Investigating drone technology in health-care products delivery in rural communities in Ghana: benefits, barriers and perceptions Open Access

The rise in rural-urban migration, where mostly young, energetic people move to urban centres, has resulted in an ageing and weaker rural population that is susceptible to poor health. Despite this, health systems in rural communities are either underdeveloped or unavailable compared to urban centres (Fan et al., 2021). A World Health Organization (WHO) survey indicates that about 56% of rural dwellers have no access to health care, and the quality of health-care delivery is low in the health facilities within those communities (WHO, 2018). This is due to the increasing demand for primary health care, the unavailability of personnel and the undersupply of health services and products (Blumenthal et al., 2020). The undersupply of health-care products in rural areas is caused by transport unavailability, road inaccessibility and long distances to these communities (Stockton et al., 2021). Besides, the lack of appropriate storage facilities in these rural areas hinders the delivery of specific health products and medicines (Laksham, 2019). Because most rural communities do not have these storage systems (Saechan and Jaworski, 2018), delivering perishable products in large quantities is uneconomical compared to supplying them to the facilities when needed.

However, some health-care products are required within the shortest possible time to save lives during emergencies. These products might arrive too late when supplied through road transport due to the nature of rural roads and the distances to these areas (Nyaaba and Ayamga, 2021). Such a situation has led to the investigation of a faster and more efficient approach to supplying health-care products such as vaccines, blood, anti-snake serum and other emergency medicines to rural communities. In such situations, Scott and Scott (2017) proposed using drones. Drones, also called unmanned aerial vehicles (UAV), are used in various applications like photography, military surveillance, disaster relief operations, industrial monitoring, media coverage and parcel delivery (Otto et al., 2018). There has been good improvement in the technical aspects of the services provided by this technology to the extent that it is currently being exploited for passenger use (Kellermann et al., 2020). Transportation, logistics supply and other services provided by UAVs are estimated to reach about $29.06bn by the year 2027 at a growth rate of 21.01% (Rejeb et al., 2021). Various industries are exploring diverse ventures on the use of drones for efficient, economical and sustainable delivery of goods and services to their clients. Amazon is currently assessing the use of drones and octocopters to supply purchased products to the doorstep of their clients (Sudbury and Hutchinson, 2016). DHL has launched a drone delivery service to aid in a faster and more economical approach to their services (Scott and Scott, 2020). Google and Facebook are also prospecting on providing internet access to emerging countries and hard-to-reach locations using drones (Chandrasekharan et al., 2016).

Another company using the advantage of UAVs is the Zipline drone delivery company. It is a multinational company with the core mandate of designing, manufacturing and operating drones to deliver medical supplies to various health facilities in selected countries worldwide. The company provides specific medical supplies, including frozen plasma, blood, cryoprecipitate, infusions and vaccines (Gangwal et al., 2019). In 2019, there was a partnership between the Government of Ghana and Zipline Company Limited to establish a drone delivery service for health-care supplies in the country. The project tackled key logistical challenges in supplying medical products to hard-to-reach communities nationwide (Damoah et al., 2021). The company signed an agreement to make 600 deliveries of medical products a day for four years at a total project cost of $12m (Asiedu, 2019). In this view, Zipline established four distribution centres at vantage locations in the country to reach the target population in Ghana (Ackerman, 2019). Critics were, however, of the view that the project amount could have been invested in developing existing health systems, such as providing hospital beds, gloves, constant water supply and improving existing buildings (Asiedu, 2019).

Three years into the project, Zipline operates six distribution centres in the country, serving 2,300 health facilities. The company has delivered about 1.7 million lifesaving medical products, a million COVID-19 vaccines and personal protective equipment and about 1.5 million child immunisation vaccines (Cudjoe, 2022). It explored the advantages of drone technology, such as speed, autonomous operations and flying over rugged terrains to achieve these successes. However, the overall efficiency and sustainability of Zipline’s operations include other factors such as the number of UAVs that could be deployed, the loading capacity of the drones, the routes and terrains to delivery sites, the possible distance of travel per round trip (Rejeb et al., 2021) and people’s attitude towards the technology are yet to be investigated. Despite the success of Zipline’s operations in Ghana, no research has synthesised the benefits, barriers and perceptions of drone technologies in the country. This research aims to fill that gap by identifying these factors in rural communities. It specifically addresses the question: What are the benefits, barriers and perceptions associated with drone technology for health-care product delivery in rural Ghana?

This study establishes these factors for drone technology in a developing country, providing a reference for other developing nations. In addition, the success of Zipline’s operations in Ghana counters critics and supports calls for its expansion. The rest of the manuscripts is organised as follows: Section 2, the literature review, covers the theories, existing literature on the topic and conceptual review. Section 3 describes the research methodology, including the sampling techniques, data collection methods and analytical tools used. Section 4 presents the results of the study, followed by a comprehensive discussion that interprets the findings in the context of existing literature. Finally, Section 5 concludes the study by summarising key findings, discussing their implications for practice and policy and suggesting directions for future research.

The Diffusion of Innovation (DOI) theory and the unified theory of acceptance and use of technology are the theories used to investigate the issue under consideration. The DOI theory explains the methods used in communicating innovations through various channels between members of the same social circles. According to Demuyakor (2020), it has been regarded as an authentic model used to determine the adoption of new technologies by various societies. The theory assesses complexity, compatibility, comparative advantage, trainability and observability. These attributes influence the perception and adoption of technology and hence can be used concerning the adoption and use of drone delivery services. Five sequential adoption segments exist in DOI: innovators, early adopters, early majority, late majority and laggers (Chen et al., 2022). The unified theory of acceptance and use of technology (UTAUT) assesses the intention of users to adopt technology and subsequently studies their usage behaviour. UTAUT is defined by constructs such as effort expectancy, enabling conditions, social influence and performance expectancy (Yoo et al., 2018). UTAUT combines the attributes of DOI, the technology acceptance model (TAM) and the theory of reasoned action (TRA) to develop a comprehensive model (Leon et al., 2021). According to Leon et al. (2021), the term “perceived usefulness” in UTAUT is the most robust framework for predicting the intention of use. Regarding drone delivery, Yoo et al. (2018) suggested the relative advantage of speed and environmental friendliness as strong predictors. Jasim et al. (2022) posited that even though UTAUT performs better in assessing the perception of people on the acceptance and use of drone delivery, especially for medical product delivery, they have been predominantly used in assessing organisational viewpoints on the use of drones.

The DOI set the guidelines for rural communities’ acceptance while UTAUT set the tone for the barriers and perceptions towards drone technology usage in health product delivery. Relating the theories to the results achieved, DOI showcases the comparative advantage (reaching remote areas, rapid mode of transport and flying over rough terrains) of drones in the delivery of health-care products in rural communities. On the other hand, UTAUT identified bad weather, drone payload capacity and shortage of trained personnel as barriers that may hinder drone health-care product delivery. however, UTAUT’s social influence (that is, using drones for health-care delivery influences the health-care industry’s development positively) overrides the barriers stated.

Drone technology refers to using aircraft that can be operated remotely or autonomously without a human pilot on board (Otto et al., 2018). Drones are equipped with various sensors, cameras and other technology that enable them to perform multiple tasks, such as surveillance, inspection, mapping and delivery (Otto et al., 2018). They can be used for commercial and military purposes. The health industry has also used drones for telemedicine and emergency medical supplies. The technology is helpful in the delivery of health-care products, such as vaccines, medications and other medical supplies, to remote or underserved areas (Javaid et al., 2022). Current research areas on the use of drone technology include optimisation in routing and scheduling, reliability, safety, energy efficiency and inventory management (Amoiralis et al., 2014; Garvanov et al., 2021; Kritskiy et al., 2018; Poikonen and Campbell, 2021). Za’im Sahul Hameed et al. (2023) emphasised that drone usage could prevent mortality in obstetric emergencies in rural areas in Sabah and Sarawak, Malaysia, because health workers have a positive perception of medical drone delivery which results in their high acceptance level.

Drones have shown significant promise in transforming health-care delivery, especially in rural areas, by overcoming logistical challenges. However, their adoption is not without barriers. Several studies have explored these challenges, particularly in relation to infrastructure, regulation and operational constraints. Koshta et al. (2022) highlighted the lack of government regulations as a major challenge in adopting drones for health-care delivery, alongside issues such as limited flight range, adverse weather conditions and a shortage of skilled labour. These operational barriers are echoed by Zhu et al. (2024), who used linear programming to identify challenges like inadequate infrastructure and long, complex distribution chains in rural areas. Similarly, Aggarwal et al. (2023) identified coordination issues with local authorities, weather-related problems and social behaviours as significant hurdles to drone-based medical supply delivery in Northeast India.

Despite these challenges, there are several economic and operational advantages to using drones in health-care supply chains. Gunaratne et al. (2022) conducted a comparative analysis of existing and proposed drone delivery systems, demonstrating that drones can significantly reduce costs, delivery times and energy consumption, particularly in regions with poor road infrastructure. This economic benefit is further supported by Ospina-Fadul et al. (2024), who demonstrated that drones in Ghana’s Western North Region could not only prevent diseases but also offer cost-effective delivery solutions through Incremental Cost-Effectiveness Ratios (ICERs). Haidari et al. (2016), in a similar study, concluded that drones are economically viable for vaccine distribution, with lower operational costs outweighing capital investment. Eksioglu et al. (2024) also explored the role of drones in optimising vaccine delivery in low-income regions with poor cold chain infrastructure, showing that strategically placed drone hubs could significantly improve vaccine access.

Further research continues to highlight the operational advantages of drones over traditional delivery methods. Zhu et al. (2024) reaffirmed that drones offer low-cost, high-speed solutions for remote rural areas, whereas Bridgelall and Tolliver (2024) found them to be more cost-effective and efficient than trucks for rural deliveries. Kaplan and Heaslip (2024) assessed the environmental benefits of drones, concluding that they can reduce delivery emissions and operational costs. Sham et al. (2022) explored the role of drones in maintaining supply chains during the COVID-19 pandemic, emphasising their environmental advantages and potential to enhance health-care resilience during crises. Azmat and Kummer (2020) also proposed using autonomous cars alongside drones to address logistical challenges faced by International Humanitarian Organizations, offering solutions to overcome traditional supply chain barriers in emergency settings. Xiao et al. (2024) expanded on this concept by modelling the combined use of trucks and drones, demonstrating that this approach improves delivery efficiency and reduces environmental impacts, particularly as rural e-commerce continues to grow.

Public acceptance is another crucial factor influencing the adoption of drone technology in health-care delivery. Tamakloe et al. (2024) examined consumer attitudes towards drones, revealing that acceptance varies based on gender, income and location, with targeted education and marketing strategies recommended to enhance adoption. Similarly, Khan et al. (2019) identified privacy concerns as a significant obstacle to consumer acceptance of drone deliveries in urban areas, advocating for the integration of privacy-enhancing features in drone systems. Nyaaba and Ayamga (2021) contributed to this discussion by reviewing literature on policies that could foster public trust and acceptance of drones in health-care delivery across Africa.

Technological advancements continue to enhance the potential of drones in health-care delivery. Damoah et al. (2021) emphasised the role of artificial intelligence (AI) in optimising drone operations, demonstrating how AI can improve the efficiency of health-care supply chains. Gupta et al. (2021) also introduced a blockchain-based drone delivery system that reduced transaction costs by 44.56% compared to traditional methods, demonstrating the potential of integrating emerging technologies to enhance drone delivery systems. Mohsan et al. (2023) proposed the use of laser power transfer and low power and lossy network techniques to address the issue of battery limitations, thereby extending the operational range and reliability of drones in medical logistics. Cheng et al. (2024) proposed a two-period drone scheduling model to mitigate wind-related delays, using adaptive afternoon schedules based on real-time weather updates. Their model minimises delivery lateness using a cluster-wise ambiguity set to account for uncertain flight times.

Drone regulation in Ghana is evolving to accommodate its expanding applications in sectors such as agriculture and health care. O’Sullivan et al. (2024) highlighted that while drones offer significant potential for medical supply chains, several barriers must be addressed before widespread adoption can be achieved, with regulatory hurdles being a key challenge. Nyaaba and Ayamga (2021) emphasised that regulatory bodies must continuously develop and update policies to keep pace with technological advancements like drone technology.

One of the primary regulatory challenges is airspace management. Olatunji et al. (2023) identified Africa’s congested airspace as a major obstacle, underscoring the need for sophisticated systems to ensure safe drone operations, including collision avoidance and real-time flight tracking. Aviation authorities such as the Federal Aviation Administration in the USA and the European Union Aviation Safety Agency in Europe have established stringent regulations governing drone operations. These rules generally restrict drones from exceeding specific altitude thresholds, prohibit flights near airports and limit operations over densely populated areas unless special authorisation is granted (O’Sullivan et al., 2024).

To address airspace challenges, Kellermann et al. (2020) proposed the creation of designated air corridors for drones, akin to roadways for ground vehicles, allowing drones to operate safely without interfering with manned aircraft. They also advocate for the integration of unmanned traffic management systems to coordinate drone movements, prevent collisions and efficiently manage airspace. The implementation of these systems is essential for ensuring that drones can be safely integrated into national airspace while complying with regulatory requirements. In Ghana, regulatory efforts are already underway. Dadey (2022) notes that the Ghana Civil Aviation Authority has implemented drone safety regulations, including provisions for drone registration, pilot certification and commercial operation approvals. However, as drone usage continues to expand, ongoing regulatory updates will be necessary to ensure safety, efficiency and public trust in the technology.

Privacy and security concerns are pivotal in shaping public trust and the adoption of drone technology. Despite existing regulations, Ghana lacks a dedicated framework for drone-specific data protection, which poses risks to citizens’ rights and raises apprehensions about data breaches and unauthorised access to sensitive medical information (Dadey, 2022; Ayamga et al., 2021). While rural settings in Ghana mitigate some privacy concerns due to lower population density, challenges remain, particularly as drone technology advances with innovations such as AI, blockchain and autonomy. Integrating privacy-enhancing technologies such as blockchain and encryption offers promising solutions to these challenges. Blockchain systems can ensure secure and tamper-proof medical supply chains, whereas encryption protects data during transmission, addressing confidentiality risks (Gupta et al., 2021). Damoah et al. (2021) emphasised that regulatory frameworks must evolve to account for these emerging technologies, including remote monitoring, autonomous flight and drone swarm operations. Effectively implementing these measures could significantly bolster public trust, enhance data security and drive the broader adoption of drone technology in Ghana, which is a key focus of this study.

This section has revealed an understanding of the factors that influence the deployment of drone technology in health-care delivery, which directly informs the research objectives of the study. It highlights benefits such as drones’ ability to overcome geographical challenges, reduce delivery times and lower operational costs (Gunaratne et al., 2022; Haidari et al., 2016; Zhu et al., 2024), which align with Zipline’s success in delivering critical medical supplies in rural Ghana. However, the literature underscores significant barriers, including limited drone payload capacity, flight range, adverse weather conditions and insufficient infrastructure (Koshta et al., 2022; Zhu et al., 2024), factors critical to understanding the limitations of Zipline’s operations in these areas. Various studies have also proposed technological advancements to address barriers such as limited battery life and poor weather resistance. In addition, the perceptions of health-care workers and local communities, which influence the adoption and integration of drone technology, emerge as another crucial aspect identified in the literature (Za’im Sahul Hameed et al., 2023; Khan et al., 2019). Despite the documented success of Zipline, there is limited research exploring these factors in the context of rural Ghana, particularly regarding public attitudes, trust and acceptance of drone technology. This study draws on the reviewed works to investigate the specific benefits, barriers and perceptions associated with drone health-care delivery in rural Ghana. By integrating findings from existing studies, this research aims to provide actionable recommendations on overcoming barriers to adoption, optimising delivery systems and ultimately enhancing health-care delivery in underserved regions.

Table 1 presents a list of benefits associated with the drone delivery of health-care products to health-care facilities. Each benefit is accompanied by a description and related papers that provide insights and evidence on the benefit. These benefits demonstrate the potential of drones to revolutionise the delivery of health-care products, particularly in remote and hard-to-reach areas, by improving access, speed and efficiency while reducing costs and environmental impact. Despite the numerous benefits of drone technology to improve access to health-care products and services, several barriers hinder their practical use in delivering health-care products to health-care facilities. Table 2 presents some common obstacles associated with using drones for health-care delivery, their corresponding descriptions and related research papers. By understanding and addressing these barriers, stakeholders in the health-care industry can better leverage the potential benefits of drone technology to improve access to health-care products and services. Table 3 outlines ten factors related to the perception of drone technology in health-care product delivery, which aligns with the UTAUT model. The model is a widely recognised framework used to understand the factors influencing individuals’ acceptance and use of technology. All components of the UTAUT model can impact individuals’ perceptions towards drone technology in health-care product delivery.

The study adopted the quantitative research approach to investigate the perception, benefits and barriers influencing drone technology in the health-care products supplied by Zipline in Ghana. The study population consist of the health workers in the 410 facilities [that is, community-based health planning and services (CHPS) compound, Health centre, Clinic, Polyclinic, Hospital and Health Directorate] served by Zipline Sefwi Wiawso. A purposive sampling was used to identify experts associated with the operations of Zipline Sefwi Wiawso, who were consulted to provide insights on the objectives of the research. The authors reached out to the Research Department of Zipline Ghana Limited and five employees from Zipline Ghana Sefwi were provided to assist with the research. These experts were selected based on their professional backgrounds, experience with drone technology and roles related to drone delivery services. The purposive sampling technique was used because it helped the researcher seek information from specific individuals with the required knowledge and experience (Lavrakas, 2008).

The experts assisted in identifying the health facilities that use Zipline’s drone delivery services in the area. It was discovered that some of the health facilities that were of interest in the study were difficult to reach and could record zero response rates during the conversation with the experts. To address this issue, the researcher considered factors such as cost of identifying population element, geographical spread and ease of data collection and selected convenience sampling as the most suitable technique for identifying the health facilities that were willing to participate in the study and their health workers readily available (Frey, 2021). Convenience sampling is often used when the population is difficult to reach or when time and resources are limited. The shift in sampling techniques aimed to include a broader range of participants who could provide practical insights based on their experiences with drone delivery services. The study explicitly acknowledges that convenience sampling can lead to biases due to the non-random selection of participants. This recognition helps in transparently presenting the limitations of the study. The strategy resulted in 70 health facilities meeting the selection criteria and being included in the study. Purposive sampling was used again to select three health workers who were directly involved in the request, receipt or administration of health-care products delivered by drones from Zipline to each of the 70 sampled facilities. Overall, the study yielded a sample size of 210 respondents.

Primary data was collected using a closed-ended questionnaire with scaled-response types. The questions asked during the survey were developed using information gathered from existing literature about the use of drones for delivering health-care products. Various factors were identified and grouped into three main categories: benefits, barriers and perceptions. The questionnaire was divided into four parts. Part A deals with the participants’ demographics, whereas Part B identifies the benefits associated with the drone delivery of health-care products to health-care facilities. Part C dives into the barriers that hinder the practical use of drone technology in delivering health-care products to health facilities. Part D outlines the perceptions of using drones for health-care product delivery (see  Appendix). The questionnaire developed from the factors was administered to the respondents using internet-based Google Forms. Five experts from Zipline were asked to check the validity and reliability of the factors derived from the literature before the main survey. These experts were employed to verify the elements that pertain to Zipline’s operations in Sefwi Wiawso and make inclusions and exclusions on those factors not captured in the literature. Validating these factors was relevant because it was adopted for developing the questionnaire used in the study. The feedback led to several refinements, including adjustments in wording, question order and the inclusion of additional items to ensure comprehensive coverage of the topic.

The survey was pre-tested with ten respondents, two from each facility, to gather pertinent input on the questionnaire and reduce ambiguity. The pre-testing of the questionnaires was done to evaluate the length, clarity and feasibility of the survey tool, as well as the time it took participants to respond to the questionnaire (Toepoel, 2017). Respondents on the pre-test noted that the questionnaires were straightforward to understand and very likely to yield useful information in the main survey. The researcher’s supervisor also made significant inputs, which enhanced the consistency of the questionnaire and the entire survey.

This study uses quantitative data analysis techniques by coding the responses received from the questionnaire. The data investigation has begun by conducting a descriptive analysis using frequency distribution and means score tests. The frequency tables summarise the distribution of the variables derived from the questionnaire. This was used to describe the profile of the respondents. It was followed by inferential analysis to better comprehend patterns and relationships in the data set. Correlation analysis was conducted to determine a significant relationship between variables and provide valuable insights into the nature of the connection (Shrestha, 2021). The inclusion of correlation analysis in the study helps to examine the degree of association between the key variables of interest and also identifies critical factors that may significantly impact the research objectives. The correlation analysis sets the basis for the factor analysis, which is preceded by the Kaiser–Meyer–Olkin (KMO) and Bartlett’s test. The KMO test evaluates the suitability of the data set for factor analysis and whether the variables are adequately correlated. A KMO value above 0.7 indicates good suitability for factor analysis (Shrestha, 2021). From Bartlett’s test, a p-value < 0.05 indicates that the variables are interrelated, supporting the use of factor analysis (Shrestha, 2021).

Principal component analysis (PCA) was chosen over other extraction methods like Principal Axis Factoring (PAF) and Maximum Likelihood (ML) estimation for the analysis as it is suitable for exploratory dimension reduction, focusing on identifying underlying common factors influencing the use of drones in the study area, cover multicollinearity detection, very versatile and computationally easier, although it assumes linearity (Bharadiya, 2023; Dorabiala et al., 2024; Abdi and Williams, 2010). PCA is useful when no specific theory guides the selection of factors (Shrestha, 2021). PAF is suitable when there is a theoretical basis for assuming that observed variables have common and unique components. In contrast, ML estimation is suitable when the data are assumed to be multivariate and normally distributed (Shrestha, 2021). The suitability of the data set for PCA was also assessed using the KMO measure of sampling adequacy and Bartlett’s test of sphericity. These tests were conducted to determine the statistical significance and appropriateness of the data set. This approach tends to validate the results obtained from the correlation results.

The extraction limit of the eigenvalue cut-off of 1 was chosen to prioritise robust and meaningful factors while minimising noise or measurement error. Auerswald and Moshagen (2019) asserted that the outcome of an initial factor analysis can be challenging to interpret because the factors extracted may correlate. This can be solved by applying a rotation that helps transform the original factor structure to a new configuration that is easier to understand and aligns the factors more independently (Auerswald and Moshagen, 2019). The commonly used rotation methods are varimax, noblemen and quartimax (Nguyen and Waller, 2022). The researcher used varimax rotation for this study due to its ability to simplify and enhance the interpretability of PCA results. A confidence interval strategy was chosen for the analysis. The strategy was used on the significant values obtained from the correlation studies and Bartlett’s test. The study used confidence intervals of 95%, and 99% to provide the necessary information on both the test of significance and variability of the estimate.

A total of 230 questionnaires were distributed to the health workers out of which 214 responded, resulting in a response rate of 93%. The data collected via Google Forms was exported to Microsoft Excel for data cleaning and organisation using specialised Excel functions. The refined data comprising 210 data sets was coded in SPSS version 26. According to Glaser (2012), including the respondent’s profile in the results helps researchers and readers understand the background of the participants, thereby adding credibility to the responses and overall findings of the research.

The analysis uses frequency distribution and percentages, as presented in Table 4. This table describes the distribution of respondents by gender. Among the participants, 99 (47.1%) identified as male, whereas 111 (52.9%) identified as female. The data indicate a relatively balanced gender representation, with females slightly outnumbering males.

The majority of respondents were between 26 and 35 years old (59.0%), followed by those aged 36–45 (27.1%). The remaining age groups had smaller representations: 18–25 years (9.0%), 46–55 years (3.3%) and 56 years and above (1.4%). These findings suggest that the study primarily captures the perspectives of individuals in their late 20s to mid-30s, though a diverse range of age groups is included.

The findings also highlight the different types of health-care facilities represented in the study. The largest proportion of respondents (45.7%) were affiliated with CHPS compounds. Health centres were the second most common facility type, accounting for 22.9% of respondents. Hospitals had a significant representation at 20.5%, whereas Health Directorates and clinics comprised 4.8% and 4.3% of respondents, respectively. Only 1.9% of participants were associated with polyclinics.

Regarding educational background, most respondents (41.9%) held a diploma, followed by 24.8% with a certificate qualification. Respondents with a bachelor’s degree accounted for 22.9%, whereas 7.1% had completed a master’s degree. A small proportion (2.4%) had acquired a Higher National Diploma (HND), and 1% had Senior High School (SHS) as their highest educational level.

Table 4 also presents the respondents’ years of work experience. Nearly half (49.5%) had less than five years of experience in their respective fields. Around 30% of respondents reported having between 5 and 10 years of experience, whereas 17.1% had 11–15 years. A small proportion (1.9%) had between 16 and 20 years of experience, and 1.4% had more than 20 years.

The mean and standard deviation for each benefit, barrier and perception, as rated by participants using Likert scales, are presented in Table 5. Higher mean scores in these three categories indicate a greater perceived benefit, a stronger perceived barrier and a more positive perception.

Under benefits, participants strongly agreed that UAVs have the capability to reach remote areas, with a mean score of 4.53. This was followed by their effectiveness as a rapid mode of transport (mean = 4.51) and their role in improving emergency response (mean = 4.43). These benefits also exhibited low variability in responses, as indicated by their lower standard deviations. The descriptive analysis further showed that all benefits identified from the literature had mean scores greater than the median value and were therefore included in further analysis.

Among the barriers, the most significant challenge identified was the difficulty of UAVs flying in bad weather (mean = 3.22), followed by limited infrastructure (mean = 3.21) and restricted payload capacity due to product quantity limitations (mean = 3.18). The analysis revealed that only five barriers had mean scores above the median value of 3 on the five-point Likert scale. Consequently, further analysis focused on these five barriers.

Regarding perceptions, respondents ranked attitude towards using technology as the most important attribute influencing the adoption of drones for health-care product delivery, with a mean score of 4.42. This was followed by complexity (mean = 4.32) and personal innovativeness (mean = 4.30). The descriptive analysis showed that only seven perception attributes had mean scores above the median value of 3. Therefore, only these seven attributes were included in further analysis.

Tables 6–8 present the relationships between factors associated with the benefits, barriers and perception of drones in health product delivery. All the tables show a positive correlation among the factors, indicating the presence of multiple benefits and few barriers when using drones in health care, also suggesting that the factors under perception are interrelated and tend to co-occur in the context of the perception of drone delivery. All the factors in the tables had lower significance levels (p < 0.01), indicating strong evidence for meaningful relationships between variables. The results establish a solid foundation for increased reliability and validity of the data set, thereby enhancing its credibility and overall robustness.

Table 9 presents the results of the KMO measure of sampling adequacy and Bartlett’s test benefits, barriers and perception. The KMO values of 0.925, 0.714 and 0.898 for the benefits, barriers and perception, respectively, indicate that the sample size is adequate for conducting factor analysis. The test statistics values for Bartlett’s test and the respective degrees of freedom and significance level (Sig.) all being 0.000 indicate that the correlation matrix significantly differs from the identity matrix, providing evidence that factor analysis is appropriate for the data. Consequently, the PCA was performed on the validated data set.

Table 10 displays the total variance in the benefits of using drones for health product delivery. Component 1, with an initial eigenvalue of 8.353, accounts for 59.662% of the variance, whereas its rotation sums of squared loadings indicate a contribution of 43.844% of the variance. Component 2 has an initial eigenvalue of 1.069, explaining 7.634% of the variance, and its rotation sums of squared loadings represent 23.452% of the variance. The remaining components with eigenvalues below 1 have smaller contributions to the variance.

The PCA results on the benefits of adopting drones for health-care product delivery are presented in Table 11. The table shows that all communalities after extraction were greater than 0.5, indicating that the extracted components explain a substantial proportion of variance in each item. Only rotated component loadings above 0.5 were retained and ranked, ensuring that the analysis focused on variables with meaningful relationships to the components. This approach enhances the clarity, relevance and interpretability of the PCA results, as highlighted by Ajtai et al. (2023).

Among the identified benefits, reduced inventory cost was ranked highest, followed by reduced inventory waste and rapid mode of transport. Although benefits related to reduced waiting time and reduced carbon footprint had the lowest loadings, they still played a meaningful role in defining the components.

The PCA results suggest that the benefits of drone delivery can be grouped into two distinct components. Component 1 is associated with the quality improvement of drone services. This includes rapid mode of transport (0.790), reduced labour cost (0.788), reaching remote areas (0.780), reduced transportation cost (0.772), improved emergency response (0.770), enhanced customer-centric delivery (0.739), reduced losses from delayed delivery (0.711), improved storage conditions (0.674), reduced product damage/contamination (0.657), reduced waiting time (0.643) and reduced carbon footprint (0.638). This component highlights how drones enhance service delivery by ensuring faster response times and improving delivery conditions, particularly in rural areas. It also contributes to sustainability by reducing carbon emissions.

Component 2, on the other hand, is strongly associated with operational and logistical efficiencies. Key benefits include reduced inventory cost (0.846), reduced inventory wastage (0.804) and increased product accessibility (0.707). This component emphasises the role of drones in lowering transportation costs, speeding up product delivery and improving access to essential health-care products. Collectively, these factors contribute to a more efficient and cost-effective health-care supply chain.

Table 12 presents the variance explained regarding the barriers associated with using drones for health product delivery. Components 1 and 2 had initial eigenvalues of 2.732 and 1.068, explaining 54.641% and 21.356% of the variance, whereas their rotation sums of squared loadings account for 45.755% and 30.241% of the variance, respectively. All the remaining components had initial eigenvalues below 1 and explained smaller proportions of the variance. These results suggest that the barriers to using drones for health-care product delivery can be categorised into two distinct components.

Table 13 presents the PCA results on the barriers associated with using drones for health-care product delivery. Similar to the findings on benefits, the communalities after extraction were all greater than 0.5, indicating that the extracted components explain a substantial proportion of variance in each item. In addition, only rotated component loadings above 0.5 were retained and ranked to ensure the analysis focused on the most relevant barriers. Among the identified barriers, limited payload capacity was ranked highest, followed by difficulty operating in bad weather and lack of trained personnel. Although limited infrastructure had lower loadings, it remained significant in defining the components.

The PCA results suggest that the barriers to drone delivery can be grouped into two distinct components. Component 1 represents logistical and technical limitations, including limited payload 1 (0.856), limited payload 2 (0.813), difficulty operating in bad weather (0.778) and limited infrastructure. This component highlights key operational constraints, such as the restricted carrying capacity of drones, their inability to function efficiently in adverse weather conditions and the challenges they face in transporting health-care products effectively.

Component 2 is associated with operational and organisational barriers, primarily the lack of trained personnel (0.815). This component underscores the human resource challenges involved in the successful deployment and management of drone delivery systems. Addressing this barrier requires investment in training programs and workforce development to facilitate the effective integration of drones into health-care logistics.

Table 14 displays the total variance explained by the perception of using drones for health product delivery. Component 1, with an initial eigenvalue of 4.079, accounts for 58.274% of the variance for both the initial value and extraction sum of squares loading. The remaining components, with eigenvalues below 1, have smaller contributions to the variance. These results suggest that the perception of using drones for health-care product delivery can be categorised into single distinct components.

Table 15 presents the PCA results assessing perceptions of drones for health-care product delivery. The table includes communalities, component matrix values and rankings of perception variables, highlighting key factors that influence the adoption of drones in health care.

The extraction values range from 0.299 to 0.724, indicating varying levels of contribution to the primary component. Attitude Towards Using Technology has the highest extraction value (0.724), suggesting it is strongly represented, whereas Invasion of Privacy (0.299) has the weakest representation.

Similarly, in the component matrix, Attitude Towards Using Technology has the highest loading (0.851), emphasising its crucial role in shaping perceptions. This is followed by Compatibility (0.844) and Performance Expectancy (0.836), reinforcing their importance in influencing acceptance. In contrast, Relative Advantage (0.569) and Invasion of Privacy (0.546) have lower loadings, indicating a comparatively smaller impact on overall perceptions.

These findings suggest that positive attitudes, compatibility and performance expectations are the most influential factors in shaping public perception of drone adoption for health-care delivery. Conversely, concerns about privacy and perceived relative advantage play a lesser role in determining acceptance.

The results in Table 5 indicate that all the benefits identified from the literature had mean scores higher than the median, suggesting that drone technology provides significant advantages for health-care product delivery. Respondents particularly emphasised the ability of drones to reach remote areas and their rapid mode of transport. This finding aligns with the work of Bridgelall and Tolliver (2024), Euchi (2021) and Nyaaba and Ayamga (2021), who highlighted the crucial role of drones in delivering medical supplies, vaccines and other essential products to remote locations that are difficult to access via traditional transportation methods.

In addition, the findings support the arguments made by Kaplan and Heaslip (2024) and Scott and Scott (2020) regarding the ability of drones to navigate rough terrains, impassable roads or areas affected by natural disasters. This capability makes them highly valuable in emergency situations where timely medical supply delivery is critical. Furthermore, the results confirm the effectiveness of drones in improving emergency response times, as highlighted by Zhu et al. (2024), Sham et al. (2022), Zailani et al. (2020) and Scott and Scott (2020). Their research underscores the ability of drones to provide faster transportation compared to conventional methods, significantly reducing delivery times from hours to minutes. The findings of this study reinforce these claims, emphasising rapid transportation as a key factor in enhancing the adoption of drones for health-care product delivery.

Further analysis, conducted through PCA in Table 11, identified two main components explaining the relationships between the observed variables. Component 1 highlights the role of drones in streamlining delivery processes, reducing costs and improving operational efficiency. In contrast, Component 2 focuses on the benefits of optimising inventory management, minimising waste and enhancing the availability and accessibility of health-care products.

Among the observed variables, reduced inventory cost had the highest rotated component loading in Component 2, whereas rapid mode of transport had the highest loading in Component 1. These loadings indicate the contribution of each variable to the formation of the principal components. Variables with higher loadings exert a stronger influence on the component’s structure and provide insights into underlying patterns within the data set analysed using PCA (Kherif and Latypova, 2020).

Based on these findings, the benefits of using drones for health-care product delivery can be broadly categorised into two key areas: reducing inventory costs and enhancing transport efficiency.

The results in Table 5 indicate that several barriers hinder the integration of drones into existing health-care product delivery systems. Among the 15 barriers examined in the literature, 5 had mean scores higher than the median, suggesting that these specific challenges are particularly significant and could greatly impact the successful implementation of drone technology in health-care delivery.

Two major challenges that emerged from the analysis were the difficulty of flying drones in bad weather conditions and limited infrastructure. These factors appear to be critical obstacles to the widespread adoption of drones for medical deliveries. Euchi (2021) highlighted how adverse weather conditions prevent customers from receiving their deliveries on time, whereas Sah et al. (2021) pointed out that drones are often unable to operate beyond the range of the delivery hub due to infrastructural constraints.

Further analysis using PCA (Table 13) identified two key components that explain the relationships among the observed variables. Component 1 emphasises challenges related to payload capacity. Rabta et al. (2018) and Sah et al. (2021) stressed that payload limitations significantly affect the effectiveness of drone delivery systems. The restricted carrying capacity of drones limits the types and quantities of medical supplies they can transport, impacting both efficiency and service reliability. In addition, Aggarwal et al. (2023) and Sah et al. (2021) explained that during adverse weather conditions, drones experience reduced stability due to aerodynamic disruptions. To maintain flight control and avoid accidents, drones must adjust by reducing their payload, further restricting the volume of medical products they can carry. This finding underscores the strong link between payload capacity limitations and challenges in operating drones under extreme weather conditions.

Component 2 highlights the shortage of trained personnel capable of efficiently managing drone delivery operations. Raj et al. (2022) emphasised that skilled personnel are essential for fully harnessing drone technology in health-care logistics. A lack of trained staff familiar with placing requests and interacting with drone systems can lead to delays or errors in medical product orders. Similarly, Jeyabalan et al. (2020) and Nyaaba and Ayamga (2021) stressed the need for targeted training programs to equip health-care workers with the necessary skills to use drone delivery services effectively. Proper training ensures proficiency in submitting accurate and timely requests, ultimately improving the efficiency of drone-based health-care supply chains. These findings reinforce the importance of addressing both technical limitations (payload capacity and weather challenges) and operational barriers (lack of trained personnel) to enhance the successful adoption of drone technology in health-care product delivery.

Goodchild and Toy (2018) affirmed that drone distance travelled determines its CO₂ emission level, whereas Euchi (2021) believed that drone usage reduces greenhouse gas emissions when used for fright purposes. Sah et al.’s (2021) study ranked harming wildlife, CO₂ emissions, visual pollution and sound pollution as the environmental issues arising from drone use. Building on the literature reviewed participants’ low ratings on ecological concerns and noise pollution, may be traced to their lack of awareness of such barriers because they are at the receiving end.

The data analysis revealed that seven out of ten perception attributes confirmed the acceptance of drones for health-care product delivery in rural areas. This is evident from the mean score values, where these seven variables had scores higher than the median. Among them, attitude towards using technology received the highest level of agreement from respondents, indicating strong support for drone integration in health-care logistics.

Further analysis using PCA (Table 15) identified one primary component explaining the relationships among the observed variables. The PCA ranking highlighted attitude towards using technology as the most influential factor shaping the overall component structure. This suggests that respondents perceive drone technology as a positive advancement in the health-care industry.

Expanding on this finding, respondents demonstrated a strong positive attitude towards drone-assisted health-care delivery, recognising its potential to enhance medical services, particularly in remote and underserved regions of Ghana. This reflects a widespread belief in the ability of drones to improve health-care accessibility, efficiency and responsiveness. Pande and Taeihagh (2021) argued that such positive perceptions align with the concept of relative advantage, where new technology is embraced based on its perceived benefits over existing methods.

Similarly, Za’im Sahul Hameed et al. (2023) and Cai et al. (2021) highlighted the growing acceptance and enthusiasm surrounding drone technology, emphasising its potential to transform health-care logistics in areas with challenging geographical conditions. Supporting these insights, Jasim et al. (2022) suggested that when drone delivery systems align well with existing health-care processes and complement the responsibilities of medical personnel, attitudes towards adoption tend to be more favourable.

These findings underscore the importance of technological compatibility, efficiency and perceived benefits in shaping attitudes towards drone adoption in health care. As drone technology continues to evolve, its ability to integrate seamlessly into health-care systems will be crucial for driving widespread acceptance and implementation.

This study examined the factors influencing the use of drone technology for health-care product delivery in rural communities, with a specific focus on the Sefwi Wiawso municipality. The findings highlight significant benefits of drone technology, particularly in reducing inventory costs and providing a rapid mode of transport. However, key barriers such as limited payload capacity and lack of trained personnel were also identified. Overall, the perception of using drones for health-care deliveries was positive, with a strong emphasis on attitudes towards technology adoption.

To address the payload capacity barrier, the study recommends that further research and development should focus on designing drones with increased carrying capacity while maintaining flight stability and safety. This aligns with the study’s findings that payload limitations significantly impact drone efficiency. Implementing lightweight materials and advanced propulsion systems can enhance payload capacity, allowing drones to transport a wider range of medical supplies. In addition, improving aerodynamics and stability features will help mitigate the impact of adverse weather conditions.

To maximise the benefits of drone technology, stakeholders should invest in education and training programmes for health-care workers and relevant personnel. Providing comprehensive training on drone operation and integration into existing health-care processes will enhance efficiency and address the study’s findings on the need for better-trained personnel.

To encourage the adoption of drone technology, it is essential to involve health-care workers in the decision-making process from the outset. Their insights can help identify potential concerns, determine suitable use cases and overcome implementation barriers. Engaging health-care professionals in the process will foster greater acceptance and ensure that drones are perceived as an enhancement to service delivery rather than a disruptive force.

Future research should explore the long-term impact of drone deliveries, including their effectiveness in improving health-care access, reducing mortality rates and enhancing overall health-care systems. In addition, comprehensive economic evaluations of drone-based health-care delivery are needed, assessing cost-effectiveness, financial feasibility and comparisons with traditional delivery methods. Understanding the economic viability and scalability of drone-based health-care solutions will be critical for policymakers and stakeholders.

Moreover, further studies should investigate existing regulations and policies governing drone technology in health care, both in Ghana and internationally. Identifying regulatory gaps and proposing solutions for better integration into health-care supply chains will be essential for developing robust legal frameworks that support safe and efficient drone operations.

In addition, drone deliveries raise concerns regarding privacy and security, particularly in safeguarding medical supplies and patient data. Establishing strong policies and security frameworks will be crucial to ensuring data protection, secure delivery operations and overall public trust in drone technology for health-care logistics.

Several limitations should be considered when interpreting the findings of this study. Firstly, the study focused exclusively on the Sefwi Wiawso municipality in Ghana’s Western North region, which limits the generalisability of the results to other areas. In addition, the study examined only health-care facilities served by Zipline in this specific area, meaning the findings may not fully reflect operations at other Zipline distribution centres in Ghana or other African countries.

Secondly, the sampling method relied on purposive and convenience sampling, which may introduce biases affecting the reliability of the study’s outcomes. These techniques, although practical for accessing the target population, may limit representation, as not all individuals had an equal chance of participation. As a result, findings could be skewed towards respondents already supportive of drone technology.

Finally, the study relied on self-reported data, which may be subject to social desirability bias, recall bias and individual interpretation of survey questions. These factors can introduce inaccuracies in responses. Future research should consider more diverse geographic locations, adopt rigorous sampling techniques and incorporate objective data sources to complement self-reported information and improve the robustness of findings.

By addressing these limitations, future studies can build on this research, contributing to a more comprehensive understanding of how drone technology can be effectively integrated into health-care supply chains.

Figure A1 

Executive summary

Rural communities are mainly associated with ageing and weaker populations. Besides, health-care facilities in such areas are mostly underdeveloped and unavailable depriving 56% of rural dwellers access to basic health care. The poor access to health care in rural communities has been attributed to road inaccessibility, transport unavailability and poor health-care storage facilities among others. One major factor contributing to this challenge is the unavailability of health-care products in these areas during emergencies. Such is the agreement between the Government of Ghana and Zipline Company Limited to deliver health-care products to hard-to-reach communities using drones. So far, Zipline operations have served over 2,300 health-care facilities delivering about 1.7 million lifesaving medical products, a million COVID-19 vaccines and about 1.5 million child immunisation vaccines. In this regard, drone technology is being accessed for the timely delivery of health-care products, primarily in remote or hard-to-reach communities. This study seeks to assess the factors that influence the use of drone technology in the supply of health-care products in the Sefwi Wiawso municipality of the Western North Region of Ghana. The research adopts a quantitative approach through closed-ended questionnaires to evaluate the benefits, barriers and perceptions associated with drones to deliver health-care products to the communities under study. The questionnaire is designed using the general factors derived from the literature. The responses received were then analysed using principal component analysis to determine the specific factors relevant to the area. The results indicate that efficiency and cost-effectiveness, inventory management and accessibility of health-care products are the significant benefits accompanying drone technology. However, the study also identified that limited payload capacity hampers the range of medical products that can be transported and quantities that can be delivered and the lack of trained personnel as barriers to using drone technology for health-care product delivery. In addition, health workers have the perception that the use of drones in the health-care industry is influenced by the attitude of health personnel towards the use of technology. Health workers have a favourable inclination towards the utilisation of drones for product delivery. They perceive drone technology offers substantial enhancements to health-care services.