Determinants of blockchain technology application in primary healthcare delivery: An integrated best-worst approach

Abstract

Primary healthcare (PHC) has become a concern in most resource constraint economies particularly in the global south where meeting this important economic need has become a burden. Given this, blockchain technology (BT), a promising transformation in contemporary service delivery, has become an alternative for service-oriented institutions to meet their desired goals. In primary healthcare, BT has been utilised to lessen the burden on medical supply chain and data management. Nonetheless, the technology seems to be a preserve of the developed economies. In the global south, not only has the complexities of the technology hindered the implementation of the technology but the understanding of its affordances has also been nascent. This study extends Task-technology fit model with the perceived e-readiness model. Drawing on the best-worst method, this paper investigates PHC managers’ decision to embrace BT for PHC delivery in Ghana. The study finds that, in order of relevance, task-technology, infrastructure and individual characteristics are the main drivers of BT adoption and implementation. The study proposes the implementation of various strategies relevant to ensuring a sustainable BT for the management of PHC in resource constraint economies.

Keywords:

• blockchain
• best-worst
• primary healthcare
• MCDM
• TTF
• PERM

PUBLIC INTEREST STATEMENT

The emergence of blockchain technology (BT) has revolutionized the digital transformation of healthcare delivery. It has the propensity to mitigate the prolonged resource constraints in the healthcare ecosystem, particularly in primary healthcare delivery, which is the first point of call for the majority of healthcare seekers in Ghana. This study explored the determinants of BT and how it could be applied in Ghana’s PHC delivery to mitigate resource constraints. Based on the 24 sub-attributes segmented into 5 main dimensions, it emerged that technological factors were the most relevant determinant. Specifically, BT technical knowledge was the most important technological task that could drive blockchain application in PHC delivery in Ghana. Strategies such as awareness creation, standardization and reducing the complexity of the blockchain are needed to mitigate the implementation challenges. The implication of this study is far-reaching given the need to meet the rising healthcare demands amidst acute resource allocation in Ghana

1. Introduction

Emerging trend in the era of industry 4.0 technologies is spurring innovations through ubiquitous interconnectivity and access to real-time data. This revolution has emerged due to heightened demand for cost-effective and resilient technologies to meet the demand of individuals in varied economic sectors (Hatzivasilis et al., 2021; Tortorella et al., 2020). An area which has gained momentum among scholars and practitioners in the past decades is blockchain (Agbo et al., 2019; Omar et al., 2021). The blockchain technology (BT) has enabled the potential to revolutionise how digital transactions are stored, retrieved and shared while enabling new business models. The decentralised nature of blockchain tends to provide users with a behemoth of benefits such as anonymity, transparency, traceability and transaction efficiency (Garg et al., 2021). Furthermore, BT provides higher security and privacy than what traditional database assumes since distributed ledgers are encrypted and eliminate intermediaries in its instantaneous transaction processes (Onik et al., 2019). BT application is gaining prominence in varied sectors including finance (Chang et al., 2020), supply chain (Kouhizadeh et al., 2021), hospitality (Bumblauskas et al., 2020) and healthcare (Prakash & Das, 2021). In healthcare, blockchain has impaired the dogma of logistical nightmares by transforming the healthcare ecosystem. For instance, BT decentralises the management of electronic medical records with controlled access to users (Krawiec et al., 2020), enables remote patient monitoring, facilitates medical supply chain and smart contracting and secures health insurance claims among others (Agbo et al., 2019; Haleem et al., 2021; Swan, 2015).

As a variant of healthcare, the primary healthcare (PHC) system is by far the most inclusive, equitable, economical and efficient strategy for improving people’s physical and mental health, as well as their social well-being (WHO, 2021). PHC is recognised as the “front door” of health systems, laying the groundwork for the development of critical public health activities (Cueto, 2004). Yet, the provenance of PHC system in most developing countries has co-existed with inadequate investment and inequitable distribution of healthcare and underdeveloped health workforce resulting in poor healthcare delivery (Asamani et al., 2021; Dassah et al., 2018). While expenditure on healthcare among developed countries keeps rising with mixed effects (Dhrifi, 2018), their counterparts in developing economies are bedeviled with rising health expenditures, inefficient supply chain and health data breaches (Owusu et al., 2021). A prior study by Moghadam et al. (2012) enumerated a number of weaknesses besetting PHC across developing economies, organisation and stewardship, financing gaps and poor information systems emanating from the unidirectional information flow among actors in the PHC system, yet proposed need-based strategies for health system-level planning aimed at curtailing the health workforce shortage and resource deficits have suffered substantial shortcomings in their implementation (Asamani et al., 2021). This implies that the recurring inequitable gap in PHC is not only structural but a combination of systemic and innovation gaps.

Given the forgone gaps in PHC systems across the global south, Gaynor et al. (2020) amplified the capabilities of blockchain in mitigating issues of data exchange, contracting and supply chain management in healthcare. Thus, we argue that blockchain serves as a promising technology for managers of PHC in Ghana. For instance, BT facilitates the secure prudent use and transfer of resources in the medical value chain. Additionally, the ledger technology helps manage pharmaceutical supply chain and facilitates the secure transfer of patient medical records. Yet, the intention to use BT among healthcare providers in the global south remains unimpressive. Furthermore, paucity of evidence exists on the drivers of BT for PHC delivery, particularly in resource constraint context, as well as accompanying strategies to overcome implementation challenges. The lack of research is surprising, given that blockchain use for PHC is crucial in strategic decision-making in achieving sustainable development goals on health (Binagwaho & Ghebreyesus, 2019; Szoszkiewicz, 2021). Hence, this study addresses the vacuum by identifying the determinants of BT adoption for PHC delivery and suggesting strategies to mitigate implementation challenges.

Existing studies have highlighted a number of challenges besetting BT integration among developing economies (Amponsah et al., 2022; Papadaki & Karamitsos, 2021). Niyitunga (2022) for instance mentioned that lack of technical expertise and infrastructure in the African healthcare system is a significant challenge to the adoption of blockchain technology. Again, Amponsah et al. (2022) highlight that there is a high cost associated with implementing and maintaining blockchain technology in the healthcare system. This, the authors contend, may be prohibitive for many healthcare facilities in Africa. While data privacy and security remain a key issue among early adopters of BT in healthcare systems, the situation is pervasive in view of the formative nature of the technology’s adoption in developing countries (Biswas & Muthukkumarasamy, 2017). Beyond the above, interoperability and regulatory challenges have beset blockchain integration in most developing economies. The lack of interoperability between different healthcare systems and the inability to share data across different platforms is another factor that demotivates blockchain integration into primary healthcare (Reda et al., 2020). Saxena et al. (2022) submit that, given the formative nature of blockchain integration into healthcare systems in developing economies, there appears to be no specific regulatory mechanism to check its implementations, which poses a challenge to the effective integration of the technology.

This paper seeks to investigate the determinants of BT integration in PHC service delivery in Ghana and also proffer strategies to mitigate intended implementation challenges. Leveraging on a mix of the Task-Technology Fit (TTF) theory and Perceived E-readiness Model (PERM), this paper presents a strategic direction for integrating blockchain within the PHC system. Further, the study evaluates and ranks the relative importance of the BT characteristics and implementation strategies based on the Best-Worst Method (BWM). The BWM is a multi-criteria decision-making (MCDM) approach that assists in determining the best weights of a collection of criteria while relying on decision-makers’ favoured opinions (Rezaei, 2015). When compared to other MCDM approaches such as AHP/ANP, BWM outperforms them in terms of consistency, minimal redundancy, total deviation and conformance (Samanlioglu et al., 2020; Sharma et al., 2021; Sofuoğlu, 2020).

This study’s contribution is in three folds: First, this study is distinct given the fact that past research has focused on BT in secondary healthcare (Agbo et al., 2019; Gaynor et al., 2020), without diving into PHC systems and overcoming strategies associated with its integration. Furthermore, while prior studies have made considerable headway in gathering insights on antecedents and challenges associated with adopting blockchain in healthcare supply chains, concerns associated with its implementation in the global south remain relatively few in existing literature given the fact that the phenomenon is still in its infancy. Moreover, there has been no research on possible new strategies to overcome the potential challenges associated with the integration and implementation of blockchain within the PHC system in Africa. Thus, we argue that the study fills a long-standing gap in literature.

Second, it is also important to note that the goal of this paper is not just to identify the use of blockchain-based applications in PHC. It also contributes to the understanding of the constraints and exigencies and suggests technical approaches, methodologies and concepts to integrate the technology. This will engender a new frontier of research on the overarching BT approaches to foster integration and implementation in developing economies. In addition, this paper introduces many new ideas that have not been studied in the past. As we said earlier, the use of blockchain in healthcare is fast evolving. As a result, there are a lot of new articles on the subject, but they do not cover it well in PHC delivery.

Third, we present a strategy to stimulate a successful integrating blockchain within healthcare supply chains, as well as the appropriate policy to address probable nascent challenges, based on blockchain and PHC literature (Cueto, 2004; Gaynor et al., 2020). At the very core, the study proposes a strategy to accelerate the implementation process for actors in the PHC ecosystem in developing economies aimed at improving the resource deficit and generating innovative incentive strategies for PHC providers. Furthermore, this study responds to recent calls to investigate the factors that influence the processes leading to health institutions' adoption of state-of-the-art technologies to annihilate the consequences of epidemics and pandemics. However, having relative sufficient resources through blockchain capabilities remains a hurdle to be overcome by managers and healthcare professionals of PHC. An indication that a preference elicitation methodology is essential in identifying the relevant factors to drive the technology adoption and implementation. Thus, leveraging the BWM in this study bridges the methodological gap in BT studies in healthcare.

The rest of the paper is structured as follows: Section 1 reviews and discusses the relevant literature on blockchain and PHC. Section 2 describes the modelling framework based on BWM as well as the data collection procedure. Section 3 discusses the results, while Section 4 has the conclusion, implications and recommendations for future research.

2. Theoretical background

2.1. Potentials of blockchain technology in primary health care

An evolving trend is the blockchain application in healthcare management (Clauson et al., 2018; Fusco et al., 2020; Onik et al., 2019; Yaqoob et al., 2021). BT’s applicability in healthcare spans patient medical record management, secure transfer of clinical records, managing pharmaceutical and medical supply chain and genomic code management among others (Haleem et al., 2021). The BT capabilities come with several advantages such as cost reduction, increase information access among stakeholders and expedite healthcare decision-making (Clauson et al., 2018; Fusco et al., 2020; Onik et al., 2019; Yaqoob et al., 2021). Clauson et al. (2018) aversed that a compromised medical supply chain has the tendency to interrupt healthcare delivery with consequences for patients’ safety and health outcomes.

Relative to PHC, Moghadam et al. (2012) in their study proffers that enhancing PHC providers’ authority via efficient contracting can encourage improvements in service quality and productivity. These goals seem unattainable, particularly when extant studies have not adequately articulated them in PHC delivery. However, blockchain smart contract may be utilised to automate auditing operations, enhance medical supply chain management, evaluate product quality and compliance with established regulations (Li, 2019; Nanayakkara, 2019). Blockchain can enable PHC provider to be autonomous in their decision-making breaking the centralised healthcare system existing in most developing countries (D. Liang et al., 2021). This autonomy can be further strengthened by classifying PHC service providers as critical care givers with strong separate legal organisations with proper management and financial autonomy to directly contract with purchasers and respond to incentives (Nanayakkara, 2019).

Studies (e.g., Oleribe et al., 2019; Willems et al., 2021) have documented interventions made to mitigate the recurrent healthcare challenges including centralized healthcare system, information leakages, inefficient medical supply and contracting. Clearly, existing health information management systems have failed to address these issues. Thus, short of providing trust, privacy, security and immutability in the managing activities in the healthcare ecosystems. Kovacs et al. (2014) report that the growing demand for pharmaceutical products particularly in lower-income countries make the drug supply chain vulnerable to substandard and counterfeiting. This according to research is estimated to cost about $200 billion annually (Clark, 2015). Yet, BT has the potential to cure this effect by offering annual savings of above $100 billion in data breach-related expenditures and a reduction in fraud and counterfeit medical supplies (Tendulkar et al., 2020).

Notably, studies on the healthcare supply chain mainly focus on issues of trust, technology-driven, transparency (Dinh et al., 2018; Khatoon, 2020). Research suggests that actors in the PHC ecosystem require stronger regulations to clearly specify the makeup of their competencies, duties and responsibilities (Binagwaho & Ghebreyesus, 2019). Nonetheless, research suggests that BT has the tendency to increase access to healthcare supplies as well as other performance parameters like equity, quality and efficiency in many circumstances (Hewa et al., 2021; Singh et al., 2019) which are key indicators to sustainable PHC delivery. Overall, we argue that assessing the capabilities of BT helps, might significantly, lessen the inequality gap in PHC system while concurrently addressing the need to combat the decades-long public health challenge.

2.2. MCDM for prioritizing and ranking of attributes

To optimise decision-making process, criteria are often prioritized or ranked to help increase the efficiency and reliability of decision-making systems (Wang et al., 2020), particularly in face of multiple attributes given scarce resources (Mardani et al., 2020). This is a common MCDM issue which makes the application of decision-making systems challenging. As a remedy, MCDM approaches are appropriate for assisting in the modelling of these components to assist researchers and practitioners in determining the most important elements among the multi-factors (D. Liang et al., 2021). Extant studies have affirmed the usefulness of MCDM approached in evaluation and selection complex decision-making processes in instances such as pandemic management (Ahmad et al., 2021), site selection for electric vehicle charging station (Ghosh et al., 2021), sustainable supply chain (Kusi-Sarpong et al., 2021), customer satisfaction through online reviews (D. Liang et al., 2021), evaluation of mobile health apps (Wang et al., 2020). Given that this paper also aims at prioritizing determinants and strategies for the implementation of BT, leveraging suitable MCDM is of essence.

Ranking of attributes in a resource constraint situations becomes challenging without determining the weights of the criteria (Parmar & Desai, 2021). Hence, the need to utilise weight evaluation-based MCDM techniques. Varied weight evaluation MCDM techniques exist in literature including Fuzzy Set Theory (FST) (Zadeh, 1965), Analytics Network Process (ANP) and Analytics Hierarchy Process (AHP) (Saaty, 1990, 1996) and the Best-Worst method (BWM) (Rezaei, 2015) have been proffered. The BWM has been proven to be an efficient method for determining weights of criteria as compared with other MCDM techniques (Hashemkhani Zolfani et al., 2019; van de Kaa et al., 2017). van de Kaa et al. (2017), for instance, contend that given the efficacies of other MCDM techniques, they fall short in the multi-attribute decision-making process, particularly during the data elicitation. Eventually, relevant segments of the information are lost which affect the impact on the decision-making process and outcome. In solving real-world problem, particularly pertaining to comparing large number of criteria, the AHP/ANP fall short due to its complexities (Pamučar et al., 2020). Due to its dynamic nature of information elicitation, this amplification of pairwise comparisons makes it challenging for decision-makers to handle (Rezaei, 2015). Using the BWM, optimal weight coefficient values are obtained with only 2n-3 comparisons in criteria pairs. A significant subset of paired comparisons eliminates inconsistencies during criterion comparison. This has an additional impact on obtaining more reliable results (in relation to the AHP), because transitivity relations are less undermined, resulting in greater consistency of the results.

3. Methodology

3.1. Identification of determinants and implementation strategies

In this study, the elicitation of the data was done in stages. First, we employed a Modified-Delphi technique employed by interviewing seven experts in the PHC ecosystem in Ghana. It must be emphasized that the seven experts were sampled purposively. This was to ensure that all relevant actors in the PHC delivery were captured. The modified Delphi technique is useful for obtaining consensus among experts from diverse professions (Dalkey & Helmer, 1963). As presented in Figure 1, prior to that, an extensive literature review was conducted to identify the determinant of BT adoption and implementation challenges in PHC. We leveraged the expertise of experts in the PHC delivery in Ghana to develop a concise checklist of indicators relevant to BT use in PHC delivery. The experts undertook a three-round modified-Delphi method before settling on the fitting determinants and strategies. First and second rounds were used to give a preliminary score on each of the indicators. The final round was used to endorse or reject the indicators deemed appropriate since the first two rounds did not settle on fitting indicators by consensus.

Figure 1. Research methodology.

Consequent to this, the criteria were regrouped into 5 dimensions (main criteria) and 24 items (sub-criteria) as shown in Table 1. The indicators were drawn from the guiding theories; the Task-Technology Fit (TTF) theory (Goodhue & Thompson, 1995) and Perceived E-readiness Model (PERM) (Molla & Licker, 2005a). The TTF, according to Goodhue and Thompson (1995), is useful in explaining how technology can be used to fulfil tasks with the overall aim of improving performance. The theory posits that there is a higher prospect for technology to have a positive influence on an individual’s performance when it is utilized and when the features of that technology and the user’s abilities satisfy the requirements of the task. The PERM, on the hand, was originally conceptualised as an e-commerce adoption model based on perceived organisational and external e-readiness (Molla & Licker, 2005a). The model embraces four main constructs namely organisational, environmental, managerial and innovative capabilities of the technology adoption. In this study, we explicate the key determinants of blockchain application by drawing on five constructs derived from these theories as shown in Table 1. Extant studies have utilised these theories in the determination of technology adoption and e-readiness (Alazab et al., 2021; Broni & Owusu, 2020). We argue that the integrated TTF-PERM provides a more comprehensive view of the criteria that determines the integration and implementation of BT in PHC. In essence, the model captures the task, technology, organisational and management strategies aimed at facilitating PHC institutions’ readiness to adopt and implement blockchain within the healthcare systems.

Table 1. Characteristics of blockchain application in PHC

Dimension and attributes	Item	Description	References
Task-Technology	TTF
Technical knowledge	TTF1	The individual’s level of understanding of BT to be able to adopt and implement.	(Choi et al., 2020; D. Liang et al., 2021; Garg et al., 2021; Haleem et al., 2021; Hassanzadeh-Nazarabadi et al., 2019; Iftikhar et al., 2021; Kouhizadeh et al., 2021; Omar et al., 2021; Rejeb et al., 2022; Yaqoob et al., 2021)
Complexities	TTF2	Efforts required to understand and easily use BT.
Data Privacy/security	TTF3	Securing the privacy/security of users and their digital asset/records in the block system.
System Scalability	TTF4	The quality and ability of the BT network to handle varied transaction throughput in a short amount of time.
System interoperability	TTF5	The BT’s ability to run synchronous and seamless with other health information management system.
Individual	IND
Access control	IND1	Users ability to have unfettered access to the system with minimal but efficient control.	(Agbo et al., 2019; Bumblauskas et al., 2020; Choi et al., 2020; Clauson et al., 2018; D. Liang et al., 2021; Haleem et al., 2021; Iftikhar et al., 2021; Kouhizadeh et al., 2021; Kusi-Sarpong et al., 2021; Rejeb et al., 2022; Yaqoob et al., 2021)
Expertise	IND2	Individual’s level of competence to utilize BT.
Dislocation of employee	IND3	Constant fear among employees of being laid off/redundant because blockchain may take over their jobs.
Commitment	IND4	The need to motivate users to use BT and reward system for continuous and effective utilization of the technology.
Resistance to Change	IND5	Employees inability to confound to the changes in their routine activities using BT.
Self-efficacy	IND6	Entails belief in having abilities or skills to perform tasks related to BT.
Trust	IND7	The believe that the system is capable of facilitating transactions per their expectations.
Environmental	ENV
Competitive pressure	ENV1	Constant demand for enhanced services from citizens or stakeholders.	(Alazab et al., 2021; Choi et al., 2020; Iftikhar et al., 2021; Kotronis et al., 2019; Molla & Licker, 2005b; Onik et al., 2019; Rejeb et al., 2022)
Awareness creation	ENV2	Making citizens aware of BT in medical technology.
Regulation/Standardization	ENV3	Aligning blockchain with standardised frameworks of the medical value chain and practices.
Time criticality	ENV4	The urgency of BT utilization in enhancing medical supply chain.
Organization	ORG		(Chang et al., 2020; Choi et al., 2020; D. Liang et al., 2021; Du et al., 2020; Garg et al., 2021; Gaynor et al., 2020; Haleem et al., 2021; Li, 2019; Molla & Licker, 2005b; Oleribe et al., 2019; Onik et al., 2019; Yaeger et al., 2019)
Facility size	ORG1	The size of the health facility and employee workload.
Ownership	ORG2	Ownership and management of the health facility.
Government/executive buy-in	ORG3	Support from executive/management in the technology implementation process.
Adequate financial resource	ORG4	Sustainable financial support for BT adoption and implementation.
Infrastructure	INF
Facilitating resources	INF1	Available resources and investments to meet the needed BT infrastructural requirements.	(Chang et al., 2020; Choi et al., 2020; D. Liang et al., 2021; Haleem et al., 2021; Kouhizadeh et al., 2021; Molla & Licker, 2005b; Omar et al., 2021; Rejeb et al., 2022; Yaqoob et al., 2021)
Availability of infrastructure	INF2	The availability of the facilitating IT resource to manage the BT.
Robust network infrastructure	INF3	Efficiency and reliability of existing network infrastructure to enhance BT system utilization.
Adequate maintenance	INF4	Provision of adequate resource to maintain the retain the BT systems.
System integration	INF5	The effectiveness of the infrastructure to ensure easy interoperability with other health information management systems.

Besides, the model incorporates other strategies to incentivise managers in healthcare and healthcare institutions in eliminating probable constraints associated with implementing BT for PHC delivery. The multiplicity of the determinants and prioritization of the strategies makes it multi-dimensional and fuzzy. Thus, the need for a reliable multi-criteria decision framework to elicit and prioritise the determining factors necessary for BT implementation. At the end of the evaluation, eight strategies were settled by the experts. Table 2 summarises all the strategies to mitigate any BT implementation challenge.

Table 2. Strategies to overcome BT implementation constraints

Strategies	STR	Description
Adequate training and awareness creation.	STR1	Providing the needed training and awareness creation for all actors in the PHC ecosystem.
Optimal technical and human resource allocation.	STR2	Optimal allocation of technical and human resources to facilitate the technology adoption and implementation.
Build assurance and security of resources.	STR3	Provide the needed security and integrity of digital resources on the system.
Increased financial resources for sustainable system implementation.	STR4	Appropriating financial resources for the management and maintenance of BT infrastructure.
Effective change management and incentive strategy	STR5	Build effective change management and reward strategies to foster commitment and continuous usage.
Government support for the deployment of blockchain in PHC.	STR6	Effective collaboration between facilities and health regulatory agencies for smooth implementation.
Standardisation of blockchain in health care.	STR7	Build policies on blockchain in tandem with standardized health care practice.
Optimize system scalability, complexity and increase data security.	STR8	Incorporate features to make blockchain user friendly while securing users’ digital assets.

3.2. BWM

Proposed by (Rezaei, 2015, 2016), the best-worst method (BWM) has been proven to be one of the efficient MCDM techniques to find weights of attributes. Given the efficiency in modeling multi-criteria decision problems, the BWM has been applied in real-world decision-making (Pamučar et al., 2020). Others have also applied the method in supplier selection (Hashemkhani Zolfani et al., 2019), supply chain (Ecer & Pamucar, 2020; Sharma et al., 2021) and healthcare evaluation (Torkayesh et al., 2021). Kusi-Sarpong et al. (2021) and leveraged the BWM to evaluate the risk associated with the implementation of big data analytics in sustainable supply chains and overcoming strategies. Rejeb et al. (2022) in their recent study, integrated a fuzzy Delphi with the BWM to unearth barriers to blockchain adoption in a circular economy. Twelve experts were selected from varied sectors to rate the blockchain barriers using the BWM. It emerged that lack of knowledge and management, reluctance to change and technological immaturity were the most significant barriers.

3.2.1. Prioritising and ranking factors

A concern in real-world decision-making is the accompanied uncertainty leading to ineffective decisions. A remedy, is to prioritise or rank criteria to optimize outcomes in decision-making. The BWM comes handy and uncomplicated compared to other MCDMs. It offers decision-makers the opportunity to express their preferences relative to their significance, implying that the optimal weights of criteria are determined through a simple pairwise comparison. This is because a relatively small number of experts or datasets are required. BWM uses smaller datasets and expert inputs, which saves time for experts and simplifies computation, resulting in more consistent results. Additionally, the BWM is found to be efficient and consistent over the AHP (Rezaei, 2015). In this study, we utilize the BWM technique because it offers an efficient means of eliminating inconsistencies in the preference experts since optimal weights are obtained. The BWM simplifies pairwise computations and determination of the weights as opposed to other MCDM (Rezaei, 2019). Ghana was used as the context of the study.

Following (Rezaei, 2015, 2016), the BWM for this study is presented as follows:

Step 1: Determination of relevant criteria. Five main criteria with corresponding sub-criteria were used in this study as shown in Table 1.

Step 2: Identifying the best and worst criterion for the main and sub-criteria.

Step 3: Expert conduct a pairwise comparison using a scale of 1 to 9. Where 1: (alternative) $\mathit{i}$ represents is equally important to (criterion) $\mathit{j}$ and $9;\mathit{i}$ is extremely more important than $\mathit{j}.$ The result is presented in a vector:

$A_B=\left(a_{\mathit{B}1},a_{\mathit{B}2},\dots a_{\mathit{Bn}}\right).$

In this case, $a_{\mathit{BJ}}$ represents the Best criterion B over criterion $\mathit{j}$.

Step 4: Each of the experts was asked to elicit pairwise comparison ratings of all the other criteria with the worst (W) criterion, as described above. This, too, will result in a vector.

$A_W={\left({\mathit{a}}_{1\mathit{W}},a_{2\mathit{W}},\dots a_{\mathit{nW}}\right)}^{\mathit{T}}.$

$a_{\mathit{JW}}$ in this case represents the preference of $\mathit{j}$ to the Worst criterion W.

Step 5: The next step is to calculate the optimal weights $(w_1^{\ast },w_2^{\ast },\dots .,w_n^{\ast })$for each of the criteria.

$\left\{\left|w_B-a_{\mathit{BJ}}{w}_j\right|,\left|w_j-a_{\mathit{jW}}{w}_W\right|\right\}.$ Following, the minimax model is determined:

min max $\left\{\left|w_B-a_{\mathit{BJ}}{w}_j\right|,\left|w_j-a_{\mathit{jW}}{w}_W\right|\right\},$

$\mathrm{s}.\mathrm{t}.{\sum }_j{w}_j=1,$

(1) $w_j\ge 0,\mathrm{for}\mathrm{each}\mathrm{criterion}$(1)

Transforming Model (1) into a linear model results:

$\min {\xi }^{\mathrm{L}},$

subject to:

$|w_B-a_{\mathit{BJ}}{w}_j|\le {\mathit{\xi }}^{{\mathrm{L}}^{\ast }},\mathrm{all}\mathit{j},$

$|w_j-a_{\mathit{JW}}{w}_W|\le {\mathit{\xi }}^{{\mathrm{L}}^{\ast }},\mathrm{for}\mathrm{all}\mathit{j},$

$\sum _j{w}_j=1,$

(2) $w_j\ge 0,\mathrm{for}\mathrm{all}\mathit{j}$(2)

Model (2) can be computed to obtain the optimal weights $(w_1^{\ast },w_2^{\ast },\dots .,w_n^{\ast })$ and optimal value ${\mathit{\xi }}^{{\mathrm{L}}^{\ast }}.$ Note that the nearer the consistency $({\mathit{\xi }}^{{\mathrm{L}}^{\ast }})$ value to zero, the better the model (Rezaei, 2016).

3.2.2. Determination of the global weights and ranking strategies

The global weights of each criterion are determined after the factors are clustered by the decision-makers to do the pairwise comparisons. The weights of each criterion are determined by multiplying the local weights of the main and sub-criteria. Experts also determine the weights of the strategies based on the same scale (1–9). The overall score of strategies is computed using Equation (3)(3) $V_i={\sum }_{j=1}^n{w}_j{u}_{\mathit{ij}}$(3) :

(3) $V_i={\sum }_{j=1}^n{w}_j{u}_{\mathit{ij}}$(3)

Where $\mathit{i}$ is the index of the strategy and $u_{\mathit{ij}}$ is the normalised score of the strategy $\mathit{i}$ with respect to the main criterion $\mathit{j}$. To be able to prioritise the strategies based on the main criteria, we employ the following models:

(4) $u_{\mathit{ij}}=\frac{{\mathit{x}}_{\mathit{ij}}}{{\sum }_j{x}_{\mathit{ij}}}\mathrm{for}\mathrm{all}\mathit{j}$(4)

(5) $u_{\mathit{ij}}=\frac{\frac{1}{{\mathit{x}}_{\mathit{ij}}}}{{\sum }_j\frac{1}{{\mathit{x}}_{\mathit{ij}}}}\mathrm{for}\mathrm{all}\mathit{j}$(5)

Note that $x_{\mathit{ij}}$ remains the actual score of the $\mathit{i}$ and $\mathit{j}$.

4. Case description

This study investigates Ghana’s PHC system integration of BT. Ghana implemented its PHC policy in 1978 to provide basic and primary healthcare for at least 80% of the population, effectively tackling disease that cause death and disability. Ghana’s PHC delivery operates a three-tier system mainly designed at the district and municipal assembly levels (Bonenberger et al., 2015). The first tier represents the district-level healthcare comprising district/municipal hospitals which serve a population of about 200,000 people in the catchment area. The district/municipal hospitals often have about 50 to 60-bed capacity and serve as the referral hospitals for curative care, laboratory and surgical operations. It is headed by a district health management team responsible for supervisory and implementation of community health services to the second and third tiers. Tier-2 PHC comprises the sub-district level where healthcare is delivered at health centres manned by medical or physician assistants, midwives and general community health nurses and other technical clinical and public health personnel. Tier-2 PHC serves nearly 20,000 populations mainly for basic curative and preventive medicine. Tier-3 is primarily community-based healthcare services (CHPS) which provide essential community-based health services through the active participation of the community members.

Given the disparity in healthcare in Ghana, the PHC system was implemented to bridge the widening gap between the urban and rural areas (Tion & Kicha, 2020). PHC implementation in Ghana has focused on curative care, where the emphasis is placed on removing the immediate cause(s) of an individual’s signs and symptoms. In spite of the progress made, the PHC system faces several setbacks challenging its successes. The challenges include inadequate health workers and logistics, insufficient clinical infrastructure and training (Tion & Kicha, 2020). The challenges nonetheless can be overcome with the deployment of innovative technology. Blockchain has been identified as a key enabler of healthcare with the prospect of overcoming logistical shortcomings (Yaqoob et al., 2021). The impact of PHC on Ghana’s economy is quite enormous, particularly the financial burden and associated challenges. This therefore requires a distributed system where caregivers can share underutilized medical assets with minimal constraints. PHC is basically patient-centred demanding constant access to clinical resources. This feature, is what blockchain seeks to enhance by providing adequate and quality care through data sharing in real-time to enhance treatment choices (Haleem et al., 2021). The decentralized database helps managers of PHC provide better records for effective decision-making with limited resources at hand. It is important therefore to investigate the relevance of employing blockchain for PHC and suggest strategies to mitigate probable implementation challenges.

4.1. Case application

After the elicitation of the factors and subsequent evaluation by the experts as demonstrated in Figure 1. Based on the integrated TTF-PERM, 24 relevant items (sub-criteria) were identified and segmented under 5 main criteria as shown in Table 1. This was after relevant items identified through an extensive literature review underwent a 3-round of expert review. The five main criteria (characteristics) were task-technology, individual, environmental, organization and infrastructure.

The Modified-Delphi techniques employed by the seven experts to undertake the identification and ranking of the various factors. We purposively sampled seven experts from key units of the health sectors that are primarily involved in primary healthcare delivery in Ghana. Table 3 presents a summary of their demographic information. In the first phase of the ranking of the characteristics using the BWM followed the identification of the criteria (i.e., the blockchain characteristics) and sub-attributes. The experts were tasked to select their preferred main criteria and sub-criteria using a 1–9 scale by selecting the best and worst. In BWM, the scale 1–9 is a linguistic term scale where 1 represents very low and 9 indicating very high. Then, they were tasked to select first, best criteria for others and second, others to the worst in a pairwise comparison as summarised in Table 4. The second phase comprised the ranking of the main and sub-criteria characteristics. The weighted characteristics listed in Table 5 were prioritized based on their individual weights. The local weight is the average of the results of all experts. The global weight was obtained by multiplying each sub-criterion local weight by the main criterion weight.

Table 3. Details of experts

Expert no.	Designation/rank	Work experience (in years)	Highest qualification
1	Senior Medical Officer—Medical	12	Master’s Degree
2	Senior Pharmaceutical Officer—Pharmaceutical	15	Master’s Degree
3	Physician Assistant—Medical	17	Bachelor’s Degree
4	Procurement Officer—Procurement and supply chain	14	Bachelor’s Degree
5	Planning Officer—Planning and Budgeting	9	Master’s Degree
6	Health Information Officer—ICT	7	Bachelor’s Degree
7	Statistics and Research Officer—Research	6	Master’s Degree

Table 4. Consistency ratio of main criteria

	${\mathit{\xi }}^{{\mathrm{L}}^{\ast }}$ expert
	1	2	3	4	5	6	7
Task-Technology	0.0298	0.0119	0.0202	0.0523	0.0656	0.0986	0.1221
Individual	0.0683	0.0462	0.0597	0.0682	0.0327	0.0987	0.0724
Environmental	0.0034	0.0267	0.0082	0.0426	0.0333	0.0326	0.0349
Organisational	0.0609	0.0008	0.0301	0.0985	0.0475	0.0537	0.0383
Infrastructure	0.1317	0.0537	0.0089	0.0422	0.0714	0.0317	0.0574

Table 5. Expert selection of best to other seven criteria

Expert	Best Criteria		TTC					IND			ENV		ORG	INF
1	ENV		8					7			1		6	5
2	TTC		1					6			6		6	6
3	TTC		1					3			5		6	7
4	TTC		1					6			7		8	8
5	TTC		1					5			7		6	6
6	INF		8					7			5		5	1
7	INF		9					7			5		7	1
Experts’ selection of others to Worst for seven respondents
Others to the Worst		Expert 1 IND			Expert 2 IND	Expert 3 ORG			Expert 4 ENV			Expert 5 ORG	Expert 6 ENV	Expert 7 ENV
TTC		6			5	5			9			7	8	8
IND		1			1	7			8			6	7	6
ENV		6			5	7			1			6	1	1
ORG		6			5	1			6			1	5	8
INF		5			5	6			6			7	6	8

Finally, was the evaluation of the relative importance attached to the strategies that will engender the adoption and implementation of blockchain for PHC as aggregated in Table 6. The experts in this instance were tasked to, ranked the aggregated eight items (strategies) using the BWM items against the main criteria on a scale of 1–9. The essence was to help construct a framework set to enhance user acceptance and continuance use of BT in PHC in Ghana. We leverage Equation 4(4) $u_{\mathit{ij}}=\frac{{\mathit{x}}_{\mathit{ij}}}{{\sum }_j{x}_{\mathit{ij}}}\mathrm{for}\mathrm{all}\mathit{j}$(4) to determine the $u_{\mathit{ij}}$ value using the normalized score. The normalized score was determined by summing the average scores of each expert rating divided by the sum of each strategy. To obtain the ranking of the strategies in (Table 7), we multiply the respective weights $u_{\mathit{ij}}$ by the weights of the five main criteria to derive $V_i.$ Table 7 presents a summary of the prioritized implementation strategy relative to each main criterion.

Table 6. Local and global weights of main and sub-criteria

Characteristics (main criteria) Sub-criteria	Sub-criteria local weight	Global weights	Rank
Task-Technology		0.401
TTF1	0.353	0.141	1
TTF2	0.284	0.114	2
TTF3	0.167	0.067	4
TTF4	0.069	0.028	14
TTF5	0.127	0.051	5
Individual		0.164
IND1	0.108	0.018	21
IND2	0.281	0.046	6
IND3	0.073	0.012	24
IND4	0.144	0.024	17
IND5	0.129	0.021	18
IND6	0.105	0.017	22
IND7	0.160	0.026	16
Environmental		0.106
ENV1	0.145	0.015	23
ENV2	0.197	0.021	19
ENV3	0.405	0.043	7
ENV4	0.253	0.027	15
Organization		0.098
ORG1	0.100	0.010	25
ORG2	0.198	0.019	20
ORG3	0.319	0.031	13
ORG4	0.383	0.038	10
Infrastructure		0.231
INF1	0.147	0.034	11
INF2	0.171	0.040	8
INF3	0.367	0.085	3
INF4	0.169	0.039	9
INF5	0.146	0.034	12

Table 7. Ranking of strategies

Strategy	Technological		Individual		Environmental		Organisational		Infrastructure
	${\mathit{V}}_{\mathit{I}}$	Rank	${\mathit{V}}_{\mathit{I}}$	Rank	${\mathit{V}}_{\mathit{I}}$	Rank	${\mathit{V}}_{\mathit{I}}$	Rank	${\mathit{V}}_{\mathit{I}}$	Rank
STR1	0.0923	4	0.1678	2	0.1666	3	0.1137	4	0.1093	4
STR2	0.0510	8	0.1398	3	0.0783	6	0.1628	3	0.1678	2
STR3	0.1821	1	0.0455	8	0.0458	8	0.0484	8	0.1785	1
STR4	0.0774	7	0.0720	7	0.0738	7	0.1736	2	0.1597	3
STR5	0.1809	3	0.0945	4	0.0837	5	0.1839	1	0.0760	6
STR6	0.0848	5	0.0787	6	0.1678	2	0.0820	5	0.0760	7
STR7	0.0815	6	0.0821	5	0.1747	1	0.0817	6	0.0850	5
STR8	0.1853	2	0.1731	1	0.0995	4	0.0704	7	0.0474	8

To ensure the robustness of the model and the reliability of the data collected, we tested the consistency ratio using the BWM. As presented in Table 4, a consistency in the model was achieved, since all the pairwise comparison of the individual main criteria by each expert in each category. This implies that the model was robust and reliable, consistent with Rezaei (2016) who posited that a consistency $({\mathit{\xi }}^{{\mathrm{L}}^{\ast }})$ value nearer to zero indicates a better model.

5. Discussions

5.1. Interpretation of weights and ranking of determining factors

As emerged from the results (Table 6), it can be deduced that among the five main characteristics, task-technology (weighted 0.401), infrastructure (weighted 0.231), individual (weighted 0.164) and environmental (weighted 0.106) appeared to be the most important determinants of BT application for PHC delivery in Ghana. The emergence of task-technology as the dominant BT determinant characteristic was expected given the technicalities surrounding BT is characterized by features such as technical knowledge, complexities and data privacy and security drive users’ continuance usage of BT (Haleem et al., 2021). Second on the priority was infrastructure. The availability of technological infrastructure to kick-start the implementation of blockchain has been identified as a major driver in most jurisdictions (Choi et al., 2020). Given that blockchain is mediated through internet, it is therefore suggestive that the needed network infrastructure and facilitating conditions are provided. BT implementation has failed in most jurisdictions due to the non-availability of robust network infrastructure (Yaqoob et al., 2021). Whereas internet penetration is high in Ghana, poor and limited access prevails particularly in the peri-urban and rural areas where PHC mostly serves (Abusamhadana et al., 2021).

Following is the individual characteristics which define the individual’s willingness to use the technology. Consistent with Alazab et al. (2021), an individual’s decision to use the technology will be contingent on factors such as his/her expertise, commitment and trust. It is instructive to note that once individuals perceive the technology to be complex, particularly when they lack the technical know-how, they become averse to the use of the technology. This is true in blockchain adoption when Chang et al. (2020) for instance cite that individuals are often reluctant to use blockchain due to their inexperiences and lack of trust. Fourth, users’ priority, was environmental characteristics. Blockchain is still evolving and at a slower pace in developing economies. As a result, a conducive environment which seeks to advance its implementation and usage is critical. Since healthcare delivery is highly regulated, actors in the healthcare ecosystem play a critical role in ensuring effective running of healthcare institutions. Conversely, organisational factors were not among the highly prioritized blockchain application determinants. This does not suggest that organisational factors were not relevant. The implication is that, in terms of preference, users did not prioritize organizational factors compared to other factors.

It must be emphasized that the conditional precedence of the BWM refers to the order in which the criteria are considered in the decision-making process. The criteria are evaluated based on their performance over the alternatives. This means that an alternative must meet a minimum level of performance on all criteria before it can be considered for selection. In other words, each of the five main criteria was evaluated independently, and the performance of the alternatives was evaluated based on how well they performed on any one of the other criteria. Essentially, an alternative can be considered for selection if it performs well on at least one of the criteria. The choice in this case is conditioned on the decision-maker’s preferences and the underlying context of the decision problem (i.e., the determinants of BT application for PHC delivery in Ghana). The focal point is the concentration on the most important criteria, and how an alternative can be selected if it performs well on at least one of them.

Relative to the sub-criteria characteristics (Table 6), technical knowledge was ranked the most important determinant of blockchain usage. The understanding of the BT and its underlying processes in Ghana is still formative. This explains why users are averse to BT. In Ghana, blockchain usage in healthcare delivery is practically non-existent although some relative usage is observed in the financial sector relative to cryptocurrencies (Broni & Owusu, 2020). This perhaps explains why the technical complexities of the blockchain emerged as the second prominent characteristics. The novelty and technicalities surrounding blockchain make it complex for users. The third most prominent characteristic was robust network infrastructure. The highly decentralized architecture of blockchain implies that nodes must connect to each other and transfer data in the transaction process. Robust network infrastructure is significant in ensuring that users will be able to connect to the nearest health infrastructure by seeking the shortest route using tokens (Bürer et al., 2019). Fourth, data privacy and security emerged as the next important sub-criteria in ranking. Given that blockchain is a shared ledger for recording transactions, data recorded is often encrypted to ensure optimal security and transparency. Protecting the security and privacy of clinical resources and accompanied data is critical because blockchain requires the collection, storage and use of massive amounts of personally identifiable and encrypted records, much of which may be sensitive and potentially harmful if compromised (Iftikhar et al., 2021). In a public blockchain ledger, for instance, privacy is always an issue and instances where security and privacy breaches occur, the individual or health institution whose records are improperly accessed may face a number integrity issues deficits (Yaqoob et al., 2021). Setting a boundary condition in terms of privacy and security in public ledger architecture is important. Column 4 of Table 5 shows the ranks of all sub-criteria, which is a representation of the magnitude of the global weights.

5.2. Prioritization of blockchain implementation strategies

The result of the evaluated blockchain implementation strategies across the five main characteristics is summarized in Table 7. In effect, this was to proffer alternative solutions to mitigate probable challenges in implementing blockchain. For the “task-technology”, “optimize system scalability, complexity and increase data security (STR8) and build assurance and security of resources (STR3)” emerged as the most important strategies. These strategies reinforce the earlier obvious task-technology characteristics: technical complexities and data privacy and security of blockchain. This implies that users can generally embrace the phenomenon when then BT tasks are demystified by reducing its task complexities while optimising the security of data recorded. Furthermore, ensuring the integrity of digital resources on the system in accordance with established standards will help to overcome task-technology difficulties.

Second, overcoming the “individual” challenges in the implementation of blockchain “optimize system scalability, complexity and increase data security (STR8) and adequate training and awareness creation (STR1)” emerged as the first and second strategies, respectively. This implies that individual users are oblivious to the complexities surrounding blockchain and hence will be demanding continuous training to augment their competencies. Besides, awareness is important to foster an enabling environment and generate buy-in among all stakeholders in the PHC ecosystem. Relative to the “environmental” challenges, “Build policies on blockchain in tandem with standardized health care practice (STR7) and effective collaboration between facilities and health regulative agencies for smooth implementation (STR6)” emerged as the top priorities to overcome blockchain implementation. As noted by (Grillos et al., 2021), Government represents a key stakeholder in the health ecosystem and the allocation of resources and regulations of the PHC lies in the ambient of Government. In effect, effective regulations will help standardise and promote the use of blockchain in the health sector.

Additional strategies identified to mitigate “organisational” challenges were “effective change management and incentive strategy (STR5) and Increased financial resources for sustainable system implementation (STR4)”. Given that implementing blockchain in the PHC delivery will dislocate some employees from their traditional jobs, they become incense and resist any effort management effort to advance the course of the technology implementation. Providing effective change management and incentive strategies will help mitigate the displacement effect and smoothen the adoption and implementation transition. Besides, the system must be regularly maintained for which adequate financial resources are needed to ensure its sustainability.

Finally, overcoming “infrastructure” challenges, “providing the needed security and integrity of digital resources on the system (STR3) and optimal allocation of technical and human resources to facilitate the technology adoption and implementation (STR2)” emerged as the first and second implementation strategies. As discussed earlier, securing a public ledger may be challenging for which optimum security for records in the block system is paramount. Even though blockchain has been touted as tamper free, concerns about its security have been pronounced due to flaws (Onik et al., 2019). It is imperative to secure every block in the network to engender trust. The provision of the needed technical and human resources needed to manage the blockchain must be provided. Obviously, the appropriate resources will enhance the optimal application of BT in PHC. It must be emphasised that inappropriate infrastructure and incompetent personnel will advertently increase the risk associated with the blockchain (Iftikhar et al., 2021). Empirically, inappropriate infrastructure coupled with unskilled technical personnel frame a negative perception about technology use affecting system performance and usability.

5.3. Study implications

This research provides several implications to both theory and practice. First, our theoretical contribution lies in the fact that this study is one of the first attempts to examine the determinants of BT integration in PHC delivery and proferring strategies towards overcoming probable challenges. The study further provides a model for prioritising the determinants and strategies to overcome blockchain implementation challenges. Implying, the study provides insights into the drivers of blockchain adoption and respective implementation strategies to mitigate the deficits in PHC infrastructure constraints in developing economies. Through the lens of the BWM, in particular, the study extricated discrete choice limitations in real world scenarios inherent in most accomplished MCDMs like AHP (Rezaei, 2019), which are commonly used. In a resource constraint setting like Ghana, adoption and implementation of BT comes with several trade-offs. Prioritising availed resources thus becomes inevitable. This implies that the BWM is a potent approach to providing reliable results by prioritising the limited number of resources considering the interactions between blockchain drivers and strategies. Leveraging the BWM therefore has several consequences in the identification and selection of resources to enhance the delivery of PHC.

Second, implementing blockchain demands several skill sets and critical technical infrastructure because organisations in Ghana lack the practical knowledge critical to BT implementation. Regrettably, the understanding of BT among healthcare institutions relative to its capabilities in mitigating the persistent resource constraints is limited. A gap thus exists concerning strategies to overcome implementation challenges. Therefore, this study has far-reaching implications for theory particularly for a developing country like Ghana which is overwhelmed with logistical challenges. The implication, consequent to this, is that the health sector of Ghana will be better equipped to implement BT to sustain its resource constraints. Grounded on the TTF and PERM theories, this study proffers that determinants of blockchain readiness relative to technological, individual, organisational, environmental and infrastructural factors have consequences on its implementation. Thus, broadening the theoretical scope to enhance the significance of the study findings to knowledge.

Furthermore, our study reinforces the TTT theory overall aim which focuses on improving the performance of technology adoption while stimulating users towards the prospects of blockchain adoption and implementation in the midst of acute resource constraints in the healthcare delivery. Furthermore, the PERM has helped identify the technical and organisational resources ready to implement BT in PHC delivery. The integrated TTF-PERM offers a means for identification of available resources that could hamper the implementation of the blockchain. This means that PHC institutions with limited resources may find it difficult to implement new approaches that could enhance their efficiency and competitiveness. Blockchain technical characteristics, like complexities, scalability and data privacy or security, play key role in its adoption and continuance usage (Alazab et al., 2021). It is therefore not surprising that technological characteristics emerged as the most important determinant of blockchain adoption and implementation in PHC delivery. In line with the tenets of the PERM, readiness of health institutions to implement blockchain is key, thus implying that the needed resource fit for its implementation must be provided. Given the availed constraints, prioritizing the availed resources becomes significant. This implies that PHC institutions must prioritize their needs in order to benefit from its capabilities.

For practice, the study provided several implications for managers of healthcare institutions in their decision to adopt and implement BT in Ghana’s health sector. The study provides a framework for managers to prioritise their needs, particularly when phased with limited resource challenge requiring trade-off. In effect, the limited resources available will be optimized to meet the requirements for PHC delivery and minimize the acute shortfalls in medical infrastructure. Identifying and addressing the blockchain drivers and associated implementation constraints is a critical task for decision-makers in the sector, and they must be aware to understand these drivers and constraints to ensure effective implementation. By employing the BWM, managers and decision analysts in the health sector can assess and prioritise BT implementation drivers and strategies to overcome accompanied constraints. In effect, the study results show that PHC managers need to know that technological, infrastructural and individual characteristics are the most important, requiring much more attention they seek to apply BT to overcome medical resource constraints. Due to the heterogeneity of BT constraints such as lack of technical expertise and non-standardization of blockchain in the PHC system in Ghana, the choice of appropriate strategy must be carefully considered in order to ensure that the right strategy is selected for execution. It is therefore important for managers of the Ghana health service to have access to this study and its proposed frameworks to make informed decisions about the implementation of BT. This, the study believes, will help Ghana and other developing economies inch towards achieving the SDGs by the year 2030.

6. Conclusions and limitations

Blockchain is one of the emerging technologies set to revolutionize the digital transformation in developing countries, particularly where resources are scarce. Healthcare delivery is not an exception given the fact that the health sector is fast undergoing digital transformation. While developed countries are increasing their effort toward blockchain application in healthcare delivery to, for instance, enhance medical supply chain management, secure patient records and streamline drug traceability, the phenomenon is yet to become popular enough to be used on a large scale in developing countries like Ghana. Amidst this, public institutions that have employed blockchain have faced several difficulties in its applications due to several challenges. This study set out to explore the determinants of BT and it could be applied in Ghana’s PHC delivery. The study hypothesized that employing BT in PHC delivery in Ghana is critical to mitigating the prolonged resource constraints in the healthcare ecosystem. This is in view of the fact that PHC delivery is the first point of call by majority healthcare seekers yet, bedevilled with acute medical resources.

Grounded on the TTF and PERM theories, an extensive cursory review by seven PHC experts identified 24 relevant determinants of BT application and implementation and eight strategies to overcome probable implementation constraints. The 24 attributes were segmented into 5 main dimensions. Using the BWM to evaluate and prioritize the concise checklist BT characteristics and sub-attributes, it emerged that technological factors were most relevant determinant of blockchain application for PHC delivery in Ghana. BT Technical knowledge was the most important technological task which could drive blockchain application in PHC delivery in Ghana. Remedying probable implementation challenge, an array of strategies was offered including awareness creation and standardization to reduce the complexities of the blockchain. The implication of this study is far-reaching given the need for rising healthcare demands amidst acute resource allocation in Ghana.

This study nonetheless had some limitations. The study’s results may not be encompassing given the context and setting. The study is also limited by the number of experts used and their designation. Given that a larger sample could be appropriate, different approaches could be utilized to explicate PHC delivery need for blockchain. Blockchain is still evolving amidst limited understanding of the concept. Hence, future studies could use other MCDM to explore the application in PHC.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work is partially supported by the National Natural Science Foundation of China (No. 72071030), the National Key R&D Program of China (No. 2020YFB1711900), the Planning Fund for the Humanities and Social Sciences of Ministry of Education of China (No. 19YJA630042) and the Social Science Planning Project of the Sichuan Province (No. SC20C007).

Notes on contributors

Sulemana Bankuoru Egala

Sulemana Bankuoru Egala is a Doctoral researcher in data mining and information management at the School of Management and Economics, University of Electronic Science and Technology of China (UESTC), Chengdu, China. He is currently an Assistant Lecturer with the Department of Informatics, Faculty of ICT at the Simon Diedong Dombo University of Business and Integrated Development Studies (SDD-UBIDS), Wa, Upper West, Region, Ghana.

Decui Liang

Decui Liang received the B.S. degree in information management and information system and the Ph.D. degree in management science and engineering from Southwest Jiaotong University, China, in 2008 and 2013, respectively. In 2012, he was a visiting Ph.D. student with the Department of Electrical and Computer Engineering, University of Alberta, Edmonton, AB, Canada.

Adjei Peter Darko

Adjei Peter Darko holds a PhD in Management Science and Engineering with specialization in MCDM analysis and optimization techniques. He is currently a post-doctorial researcher at the Psychology Department of Zhejiang Normal University, Jinhua 321000, China.

Dorcas Boateng

Dorcas Boateng holds a BSc in Computer Science and MPhil in MIS. She is currently pursuing her PhD in Information Systems at the Operations and MIS Department of the University of Ghana Business School. Her career objective is to conduct outstanding academic research, publish in premier academic journals, excel in the classroom and community service.

Haleem Yahaya

Haleem Yahaya is a research fellow with the Faculty of ICT, SD Dombo University of Business and Integrated Development Studies (SDD-UBIDS), Wa, Ghana. His research interest is in digital governance and information management.