2.1. Drone technology use in healthcare

Drone technologies have been defined in the literature in diverse ways according to the context [4–6,18]. In the area of agriculture, drone technologies are essential unmanned aerial vehicles (UAV) used for observations, sensing, spraying pesticides and providing information to farmers [3,19]. In healthcare, drone technologies have been defined as UAV used for the transport of medical goods and services (e.g., emergency blood supplies, vaccines, medicines, diagnostic samples, and even organs) [6]. Common features identified among the various definitions submitted by authors are that drones are unpiloted, unmanned, aircrafts or aerial vehicles and remotely controlled [5,6,20,21].

The healthcare supply chain is faced with the challenge of poor distribution networks [22], unavailability of appropriate transport for logistics collection [8], and supply chain complexities [23]. This has resulted in health facilities adopting drone technologies for healthcare deliveries within and across borders [21,24,25]. A study conducted by [10] revealed that medical supplies in Ghana were hindered by the difficulty of transporting medical supplies from the central point to the point-of-care and other remote areas. In the quest to overcome this challenge, the Government of Ghana collaborated with a US-based company Zipline Technology to use drones in the supply of medical supplies for emergencies across the country. The report showed that the introduction of drone technologies for medical supplies in Ghana has significantly impacted positively on the emergency health delivery system. For example, the country of Ghana used drone technologies to respond to the novel COVID-19 pandemic by shuttling medical supplies to health facilities and suspected COVID-19 patients [9]. According to the authors, the deployment of drone technologies saved the lives of the general population.

2.2. The role of the environment in reverse innovation for drone technology use in healthcare

Reverse innovation (RI) has been defined in literature as new ideas and solutions that are accepted and evaluated in low and middle-income countries (e.g., Ghana) before spreading to high-income nations (e.g., Canada). RI has its foundational focus on business development and economics and was originally termed as “innovation blowback” in 2010 [12,26]. Studies have shown that financing, governance, health information systems, leadership, collaborations for novel diagnostics, medicines, vaccinations and health service delivery using drone technologies are all instances of effective reverse innovation [12,27,28]. Studies have shown that, reverse innovations is characterised by commodity performance that must be adapted to an acceptable level at a fraction of the cost of the adopted technology [29]; sustainability, which favours the deployment of technologies such as drone use in the delivery of healthcare [7]; removal or flexibility in legal and regulatory barriers that impede the implementation of new technologies and prevent rapid market access [17]; and meeting local geographical and environmental needs like the delivery of medicine to inaccessible populations during rainy seasons using drones [30].

The current economic disparity between developing markets and developed countries drives reverse innovation, which gives specialized answers to challenges that have not previously been handled in an affordable or culturally acceptable manner [28]. Unlike developed countries with strict regulatory barriers on the implementation of new technologies and aviation, developing countries have flexible policies and regulations that make it possible to implement drone technologies in the healthcare sector [16]. An instance of reverse innovation in a developing country is the deployment of drone technologies to transport health supplies with a large coverage in Ghana [10]. This was made possible because existing aviation laws and regulations on drones in Ghana are very flexible, making flying and delivery by drones very easy [9], compared to the advanced countries like the United States of America, which have strict laws and regulations on aviation. Such countries have to emulate strategies in the implementation of drone technologies in developing countries within the healthcare supply chain, making the theory of reverse innovation possible [16]. Experiences and lessons from reverse innovation in global health show that collaborations between developing countries and developed countries can result in system-wide advantages [30].

2.3. Actual and continuance usage of drones in healthcare

The actual and continuance usage of drone technologies explains the decision of users to either use and continue using drone technologies in healthcare delivery. Empirical evidences show that, the adoption and actual usage of drone technologies hinges on user satisfaction, performance [29], sustainability [7], flexibility in regulations and the implementation of new technologies [17,31], and meeting environmental needs [30]. This explains that the unavailability of such flexibilities and abilities in a country may impose difficulties in drone technologies adoption and usage which explains why developed countries are unable to easily implement drone technologies in the healthcare supply chain [9]. It can therefore be inferred that, the adoption and continual usage of drone technologies in the healthcare supply chain of developing countries like Ghana (transports health supplies) and Rwanda (transports blood) is as a result of the aforementioned characteristics. A study conducted by [32] in examining information systems continuance usage revealed that, actual usage of information systems or technologies plays a mediating role in determining continuance usage behaviour. This implies that, until the consumers actually consume or are involved in the drone experience either by themselves, families or relatives, they will still hold on to their prior usage expectations [33]. And this makes it necessary to examine the conditions for the adoptions and continual usage of drone technologies in healthcare delivery leveraging recent developments in developing countries (reverse innovation).

2.4. Theoretical framework

The study is grounded in the expectation and confirmation theory (ECT) and the technology organization environment theory (TOE). The ECT posits that satisfaction is directly influenced by disconfirmation of beliefs and perceived performance, and is indirectly influenced by both expectations and perceived performance by means of a mediational relationship which passes through the disconfirmation construct. Specifically, users compare their pre-purchased expectation and the post-purchase perceived performance as a means in determining satisfaction (confirmation) of a product or service, which will lead to repurchase intentions [30,34]. The theory has four (4) main constructs namely; expectations, performance, disconfirmation and satisfaction. Expectations serve as the anchor for comparison (anticipations), that is what clients will use to evaluate performance from a disconfirmation judgment [35,36]. Performance is users’ evaluation after purchase or usage of a service (e.g., perception of product quality). If a product meets or outperforms expectations (confirmation), post-purchase satisfaction will result. If a product falls short of expectations (disconfirmation), the consumer is likely to be dissatisfied [33]. ECT has a limitation in explaining the information system expectation formation process [37]. For example, ECT primarily focuses on the individual level of analysis and may overlook the social and contextual factors that influence expectations. In the context of information systems, expectations are often shaped by factors such as social influence, cultural norms, organizational context, and technological advancements. It does not adequately capture these external influences on expectation formation [38]. In addition, ECT is centered around consumers repurchasing intentions of goods and services, handling of the assumption and attitudes toward the goods and services attributes or performance [39]. It does not capture the information systems quality factors (repair quality, information quality and system quality) and subsequently does not explain the thought of end-user satisfaction of information system frameworks [40]. We draw on the ECT to argue that users’ satisfaction with drone technologies in healthcare delivery drives its usage leading to their intention for continuous use.

TOE is a theoretical framework that explains technology adoption in organizations and describes how the process of adopting and implementing technological innovations are influenced by the technological—all of the technologies that are relevant to the firm either in use or available on the market [41]; organizational- the characteristics and resources of the firm and the amount of slack resources [42]; and environmental contexts- structure of the industry, the presence or absence of technology service providers, and the regulatory environment [3,31]. In the area of healthcare, there is an indication of reverse innovation where the environmental factors contribute immensely to the success story of its applications in developing countries [43] and laws and regulations on aviation are not very strict. So, the continuance usage intentions will be higher than those in the developed countries where the environmental factors are not favorable for the use of drones within the healthcare supply chain [44]. We argue that the adoption and continuous usage of drone technologies are influenced by the environment and that there is an indication of reverse innovation where the environment contributes immensely to drone technologies applications in healthcare delivery in developing countries.

2.5. Research model and hypothesis

2.5.1. Relationship between technology adoption, confirmation of expectations and user satisfaction.

Empirical evidence shows that firm adoption of technologies is influenced by their perception of usefulness, technical and organizational compatibility, complexity and learning curve, pilot test/experimentation, and visibility/imagination [45,46]. We have presented the conceptual framework in Fig 1 and validated model in Fig 2. [47] asserts that customer satisfaction is determined by the ease with which they can utilize the technological service. The outcome may be that ease of use is a critical component of customer satisfaction with technological services, as evidenced in the study of [47]. An easy-to-use system will encourage higher system utilization, and users will be more inclined to embrace it if it is perceived as such. [48] observed that, in order to sustain connectivity and communication between smart city services (SCS), it is necessary to promote the adoption and usage of SCS technology. Users must be able to see the benefits of accessing and completing SCS as an improvement in their overall quality of lives [49]. The perceived benefits include the perceived comfort, usefulness, convenience, and safety connected with the delivery of SCS [50]. Increased customer satisfaction is associated with lower price sensitivity and greater reputation efficacy [33]. The complexity of smart services is linked to the discomfort experienced by users who utilize SCS technology. Users’ dissatisfaction is likely to have a negative impact on their capacity to use the channels, as well as on their perceptions of SCS in general [51,52]. [53] opined that the experience of consumers with technology application is a major element in measuring consumer satisfaction with products and services.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Conceptual Framework.

https://doi.org/10.1371/journal.pdig.0000266.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Validated structural model.

https://doi.org/10.1371/journal.pdig.0000266.g002

The term "confirmation of expectations" was used by [54] in the confirmation of expectations theory to characterize the confirmation or disconfirmation of a person’s pre-judgments (initial expectations) about a product or service [37]. According to [46], user satisfaction is determined by two elements: confirmation of expectations after actual usage and post-adoption expectations. Users’ satisfaction may be measured against their expectations before they begin using an expectation or service [54]. Users’ confirmation of expectations means that users received the expected value from their information technology (IT) usage experiences, which has a positive impact on users’ satisfaction with their initial IT use [55]. Following the expectations confirmation model, the degree to which information system users confirm their expectations has a positive impact on their perceived usefulness and satisfaction with information systems (IS) [56]. After purchasing a product or service that meets or exceeds a person’s initial expectations, the confirmation is positive, resulting in an increase in product sales, which is referred to as post-adoption satisfaction. In every other case, the confirmation is negative, which lowers post-adoption satisfaction [57]. Following the arguments put forward on the relationship between technology adoption, confirmation of expectations and user satisfaction, the study hypothesize:

1. H1a: There is a positive relationship between technology adoption and user satisfaction.
2. H1b: There is a positive relationship between confirmation of expectation and user satisfaction.

3.1. Measures

The study explores how the environment influences the adoption, usage and continuance usage of drone technologies in the healthcare industry. The study items were sourced from extant literature. A number of considerations were made in the generation of the scale items in this step, including the extensive review of related literature, availability of existing measurement scale items, and how well the chosen item suits the hypotheses outlined in the previous chapters. These scale measurement items were carefully scrutinized to eliminate inappropriate items or items that do not measure the measurable construct in the hypothesis correctly. The initial measuring items were emailed to health professionals, drone service providers and information systems academicians for review. In total, seven (7) measuring items were used to realize the study objectives–technology adoption (measured with perceived barriers, perceived usefulness, perceived benefits, complexities and compatibility), user satisfaction (measured with prior usage and information system expectation), confirmation of expectation (measured with user expectations, perceived information system usage and perceived ease of use), actual usage, continuance usage, environment (measured with government regulations, trade partners and access to resources) and organization (measured with just-in-time delivery, creativity, efficiency, innovation and leadership). To facilitate the use of statistical analyses, a seven-point Likert scale with 7 representing “strongly agree” and 1 representing “strongly disagree” was used to measure the indicators. Demographic level data on firm characteristics were also collected and examined.

3.2. Sample and data collection

4.1. Demographic results

The study revealed that 232 (74.40%) of the respondents were males, and 80 (25.60%) of the respondents were females indicating that drone technology services have been deployed mostly in male dominated health facilities. This could be because there are less women with technological capabilities in Ghana. Also, 34 (10.90%) of the respondents were 20 years or less, 55 (17.60%) of the respondents were between the ages of 21 and 30 years, 185 (59.30%) of the respondents were between 31 and 39 years, 11 (3.50%) of the respondents were between the ages of 41 and 50, 2 (0.60%) respondents were between the ages of 51 to 60, and 25 (8.00%) of the respondents were over 61 years of age. Regarding educational level, 62 (19.50%) of the respondents were high school graduates, 126 (40.40%) bachelor degree holders, 56 (17.90%) professional certificate holders, 15 (4.80%) master’s degree holders and 11 (3.50%) doctorate degree holders. Most of the respondents serve in private health facilities 212 (67.9%) indicating that the deployment of drone technology services for health delivery in Ghana is mostly seen in the private sectors. Respondents from public health facilities stood at 100 (32.10%). With respect to individual participants personal experience with drone, we found that 240 (76.90%) of the respondents have not had a direct experience with drone deliveries while 72 (23.10%) have had experience with drone deliveries. This indicates that while health facilities leverage drone delivery services, majority of the staff have not directly experienced drone deliveries. This may be because, the health facilities have trained staff who pickup drone deliveries. The full demographic data is presented in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic statistics.

https://doi.org/10.1371/journal.pdig.0000266.t001

4.2. Assessment of measures

Widely, the variance and covariance-based structural equation modelling techniques are used for model estimation and assessment. In this study, the variance-based approach (VB-SEM) was adopted for the model estimation and assessment leveraging SmartPLS—a partial least square structural equation modelling (PLS-SEM) software. Studies have shown that PLS-SEM has a high statistical power [71,72]. This implies that PLS-SEM is more likely to identify relationships as significant when they are indeed present in the population. This justifies the authors choice of variance-based approach with SmartPLS. Internal consistency tests (Cronbach’s alpha and Composite Reliability), convergent validity tests (Average Variance Extracted), and discriminant validity tests (Fornell and Larcker criterion and Cross loading and Heterotrait-Monotrait Ratio) were adopted to assess the validity of the constructs. A summary of the assessment results is presented in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Measures assessment results.

https://doi.org/10.1371/journal.pdig.0000266.t002

Convergent validity test was carried out to measure the degree to which each measure correlates favourably with an alternate measure of the same construct. The test was carried out by accessing the resulting values of the Average Variance Extracted (AVE). The results showed that all the measures had their AVE statistics higher than the recommended threshold of .50 [73] indicating that all the measures are adequately valid.

Further, the discriminant validity test was carried out to examine the extent to which each construct is truly distinct from the other constructs. Fornell and Larcker criterion, cross loading and Heterotrait-Monotrait Ratio (HTMT) were used to assess the discriminant validity. According to [73] criterion, discriminant validity is ascertained when the square root of AVE for a construct is greater than its correlation with all other constructs. It was shown that the square root or leading diagonal elements were all greater than the corresponding constructs estimates (correlation coefficient of each latent variable), providing a strong support for the establishment of discriminant validity. Also, all the item factor loadings are heavier on the underlying construct to which they belong rather than the different construct in the study [74], providing further evidence for discriminant validity using the cross-loading statistics. Heterotrait-Monotrait Ratio (HTMT) for the constructs were also less than the acceptable threshold of 0.90 [75]. Finally, an internal consistency test was carried out to examine how the collection of indicators accurately measured the latent variables. Cronbach’s alpha and composite reliability statistics were used for the investigation. The examination of the data revealed that all the constructs had a Cronbach alpha higher than the recommended threshold, 0.7 [72] except technology complexity, (that reported alpha value 0.689). However, all the constructs including technology complexity had composite reliability test values exceeding the recommended threshold 0.7 [72]. Hence, construct reliability is established.

4.3 Findings

The previous section established the validity of constructs indicating that they are reliable enough to generate accurate insights for the study. This section examines the relationship between the validated constructs following the proposed hypothesis. The results of the structural model are presented in Table 3 below:

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Results of structural model.

https://doi.org/10.1371/journal.pdig.0000266.t003

An assessment was carried out to validate the proposed hypothesis (H1a = There is a positive relationship between technology adoption and user satisfaction; H1b = There is a positive relationship between confirmation of expectation and user satisfaction; H2 = There is a direct relationship between user satisfaction and actual usage. H3 = There is a positive relationship between actual usage and continuance usage; H4 = There is a direct relationship between organization and continuance usage; H5a = There is a positive relationship between environment and user satisfaction; H5b = There is a positive relationship between environment and actual usage; H5c = There is a positive relationship between environment and continuous usage; and H6 = Environment moderates the relationship between actual usage and continuous usage).

The findings of the assessment as shown in Table 3 above revealed that there is a positive relationship between technology adoption and user satisfaction (H1a: β = 0.425, p = 0.000, supported). Confirmation of expectation was found to positively relate with user satisfaction (H1b: β = 0.288, p = 0.000, supported). User satisfaction was found to positively affect actual usage (H2: β = 0.412, p = 0.000, supported) which eventually leads to continuous usage (H3: β = 0.340, p = 0.000, supported). Organisational variables like just-in-time delivery, creativity, efficiency, innovations, and leadership were found to have a direct relationship with information systems continuance usage (H4: β = 0.507, p = 0.000, supported). It was also revealed that there is a direct relationship between the organisation’s operational environment (government regulation, trading partners and, access to resources) and user satisfaction (H5a: β = 0.174, p = 0.018, supported); and actual usage (H5b: β = 0.412, p = 0.000, supported). However, the hypothesis for the relationship between the environment and continous usage was not supported (H5c: β = -0.033, p = 0.731, not supported). Finally, the relationship between actual usage and continuous usage was found to be moderated by the environment (H6: β = -0.054, p = 0.020, supported).

5.1. Relationship between technology, confirmation and user satisfaction

The study showed that there is a significant and positive relationship between technology and user satisfaction (H1a). Again, there is a positive relationship between confirmation and user satisfaction (H1b) which showed consistency with empirical literature and provided further confirmation to the research theories. First, studies have revealed a positive relationship between users’ confirmation of their perceived usefulness and satisfaction [76,77]. This implies that if users believe that an IS is very useful and the actual use experience corresponds or goes beyond their initial expectation, the confirmation that exists leads to user satisfaction [78,79]. This is because the expected benefits of IS use are realized [80]. With regard to technology, studies have highlighted that the complexity of smart services is related to users’ discomfort with smart technologies. Thus, when users’ feeling of discomfort with technology is high, it is likely to have a negative impact on their ability to use smart services, and as such, will have negative attitudes towards its adoption [81]. As a result, the success of smart services should not focus only on the quality of the actual service delivered to the users but also on the complexity or simplicity of the software that drives the smart systems interface hardware. This may also be extended in the adoption of drone technology usage and user satisfaction. This implies that a high level of functionality of drone technology will result in improved levels of user satisfaction. In order to eliminate or, minimize the complexity of smart service delivery channels, the interface technologies should be efficient and personalize it in order to give the users sufficient choices and information to make real-time decisions and accomplish their service needs easily and timely without being confounded by the smart service delivery technologies [82]. In relation to the confirmation of expectations, it is key that smart service delivery providers like drone delivery technologies involve and interact with users through the whole cycle of service development to minimize user challenges [48]. Expectation before initial usage provides a baseline for users to evaluate their satisfaction. Users’ confirmation of expectations suggests that the users obtained expected value through their usage experiences with the IS and thus leads to a positive effect on users’ satisfaction with their initial IS use [55].

5.2. Relationship between user satisfaction and actual usage

The study showed that there is a positive relationship between user satisfaction and actual usage (H2) as hypothesized. This finding is consistent with empirical literature and provided further confirmation to the research theories. [83] discusses a lower bound to satisfaction below which users may discontinue system use, which gives some direct insight into how user satisfaction influences system usage. Using path analysis, [84] argued that user information satisfaction (UIS) could lead to system usage. In general, perceived user-friendliness or ease-of-use of technological innovations offers the strongest explanation of the impacts of such technologies on users’ satisfaction and behavioral intention to use such innovations [48]. According to the expectation-confirmation theory, consumer’s satisfaction refers to their cognitive and affective state of fulfillment after a purchase. It is a key determinant of repurchase intention [55]. As such, by extending the interpretations to the findings of this study, by achieving higher levels of user satisfaction, there will be an enhancement in the actual usage and drone technology for the delivery of health services in Ghana.

5.3. Relationship between actual usage and continuance usage

The study revealed that there is a positive relationship between actual usage and continuance usage (H3) which is consistent with empirical literature and provided further confirmation to the research theories. Users’ experience of technology application is a major criterion for the evaluation of customer’s satisfaction with the services [85]. As such, improved technological innovation and functionality increase user satisfaction. Studies have shown that user information satisfaction (UIS) could lead to system usage [84]. This argument has been extended by [86] and posits that satisfaction is an intrinsic motivator in using information systems. Thus, the actual usage of systems is promoted. Not much had, however, been done to support the position of actual usage supporting continuance usage. We argue from the factors which promote actual usage and explain that through an actual interaction and experience with technology, users may directly experience its benefits, confirm its usefulness and further clear any perceptions about barriers and constraints leading to the continuance usage of the technology.

5.4. Relationship between organization and continuance usage

The study revealed that there is a significant and positive relationship between organization and continuance usage (H4). This indicates that the organizational variables just-in-time delivery, creativity, efficiency, innovations, and leadership [51] influences the continuance usage of drone technologies in healthcare delivery. In terms of organizational issues, [87] opine that critical consideration be given when implementing large and complex information systems and technology, particularly in regard to top management support and facilitating conditions for the information systems and technology development within medical environments. Extending the findings in the adoption of drone technology, organizational factors will rather determine the continuance usage of drone technology.

5.5. Relationship between environment, user satisfaction, actual and continuance usage

The study showed that there is a significant and positive relationship between environment and user satisfaction (H5a). In addition, there is a significant and positive relationship between environment and actual usage (H5b) however, the relationship between environment and continuance usage (H5c) was not significant. As suggested by [88], environmental issues represent the current operating environment of the healthcare industry, and it is worth mentioning that government policies should be particularly considered when implementing the new information systems and technologies within medical environments. In the context of environmental factors, particularly in government regulations, [88] confirms that government policy has a significant impact on hospitals’ information systems adoption decision. If the government keeps its promise on developing the information systems, this will cause users to have a positive feeling that the IS/IT will be useful. However, the continuance usage may not necessarily be impacted by these factors. It can therefore be said that environmental factors contribute to actual usage of drone technology service. External environmental factors like government policy, trade partners, access to resources may have already been contributory to the actual usage. Thus, a zero or reduced levels of such factors will probably lead to continuance usage, as other internal factors of the service will be more significant towards continuance usage.

5.6. Moderating role of environment in the relationship between actual and continuance usage

It was revealed that the environment moderates the relationship between actual usage and continuance usage (H6). From the results, this relationship was found to be significant indicating that the relationship between actual and continual usage of drone services in healthcare delivery is contigent upon environmental factors such as access to resources [86,89,90]. However, this interaction was weak and negative suggesting that reducing the influence of environmental factors will rather increase the impact actual usage have on the continuance usage of drone technology in the delivery of healthcare services. Since no such study is attributed to this finding, no connection can be made with previous studies. Nonetheless, since earlier findings confirmed the limited interaction of environmental factors and continuance usage, the weakness of the moderating effect can provide further confirmation to the limited effect environmental factors have on the continuance usage of drone technology services in the healthcare delivery services.