The design employed was selected to enable a systematic assessment of the effectiveness of the patient referral system across various health facilities and communities within Mbala District at a single point in time.

The study was conducted in Mbala District, one of the more remote districts in Zambia, located in the Northern province. The district has a total population of approximately 166 766 and is largely rural, characterised by difficult terrain, poor road networks and limited access to higher-level health services. Mbala has 29 health facilities, including rural health centres, health posts and Mbala General Hospital which serves as the referral centre for the area.

The study population comprised two main groups: (1) health facility in-charges responsible for coordinating patient referrals and (2) community members from villages served by the sampled health facilities. The inclusion of community members aimed to gather first-hand perspectives on their experiences with the referral system.

Out of the 29 health facilities in the district, 18 were purposively sampled to ensure a representative mix of both hard-to-reach and more accessible facilities. At each of these facilities, the health facility in-charge was selected as a respondent. In addition, in every village visited during the study, 10 community members were interviewed, resulting in a total of 180 community respondents. Community members were selected using simple random sampling from households within the catchment area of each sampled facility. Data collection was completed within a 2-week period through a well-structured and coordinated fieldwork plan. A team of nine trained field assistants were deployed, organised into three teams of three members each, with each team responsible for covering six health facilities, ensuring efficient coverage of all sampled sites.

Data were collected using two tools: structured questionnaires administered to in-charges of 18 sampled health facilities, and community interview schedules conducted with 10 residents in each village visited. The questionnaires explored referral transport, communication systems, emergency response and staffing, while community interviews focused on accessibility, timeliness and satisfaction with referrals. Data collection spanned 2 weeks and was conducted by trained field assistants. Ethical standards, including informed consent and confidentiality, were strictly observed. Key variables included referral transport availability, communication systems, emergency response capacity, staffing levels, referral timeliness, accessibility of services and client satisfaction. These variables were used to assess the efficiency and effectiveness of the patient referral system. The data were coded and analysed using Jeffreys’s Amazing Statistics Program (JASP) version 0.19.3 statistical software to generate descriptive statistics and identify trends in the effectiveness of the referral system across Mbala District.

Ethical clearance to conduct this study was obtained from the Eden University School of Medicine Research Ethics Committee (No. REF. No. EdenUSMC-01-06-2025).

Regression analysis revealed that road condition, transport availability, communication reliability and facility type were important factors in the referral system performance across health facilities in Mbala District. The results showed varying effects of these factors on both referral delay and referral satisfaction. Facilities with better road access, more reliable transport and dependable communication systems generally experienced shorter referral delays and reported higher levels of satisfaction with the referral process. Differences were also observed across facility types, highlighting the influence of infrastructure and operational capacity on the efficiency and perceived quality of referral services.

In the referral delay model, the following coefficients were observed:

The negative and statistically significant coefficient for road conditions implies that as road conditions improve (from poor to good), referral delays decrease by approximately 1.64 h, all else held constant. Similarly, communication reliability shows a strong and significant effect: for every unit increase in communication reliability (e.g. from poor to moderate), referral delays reduce by nearly 1.89 h. Although transport availability and facility type showed negative coefficients suggesting that they help to reduce delays, their p-values (0.224 and 0.478, respectively) indicate that these associations were not statistically significant at the 5% level.

In the referral satisfaction model, the coefficients were:

Here, communication reliability again stood out with a statistically significant positive effect. A one-point improvement in communication reliability is associated with a 0.63-point increase in satisfaction, on a 5-point scale.

The results in Mbala District highlight several systemic weaknesses: the presence of only one operational ambulance for the entire district; poor road infrastructure that becomes impassable during the rainy season; and unreliable mobile communication networks that impede timely coordination during emergencies. Compounding these logistical barriers is the absence of resident medical doctors in rural health facilities, meaning that critical cases cannot be managed locally and must rely on the already-strained referral pathway. Unreliable sources of power in rural facilities further limit their capacity to stabilise patients before transfer.

The distribution of health facilities by distance from the general hospital offers critical insights into the structural limitations of the referral system in Mbala District. The spatial distribution poses significant challenges for timely patient referrals, especially in emergency situations. Evidence from rural health systems in sub-Saharan Africa consistently indicates that greater distances from referral centres are associated with increased mortality and delayed care, particularly in obstetric and paediatric emergencies.¹¹ Kruk et al. argue that physical access is among the most persistent barriers to achieving equitable health outcomes in low-resource settings.¹² In Zambia, the Ministry of Health recognises distance as a major determinant of care-seeking delays, as highlighted in the Zambia Demographic and Health Survey (ZDHS).¹³ Poor road networks, limited transport options and high fuel costs further compound the issue.¹⁴ Furthermore, long distances have been shown to reduce the frequency of completed referrals, as patients may abandon transfer because of travel difficulty or cost.¹⁵ In Mbala’s context, these constraints are intensified by the presence of only one district ambulance and poor communication network. To address this, targeted interventions such as investment in community-based transport solutions and improved road infrastructure are essential. Equally important is integrating telehealth and mobile-based coordination to bridge the gap for distant communities.¹⁶

The level of power outage in Mbala District represents a critical operational challenge, particularly for managing emergency referrals that require refrigeration of drugs, lighting during night-time emergencies and the use of diagnostic equipment. In rural health systems, reliable electricity is essential for effective service delivery. Power outages are directly linked to reduced healthcare quality and increased maternal and neonatal risks.¹⁷ The World Health Organization (WHO) also emphasises the role of electricity in supporting functioning cold chains, communication systems and life-saving interventions. Mbala’s grid-connected facilities, while technically electrified, remain functionally underserved because of prolonged power outages, with only a few facilities supplementing grid electricity through solar energy systems.⁸ Moreover, loadshedding disproportionately impacts referral systems, as communication tools (e.g. phones, radios) often depend on electricity. As observed by Scott et al., electricity access significantly enhances the operational readiness of rural health facilities.¹⁸ Without reliable power, even facilities near the referral hospital, such as Zombe (18 km), remain handicapped in managing emergency care. Therefore, investing in alternative sources such as solar systems with battery storage is vital. This aligns with recommendations by the Sustainable Energy for All initiative (SEforALL), which promotes decentralised energy solutions for rural health infrastructure.¹⁹

Various studies conducted in rural and resource-constrained settings, particularly in sub-Saharan Africa, show that referral systems are influenced by multiple factors, including infrastructure, transport, communication and human resource capacity. These factors highlight the need for integrated, evidence-based strategies to strengthen referral systems across remote districts, as discussed in the subsequent sections.²⁰ As shown in Table 4, a moderate negative correlation between distance to the hospital and referral success (r = –0.52) reflects barriers associated with geographic isolation. Numerous studies confirm that long distances to referral centres significantly delay or prevent access to advanced care. Kruk emphasises that geographical barriers are among the most persistent determinants of poor health outcomes in rural areas, highlighting the negative consequences of delayed access to emergency services because of distance.²¹ In Zambia, the Ministry of Health, in collaboration with development partners, has consistently documented through the ZDHS that women in remote areas are less likely to access timely obstetric and emergency care. A strong positive correlation between staff preparedness and referral success (r = 0.64) indicates that frontline health worker competence, including knowledge of referral protocols, triage ability and emergency stabilisation, plays a pivotal role in saving lives. Waiswa’s research team found that in rural Uganda, maternal referral outcomes were significantly influenced by the responsiveness and preparedness of health personnel.²² Studies have highlighted the essential role of emergency transport in effective health system functioning. Awoonor-Williams et al. demonstrated in Ghana that increased availability and reliability of ambulances improved patient transfer rates and outcomes in rural districts.²³ Likewise, Kipuro et al., indicated that districts equipped with functioning ambulances report higher completion rates of referral cases, although this is conditional on the availability of fuel, trained drivers and operational support.²⁴ Network coverage exhibited the strongest positive correlation with referral success (r = 0.89), emphasising the indispensable role of communication. Studies have consistently identified reliable communication as a critical enabler of timely referrals. Atukunda et al. reported that rural health workers in Uganda using mobile phones were able to reduce delays and coordinate emergency responses more efficiently.²⁵ Similarly, WHO and United Nations Children’s Fund (UNICEF) have underscored that robust communication systems are vital for emergency preparedness, especially in underserved and remote areas.²⁶

The referral satisfaction model provides key insights into the determinants of user satisfaction with the patient referral system in Mbala District. Among the predictors evaluated, communication reliability emerged as the only statistically significant factor (β = 0.63, p = 0.002), indicating a strong and positive relationship with satisfaction. A one-point increase in the communication reliability score corresponds to a 0.63-point rise in satisfaction on a 5-point scale. This finding is consistent with evidence suggesting that timely, reliable communication enhances user confidence in referral systems and reduces frustration associated with delays and uncertainty.²⁷ In contrast, transport availability (β = 0.27, p = 0.176), road condition (β = 0.19, p = 0.234) and facility type (β = 0.21, p = 0.342) were positively associated with satisfaction but did not achieve statistical significance. These results suggest that while physical infrastructure and institutional capacity are important, they are secondary to the perceived effectiveness of communication channels in shaping satisfaction. As highlighted by Peters et al., patient satisfaction in low-resource settings is often driven by system responsiveness and accessibility rather than facility classification.²⁰ This aligns with research by Ouma et al., who found that weak communication mechanisms are often more limiting than geographic distance in the success of rural health interventions.²¹ Similarly, Kruk et al. emphasise that functional systems where patients are guided, informed and reassured foster trust and satisfaction, regardless of structural limitations.¹² Moreover, WHO advocates for integrated digital health tools to facilitate real-time coordination and feedback between service points.²⁶ Therefore, enhancing communication infrastructure and protocols in rural Zambia could serve as a high-impact intervention for improving referral satisfaction, even in the absence of immediate improvements in roads or transport.

The cross-tabulation presented in Table 5 explores the relationship between community awareness of the patient referral system and their reported satisfaction with the referral services in Mbala District. Among the respondents, a higher proportion of those who were aware of the referral system reported not satisfied compared to those who were not aware. The results suggest that while awareness may influence satisfaction at a superficial level, it is not, in itself, a strong determinant of how community members perceive the referral system. These findings underscore that satisfaction with healthcare services, particularly in rural and low-resource settings, is often influenced by a constellation of factors beyond mere awareness. According to Peters et al., physical accessibility, financial barriers, quality of care and responsiveness of healthcare workers play a more decisive role in shaping user experiences.²⁰ Furthermore, Atkinson et al., emphasise that in health systems where logistical challenges such as transport, delays and poor infrastructure persist, awareness without corresponding system performance improvements does little to improve satisfaction.¹⁵ Nonetheless, community awareness remains a foundational component of health system engagement.²⁶ It enhances health-seeking behaviour and enables community members to better understand their rights and the services available to them. Therefore, while awareness alone may not guarantee satisfaction, it is a prerequisite for facilitating timely and appropriate referrals. Efforts to enhance satisfaction should therefore adopt a holistic approach combining awareness campaigns with improvements in emergency transport, infrastructure, health worker capacity and communication systems.

This study is limited by a small sample size, which may reduce statistical power and obscure significant relationships. Furthermore, reliance on self-reported data introduces potential bias, and the cross-sectional design prevents causal inferences. Unmeasured variables, such as fuel availability and staff workload, may also influence referral outcomes. The use of only one respondent per facility may have introduced response biasness.