themes emerged.

Consistent with the HCD approach, affinity mapping was applied in the early analysis stages to organize qualitative insights into thematic clusters. Utilizing Miro, a real-time collaboration tool, the research team grouped related ideas, experiences, barriers, and potential solutions based on identified patterns from transcripts and field notes. This process allowed for the organization of extensive qualitative information into coherent themes while capturing significant quotations and examples of workarounds used by providers in clinical practice.

The generated themes informed the coding framework and guided further coding in NVivo. Themes were refined into broader categories that highlighted competency gaps and systemic barriers to MgSO₄ administration in EmONC settings.

Through analysis and reflection, actionable insights beyond descriptive themes were identified, revealing underlying motivations, challenges, and contextual factors shaping provider behaviors. These insights highlighted opportunities to enhance EmONC service delivery and guided the co-creation of practical solutions during the HCD process.

2.7. Trustworthiness of the study

The trustworthiness of this study was established by following key qualitative research criteria. To enhance credibility, all interviews and focus group discussions were audio recorded, and field notes were taken immediately after each session. Verbatim transcripts were created shortly after data collection to ensure that participants’ responses were accurately represented. Trained transcribers from KUHeS handled the transcription and translation, ensuring precision and fidelity to the original data. A qualitative research expert reviewed the transcripts alongside the audio recordings to confirm their completeness and clarity.

Dependability was achieved by maintaining a thorough audit trail that detailed coding decisions, analytical procedures, and any adjustments made to the codebook during analysis. Before data collection began, the research team received training from an HCD expert, covering essential HCD principles such as empathy research, insight generation, prototyping, and iterative analysis. This training helped ensure consistent adherence to the methodology throughout the study.

Multiple researchers independently coded the data to reinforce confirmability. We held ongoing discussions within the team. This helped us mitigate personal biases. Transferability was supported by providing comprehensive details about the study setting, participant demographics, and research methods. This was done to enable readers to assess how the findings might apply in similar situations. To enhance credibility, member checking was used. Summaries of the findings were shared with selected participants. This allowed them to confirm that the interpretations accurately captured their experiences and viewpoints.

2.8. Ethical considerations

Ethical approval for this study was granted by the College of Medicine Research and Ethics Committee (COMREC) in Malawi (Protocol No: P.01/23/3723). Additionally, administrative permissions were secured from the three District Health Offices involved and from the in-charges of the selected BEmONC facilities.

Before participation, all participants received comprehensive details about the study’s objectives, procedures, potential risks, and expected benefits. They were given ample time to review the information and ask any questions before deciding to participate. Written informed consent was collected from all participants before data collection commenced. Participation in the study was voluntary. Participants were informed that they could withdraw at any time. There were no penalties for deciding to withdraw.

To uphold confidentiality, identification codes were used instead of names on all data collection tools and transcripts. Audio recordings, transcripts, and field notes were securely stored in encrypted, password-protected files, accessible only to members of the research team. These measures were implemented to ensure participant privacy. Data confidentiality was rigorously maintained throughout the study.

3.1. Participant demographics

A total of 48 participants were involved in the study through eight HCD workshops, which included focus group discussions conducted across three districts. Table 1 summarises the distribution of participants by facility type and professional role.

OPEN IN VIEWER

Table 1.

Distribution of FGD participants by facility type and professional role (N = 48).

Facility designation	Participant group	Number (n)	Professional role description
BEmONC	RNMs	6	All were maternity in-charges
	NMTs	6	All worked in maternal and neonatal units
	Facility In-charges	6	3 Clinical Officers; 3 Medical Assistants
CEmONC	COs	6	All were maternity COs; 1 was an anaesthetic CO and head of theatre
	RNMs	12	Included maternity in-charges, Safe Motherhood Coordinators, and midwives
	DHMT	12	2 DEHOs, 2 DHSS, 2 DMOs, 2 DNMOs, 2 Nursing Officers, 1 HSA, 1 Safe Motherhood Coordinator

3.2. Overview of results

Following participant engagement across eight HCD workshops and focus group discussions, a detailed thematic analysis was conducted to explore competency gaps in the administration of parenteral MgSO₄ among EmONC providers. Drawing on insights from RNMs, COs, NMTs, facility in-charges, and DHMT members, the findings reveal complex and interrelated challenges ranging from individual knowledge deficits, practical performance inconsistencies, and systemic barriers within maternal emergency care settings.

The analysis yielded three interrelated thematic domains: knowledge, performance, and confidence.

3.3. Themes generated from the study

Three major themes were identified from the thematic analysis: knowledge, performance, and confidence. Each theme represents a distinct but interconnected dimension that influenced providers’ ability to administer MgSO₄ safely in emergency obstetric care, as illustrated in Table 2.

OPEN IN VIEWER

Table 2.

Summary of themes and subthemes on competency gaps in MgSO₄ administration among EmONC providers.

Major theme	Subthemes	Description of finding
Knowledge: Understanding of MgSO4 Protocols	• Incomplete recall of maintenance dose and monitoring requirements	Participants demonstrated partial theoretical understanding of MgSO4 protocols. While most providers recognised the indication and loading dose, many struggled to recall maintenance dosing intervals, monitoring requirements, and toxicity signs. Knowledge gaps were further influenced by a lack of refresher training, ambiguous guidelines, and the absence of simplified reference tools such as job aids or standardised SOPs.
	• Limited knowledge of toxicity signs and management
	• Confusion on dosage, dilution, and routes of administration
	• Ambiguities in protocol wording and lack of updated reference materials
Performance: Translation of Knowledge into Clinical Practice	• Inconsistent dilution and administration techniques	Although some providers were familiar with MgSO4 protocols, their application in clinical practice varied. Monitoring practices such as reflex checks, urine output assessment, and respiratory monitoring were often skipped due to workload pressures or equipment shortages. NMTs reported minimal practical exposure, while providers highlighted lack of mentorship and routine emergency drills to maintain competency.
	• Limited hands-on experience among some cadres
	• Omission of recommended toxicity monitoring practices
	• Lack of clinical mentorship and emergency simulations
Confidence: Psychological Readiness and Clinical Autonomy	• Hesitation to administer MgSO4 during emergencies	Providers frequently expressed hesitancy and anxiety when administering MgSO4, particularly in resource-limited settings or when senior support was unavailable. Even when the drug was available, some providers preferred referring patients due to lack of confidence. Limited availability of antidotes, equipment, and clear escalation protocols further contributed to reluctance in administering the medication.
	• Fear of adverse outcomes and lack of backup support
	• Reluctance to administer the drug when working alone
	• Reduced confidence due to lack of antidotes or monitoring equipment

Each domain is presented in detail, enriched with illustrative participant quotes and triangulated perspectives across cadres. Furthermore, the HCD approach enabled the co-creation of contextually relevant solutions, which are integrated into the findings.

3.3.3. Theme 3: Confidence-psychological readiness and clinical autonomy

Participants reported hesitancy in administering MgSO₄, particularly during emergencies or when working alone. One RNM stated, “If I can’t remember every step, I’d rather wait for another midwife.” A Clinical Officer described similar concerns, noting, “You’re thinking, if I give and she crashes, will there be oxygen? Will there be backup?”

NMTs also reported reluctance to administer MgSO₄ in the absence of senior staff. One NMT explained, “If I’m alone and there’s no doctor, I hesitate to give MgSO ₄ … what if something goes wrong?” Some providers referred to the absence of calcium gluconate as a factor influencing their decisions. An RNM stated, “If calcium gluconate is there and I know the steps, then I won’t delay. But if not, I’m scared to use it.”

DHMT members reported similar observations during supervision. One noted, “It’s not just about the BP machine not working. Even when it’s there, they still don’t administer. The issue is, they’re not competent, they’re not comfortable.” Facility in-charges also described challenges related to infrastructure and equipment. One reported, “We face challenges with BP machines; sometimes we don’t have functioning ones or even batteries,” while another added, “We also lack urine dipsticks for diagnosing preeclampsia, which makes us unsure how to proceed.”

Participants identified several facility-level actions during the workshops, including emergency preparedness training, visible SOPs, designation of MgSO₄ champions, and stocking of antidotes within maternity units. Additional suggestions included emergency kits containing checklists and SOPs, facility-level mentorship through clinical champions, and reflection sessions during post-emergency debriefings (Table 3).

OPEN IN VIEWER

Table 3.

Triangulated findings on MgSO₄ competency gaps with DHMT and HCD-informed solutions.

Domain	Key provider insights (RNM/NMT/DHMT)	HCD synthesis (user experience insight)	System-level barriers	HCD-informed solutions
Knowledge	• Incomplete recall of MgSO4 maintenance dose, toxicity signs, and monitoring procedures • Confusion on dosage, route, and dilution • Ambiguity in protocol interpretation across cadres	• Low confidence despite exposure to MgSO4 • Preference for visual aids and mobile guidance to support recall in emergencies	•Inconsistent or outdated SOPs • Ambiguous drug labelling and guidelines • Inconsistent training on revised formulations	• Laminated emergency job aids • Simplified, standardised SOPs • Mobile-based refresher modules • Training on updated MgSO4 formulations
Clinical Performance	• Missed monitoring steps during administration • Variation in dilution and administration sequence • Limited hands-on experience post-training • Preference for referral in severe cases	• Skill gaps persist despite training • Confidence improves with practice-based learning	• Lack of structured mentorship • Absence of routine simulation in clinical workflow • Weak supervision systems	• Weekly simulation drills • Peer mentorship systems • Low-fidelity simulation kits • Integration of mentorship into ward routines
Confidence & Decision-Making	• Hesitation in emergencies without senior support • Anxiety when working alone • Fear of adverse outcomes and complications	• Confidence strongly linked to visible protocols, antidotes, and support systems	• Stock-outs of antidotes (e.g., calcium gluconate) • Limited monitoring equipment (BP machines, urine dipsticks) • Unclear escalation pathways	• Ensure availability and visibility of antidotes • Designate MgSO4 clinical champions • Bedside coaching and supportive supervision • Facility-specific emergency escalation protocols

4.1. Knowledge-related gaps

The knowledge gaps highlighted in this study illustrate systemic challenges that go beyond individual provider performance. In Malawi, issues such as irregular refresher trainings, limited supervision, and unclear guidelines have consistently been identified as obstacles to effective use of MgSO₄.^12,21,22 Similar findings in LMICs demonstrate that fragmented knowledge often stems from inadequate dissemination of protocols, insufficient pre-service and in-service training, and weak clinical mentorship.^11,23–27 This indicates that the underutilisation of MgSO₄ is not an isolated issue in Malawi but part of a larger pattern of structural and educational deficiencies in emergency obstetric care.

These knowledge gaps have significant implications for maternal and fetal outcomes. Not recognising toxicity or failing to administer correct maintenance doses elevate the risk of maternal morbidity and mortality, while delays in treatment can jeopardise fetal survival.^11,23 Therefore, enhancing provider knowledge is both a technical necessity and a crucial intervention for protecting maternal and neonatal health. MgSO₄ must be viewed as a multidimensional intervention, a life-saving drug whose effectiveness hinges on provider confidence, clear guidelines, and robust systems of training and supervision.^24,28,29

Participants expressed a strong preference for practical, visually appealing job aids as a feasible solution to these challenges. Tools such as laminated dosage charts, wall-mounted SOPs, and structured refresher training were valued for supporting immediate recall during emergencies, thereby reducing reliance on memory under pressure. ³⁰ Research shows that simulation-based training and visual aids can enhance protocol adherence, improve decision-making, and reduce cognitive load in high-stress situations.^5,31–33 By incorporating context-sensitive, point-of-care supports, healthcare providers are better prepared to respond decisively and consistently in obstetric emergencies.

Overall, the findings suggest that addressing these knowledge gaps necessitates a dual strategy: reinforcing formal training while embedding decision-support tools into routine practice.^{5,25–27,32,33} Such approaches not only enhance provider readiness but also lead to measurable improvements in maternal and fetal health outcomes. In resource-limited settings, where supervision may be sparse and emergencies require swift action, equipping providers with clarity, confidence, and accessible tools is vital for maximizing the effectiveness of MgSO₄ in lowering maternal and neonatal mortality rates.

4.2. Clinical performance challenges

The inconsistencies in the clinical application of MgSO₄ observed in this study highlight the gap between theoretical awareness and practical performance. Deviations in dilution techniques, skipped toxicity monitoring, and variations in administration methods were frequently attributed to systemic pressures such as high patient loads, time constraints, and limited resources, factors previously documented in Malawi ¹² and echoed across other low-resource settings. Evidence shows that weak supervision, irregular training, and resource constraints undermine adherence to emergency obstetric protocols and contribute to variability in MgSO₄ use.^23,34,35 These regional and global findings reinforce that clinical performance challenges are not isolated but systemic, requiring both context-sensitive training and supportive health system structures to ensure consistent and safe application of MgSO₄.

Junior providers and NMTs appeared particularly vulnerable, with many reporting minimal hands-on experience and, in some cases, no administration of MgSO₄ post-deployment. This finding reflects broader evidence that insufficient simulation-based learning and weak supervision undermine skill retention and adherence to emergency obstetric protocols.^23,34,36 The absence of structured mentorship and standardised clinical drills further compounds variability in practice, echoing global patterns across LMICs where poor supervision and irregular capacity building contribute to inconsistent maternal emergency care. ³⁷

These performance gaps have direct implications for maternal–fetal outcomes. Inadequate monitoring or incorrect administration increases risk of maternal toxicity, delays life-saving treatment, and compromises fetal survival. Strengthening clinical performance, therefore, represents a multidimensional intervention: improving provider confidence and skill retention while simultaneously safeguarding maternal and neonatal health.^23,34,38,39

Through the HCD process, participants proposed context-sensitive strategies to overcome implementation barriers, including weekly simulation drills, bedside coaching, and low-cost competency checks embedded into routine workflows. Such participatory solutions resonate with evidence showing that simulation-based learning, particularly when combined with structured peer feedback and mentorship, enhances provider confidence, skill retention, and emergency preparedness.^38–40 Providers emphasised peer-led simulations and regular mentorship rounds as sustainable mechanisms for reinforcing knowledge, addressing skill decay, and cultivating a supportive learning environment. ³⁵

These insights align with emerging global applications of HCD in maternal health, which emphasize co-created, context-specific interventions as transformative pathways for improving emergency obstetric and newborn care.^19,41,42 By embedding simulation, mentorship, and competency checks into routine practice, health systems can bridge the gap between knowledge and consistent clinical performance, ultimately reducing maternal and neonatal mortality in resource-constrained settings.

4.3. Confidence and psychological readiness

Low provider confidence and lack of psychological readiness emerged as critical barriers to effective MgSO₄ use, particularly among NMTs. Reluctance to administer the drug was often rooted in fear of patient deterioration or complications, especially in facilities lacking essential resources such as calcium gluconate, functional blood pressure monitors, or urine dipsticks. This hesitation frequently translated into delayed administration or complete avoidance of MgSO₄, even in cases of severe pre-eclampsia or eclampsia. Facility in-charges and District Health Management Teams corroborated that the root issue was not drug availability but provider confidence and readiness to act.^12,24

Similar patterns have been documented across other low-resource settings. Emotional stress and fear of adverse outcomes have been shown to significantly influence clinical decision-making under pressure. ⁴² In Iran, Gholipour et al. ⁴³ highlighted how junior staff often struggled with emergency obstetric decisions due to anxiety and inadequate supervision. Evidence from Tanzania, Ethiopia, and Kenya evidence shows that insufficient mentorship, weak supervision, and lack of psychological safety undermine provider confidence and contribute to variability in emergency obstetric care.^{25,30,34–36} These findings reaffirm that psychological barriers often outweigh supply challenges, directly affecting maternal and fetal outcomes by delaying or preventing timely administration of life-saving interventions.

The HCD sessions in this study revealed practical, facility-level strategies to build confidence and psychological readiness. Suggested interventions included prominently displayed SOPs in maternity wards, emergency kits redesigned to include antidotes such as calcium gluconate, and the designation of “MgSO₄ champions” to mentor colleagues during critical cases. Structured debriefings following obstetric emergencies were also recommended to reinforce learning and cultivate psychological safety. These participatory solutions align with broader evidence: McClintock et al. ⁴⁴ emphasised that institutional support and psychologically safe environments are foundational to clinical confidence, while Salas et al. ³⁶ demonstrated that coaching and feedback reduce decision paralysis in high-stakes scenarios.

Globally, HCD approaches have been shown to strengthen provider readiness by embedding visible support systems and practical tools into routine workflows.^19,41,42 By integrating mentorship, peer-led simulations, and structured debriefings, health systems can address both technical and psychological barriers. This dual approach not only enhances provider confidence but also ensures timely and consistent use of MgSO₄, improving maternal and neonatal survival.

4.4. Implications for practice

The findings underscore the need for integrated, systems-based interventions that combine training, clinical tools, and infrastructure enhancements with participatory approaches. The HCD methodology proved effective in engaging frontline workers in identifying and prioritizing feasible strategies tailored to their workflows. Practical interventions such as job aids, simulation-based learning, peer mentorship, and visible emergency protocols offer promising opportunities to improve provider competence and readiness. Importantly, solutions that emerge from provider-led design processes are more likely to be adopted, sustained, and effective in real-world settings.

4.5. Limitations

This study’s strength lies in its triangulated approach, incorporating diverse perspectives from frontline providers, supervisory teams, and participatory design sessions. The use of HCD enriched the data by uncovering not just competency gaps but also provider-generated strategies for improvement. However, the study was limited to selected districts in Malawi and used a qualitative design, which may limit generalizability. Future research should include mixed-method or quantitative evaluations to assess the impact of suggested interventions on clinical outcomes and maternal health indicators.