The adoption of business intelligence systems in small and medium enterprises in the healthcare sector: A systematic literature review

Abstract

In this era of industrial revolution (IR 4.0), coupled with the emergence of the deadly global pandemic of COVID-19, healthcare organizations are compelled to embrace new technologies for their routine decision-making. The Business Intelligence System (BIS) is one of the emphasized innovations, and due to its potential to provide more intellectual information for decision-making processes, has attracted the interest of industry analysts and policymakers. Literature suggests that the BIS is integrated by organizations in different sectors, but most BIS initiatives struggle to produce the anticipated outcomes. Moreover, the adoption of BIS in SMEs generally and healthcare specifically is rather insignificant. This is due to numerous factors. It is therefore necessary to discover and analyze the essential determinants affecting the adoption of BIS in healthcare SMEs. Therefore, this study tries to tackle this gap by exploring the relevant factors for BIS adoption using a systematic literature review (SLR) and an expert-ranking survey of 63 studies that were published in Scopus and WoS databases from 2011 to 2020. A total of 22 determinants are identified and sent to 15 experts. The data that were gathered from these experts were analyzed using SPSS. The results of the analysis indicated 15 determinants were significant, and one determinant was added by an expert. Consequently, a theoretical structure has been developed based on technology, organization, environment, and CEOs determinants and theories. The results of the current study will deepen the current BIS literature and advance the understanding of the significant elements of BIS adoption decision.

Keywords:

• Systematic literature review
• business intelligence systems
• TOE
• SMEs
• healthcare

PUBLIC INTEREST STATEMENT

One of the major challenges for business executives is effective decision-making. Owing to the need for greater precision and coherence of the decision, the adoption of technological solutions to get information more simply and intuitively is evolving. In this regard, BI is tool that extracts, transforms, and enables data to be traversed to support managers in making a coherent and sound decisions. Similarly, in today’s healthcare market producing hugely increasing data volumes, there are serious pressures on private healthcare organizations to adopt BI solutions. It is believed that sensible and maximum use of business intelligence systems can transform data into desired knowledge that can boost operational efficiency and patient outcomes. Therefore, there are numerous articles identifying the anticipated benefits of BI reinforcing the need for study in this domain. This study conducted systematic literature to identify the most relevant factors influencing the adoption of BI in the healthcare SMEs.

1. Introduction

In the age of IR 4.0, coupled with the emergence of COVID-19 which has affected almost every part of our daily lives, business environments are becoming dynamic. Therefore, businesses need advanced developments in technological innovation for rapid response to the competitive markets (Ahmad & Miskon, 2020a; Hojnik & Ruzzier, 2016). The advent of BIS was propelled by swift technological development and internet proliferation in the mid-90s (Ain et al., 2019; Xia & Gong, 2014). BIS is widely recognized as a comprehensive collection of methods, systems and methodologies that empower organizations to combine and evaluate broad data sets to determine their vulnerabilities, strengths, and possibilities (Chang et al., 2015; Combita Niño et al., 2020; Harrison et al., 2015). As an information system (IS), BIS promotes decision-making through the control, collection and incorporation of unstructured and organized data; the handling of massive databases such as big data; the provision of ad-hoc searches, forecasting, monitoring, and analysis solutions; and the support of advanced computing technologies for new knowledge discovery by end-users (Ishaya & Folarin, 2012; Ain et al., 2019) by processing, summarizing, screening and data convergence from various channels (Veeramisti et al., 2020). Due to hyper-competition and technical developments in Big Data in contemporary trade (Cheng et al., 2020; Zhao et al., 2014), a number of decision-making bodies, including company leaders, chief information officers and chief executive officers (CEOs) have identified the BI technology as among the best technological priority (Arnott et al., 2017; Ain et al., 2019; Yeoh & Popovic, 2015).

BI technologies have gained considerable interest from the industry (Trieu, 2017). This is reflected by a dramatic growth in the global BIS industry, which rose by almost 7.3% in 2017, with sales of up to 18.3 USD billion and is projected to hit 22.8 USD billion by the end of 2020 (Ul-Ain et al., 2019). Despite its extreme significance, wide market growth and increasing investment, the planned outcomes were not achieved by more than 70 percent (Boyton et al., 2015; Puklavec & Oliveira, 2014; Puklavec et al., 2018; Ul-Ain et al., 2019) and previous investigations have shown that businesses are not able to benefit from the actual benefits of the implementation of BIS (Boyton et al., 2015; Liang & Liu, 2018; Ain et al., 2019; Yeoh & Popovic, 2015). Further, so many corporations, particularly in developing economies are yet to employ BIS initiatives because of a paucity of knowledge of the advantages, a dearth of skills and knowledge, and a scarcity of budget. Such enterprises are striving hard to find the right determinants for efficient convergence of their BIS (Liang & Liu, 2018; Sun et al., 2018). Further, academic investigations have flourished in relation to data analytics such as BIS (Sivarajah et al., 2017; Zheng et al., 2014). The tactical, and strategic approaches to the effective establishment and implementation of BIS (Ahmad, Miskon, Alabdan et al., 2020; Zheng et al., 2014) are still being addressed by practitioners and academics in a wide number of published studies. Thus, a broad stream of BIS research has been published to date on review and bibliometric studies on BI and analytics (Grublješič & Jaklič, 2015b; Liang & Liu, 2018). For instance, a comprehensive literature review was established by Hatta et al. (2015) on BIS adoption in SMEs from 2009 to 2015. Further, a two-decade systematic literature review (2000–2019) was recently published by Ain et al. (2019) who comprehensively explored the implementation, usage, and effectiveness of BIS. However, from this viewpoint, the current body of information is incomplete.

The literature shows that BIS adoption decisions are strongly affected by the consideration of the relevant determinants (Magaireah et al., 2019; Yeoh & Popovic, 2015). Yet, there is no systematic analysis available on the adoption/acceptance perspective of the most relevant determinants, theories and models at the level of individual or organization. This SLR therefore attempts to fill the current gaps through the lens of technological, organizational, environmental (TOE), and CEO’s perspectives. This review offers a detailed summary of BIS studies from 2011 to 2020, presenting the recent evolution of BIS-relevant theories/framework/models and critical determinants that contribute to the substantive recommendations. The suggested structure would direct business practitioners to consider what kind of determinants and theories are required to take the highest priority into account in order to maximize the true meaning of BIS. It can also help to enhance the capacity of academics and practitioners to work within the theoretical field adequately and effectively (Alter, 2017). The goal of this paper is to discuss the BIS and explore the factors influencing its adoption in in the healthcare sector As such, the paper answers the following questions: “What are the factors that influence BIS adoption in the Healthcare SMEs sector?” and “What are the most relevant theories/framework/models in literature regarding the significant factors influencing the for the BIS adoption?”

This article introduces the results of BIS adoption in SMEs in the healthcare sector in an attempt to add to the sparse literature on BIS in emerging economies and to compare trends to those in other areas of the world. The article comprises seven sections, deliberating the concepts and tenets that shape the BIS adoption. The first section provides the background of the study and reviews literature generated in developing countries, on BIS adoption and implementation over the past decade. The second section describes the methodology of the study, including an outline of the research methodology used, as well as analysis of the results of the systematic literature review (SLR) used in identifying factors, supported by experts’ evaluation. The third section details the results of the survey conducted to evaluate the factors. The fourth section discusses the construction of a conceptual framework, including a description of the TOE dimensions and CEO’s characteristics and their factors. In the fifth and sixth sections, the theoretical and practical implications are discussed, respectively. The research limitation and possible recommendations for research are discussed in section seven. Finally, section eight provides the conclusion.

2. Methodology

Literature review is a thorough method that serves as the basis for every research which helps to advance science incrementally based on prior findings (Al-emran et al., 2018). SLRs are a method of synthesizing empirical data to address a specific research issue in a straightforward and reproducible manner, while attempting to incorporate all published evidence on the subject and evaluate the validity of this evidence (Lame, 2019). Basically, knowledge advancement must be built on prior work. Therefore, to push the knowledge boundary, we must know where this boundary is. By reviewing important literature, we recognize the width and intensity of the current body of work and pinpoint gaps to explore (Xiao & Watson, 2019). This method aims to uncover references relevant to a subject under review and provides a vital contribution to the research significance.

This systematic review follows Xiao and Watson (2019) guidelines for conducting SLR. The primary explanation for adhering to these guidelines is due to the fact that these guidelines offer evidence-based support for the issue under investigation., and it served as a well-known guide for directing a vast number of systematic reviews (Manz, 2019). This current study conducted an SLR aimed at identifying the important factors that affect BI adoption/acceptance in SMEs in the healthcare sector, and suggests a framework encompassing the relations among the pertinent determinants. An SLR was adopted to accomplish this as it reveals all the pertinent subject matter on a given topic to be examined deeply, as well as letting other unknown concepts to be discovered. Figure 1 shows the methodological steps followed to construct the framework of the study through SLR and experts’ evaluation.

Figure 1. SLR methodological steps followed

The process followed in this SLR is specified in the following sub-sections.

2.1. Data collection for SLR

This study used seven databases as data sources (Emerald, Sage, Elsevier, IEEE, Taylor & Francis, Inderscience, and Springer). These databases are regarded as the major and all-inclusive existing databases of peer-reviewed high impact journals. In the beginning, the study used the following keywords and search terms, both combined and separate, using the Boolean operators “AND” and “OR” and advanced search: “business intelligence system adoption;” “business intelligence system acceptance;” “business intelligence system;” “factors influencing business intelligence system;” “factors affecting business intelligence system;” “business intelligence system in SMEs;” and “business intelligence system in healthcare”.

2.2. Inclusion criteria

The articles, carefully chosen, provide data that is considered substantial enough for inclusion in the review. The amount of BI adoption study has grown considerably owing to the need to make good decisions in the organization. Thus, the review includes only articles published since 2011. To guarantee the quality and impact in the domain of BI, the authors selected only those articles published from journals indexed in Web of Science (WoS) or Scopus. All the articles selected encompass BI studies and/or studies in related systems such as business analytics. Further, this study only included original empirical studies and conceptual frameworks, written/published in the English language. Other types of articles such as meta-analysis and systematic review were also included.

2.3. Extraction of data

The preliminary stage of the analysis included checking for redundant data. Subsequently, the abstracts were appraised based on the inclusion criteria. If the article was still relevant, the methodology and discussion section were read and recapitulated. The application of open coding was used via the Excel and Mendeley software. Figure 2 shows the SLR framework, portraying the selection process for the papers from the databases. In Step 1, the study identified 490 articles (Emerald, 30; Sage, 45; Elsevier, 103; IEEE, 130; Taylor & Francis, 39; Inderscience, 73; and Springer, 80). In Step 2, 65 articles were considered not relevant to this study and thus excluded in the sample. In Step 3, the abstracts of the 425 remaining articles were assessed to detect any paper that should not be included, resulting in the elimination of an additional 191 articles, leaving 234 articles. In Step 4, introduction of each article was checked comprehensively against the inclusion criteria, resulting in the elimination of a further 113 articles; the total of the remaining 121 articles were then assessed, leading to the exclusion of another 58 articles based on quality. In Step 5, the remaining 63 articles that fully met the inclusion criteria were, thus, chosen.

Figure 2. PRISMA flowchart for the selected articles

To confirm that these studies used a coherent description of factors affecting BI adoption, the definitions and items adopted to measure the factors of adoption were reviewed. The pool of articles included in this study matched in terms of research question, aim, frameworks adopted, and findings. The definitions used and the alignment of these definitions to the measurement adopted were appraised to ensure that the BI factors examined by the different researchers were largely analogous. The 63 articles included through the SLR highlighted the relationships of factors influencing BI adoption in many organizations that had a similar context with the current study. In this way, the key variables that have a positive influence on BI adoption were assembled.

Table 1 shows the total of the 63 selected articles from six databases, which were classified based on their quality either from SCOPUS or WoS.

Table 1. Results of the SLR

Database	Authors	No. of papers selected	Quality of paper
Emerald	(Ahmad, 2015; Daradkeh, 2019; El-Adaileh & Foster, 2019; Puklavec et al., 2018; Wang & Byrd, 2017)	5	SCOPUS & WoS
Elsevier	(Bach et al., 2016; Božič & Dimovski, 2019; Côrte-Real et al., 2014; Llave, 2017; Pejić Bach et al., 2019; Ain et al., 2019)	7	SCOPUS & WoS
IEEE	(Adeyelure et al., 2016; Alarmouty & Fraihat, 2019; Anjariny et al., 2016; Apraxine & Stylianou, 2017; Chaveesuk & Horkondee, 2015; Chichti et al., 2016; Chuah, 2010; Concepcion et al., 2019; Daryaei et al., 2013; Elhassan & Klett, 2016; Harb & Alhayajneh, 2019; Jalil et al., 2019; Magaireah et al., 2017; Moyo & Loock, 2017; Shahid et al., 2017; Siemen et al., 2018)	16	SCOPUS & WoS
Inderscience	(Ghaida, 2018; Hou, 2013, Hou, 2014; Jahantigh et al., 2019; Mathew, 2012; Nasab et al., 2017; Nofal & Yusof, 2016; Pool et al., 2018; Rouhani & Mehri, 2016; Yeoh, 2011; T. E. Yoon et al., 2017)	11	SCOPUS & WoS
Sage	(Banerjee & Banerjee, 2017; Bhatiasevi & Naglis, 2020; Han et al., 2014; Hou, 2016)	4
Springer	(Adeyelure et al., 2018a; A. Ahmad & Hossain, 2018; Ahmad & Miskon, 2020b; Aldossari & Mukhtar, 2019; Caserio & Trucco, 2018a, Caserio & Trucco, 2018b; Ferrari et al., 2011; Indriasari et al., 2019; Moyo & Loock, 2019; Shen et al., 2012; Venkatraman et al., 2018; Venter, 2019)	12	SCOPUS & WoS
Taylor & Francis	(Adeyelure et al., 2018b; Boonsiritomachai et al., 2016; Grublješič & Jaklič, 2015a; Olszak, 2016; Passlick et al., 2020; Verkoou & Spruit, 2013; Wang, 2014; Yiu et al., 2020)	8	SCOPUS & WoS

2.4. Factors extraction

The study checked through these 63 identified articles and extracted 22 important factors which influence the BIS adoption including: perceived security, relative advantage, compatibility, system quality, trialability, observability, social influence, top-management support, facilitating conditions, government policy, organizational readiness, organization size, complexity, data quality, vendor support, absorptive capacity, manager’s innovativeness, manager’s IT knowledge, IT infrastructure, organizational resource availability, competitive pressure, and government support.

2.5. Experts’ evaluation

According to Mosweu et al. (2016), the experience and integrity of the experts play a key role in choosing the essential factors that affect technology adoption. Additionally, these experts’ evaluation method has shown encouraging results in previous examinations when applied to identifying factors for the adoption of information systems by different organizations in both developed and developing economies (Boonstra et al., 2014; Gagnon et al., 2012; Gruenhagen & Parker, 2020; Hawash et al., 2020; Mosweu et al., 2016; Mukred et al., 2019; Schneider & Sunyaev, 2016). Therefore, once the preceding was done and the factors that could influence BI adoption were identified, they were sent via e-mail to 15 experts from IT, SMEs and Healthcare fields to assess and endorse the factors and/or suggest the addition or removal of any relevant or irrelevant factors, respectively. A description of each factor was added in the questionnaire for easy understanding and appropriate response. The experts’ evaluation was administered in May 2020 using an online survey. The experts were asked to anonymously appraise the importance of each factor in relation to BI adoption on a 5-point Likert scale (1 = very low importance to 5 = very high importance). A description of each factor was added in the questionnaire for easy understanding and appropriate response. The experts’ evaluation was administered in May 2020. All respondents were experts in the IT, SME, and Healthcare fields with at least 3 years’ experience in their respective discipline. Further, they were all PhD holders, and had published at least one article indexed in either the WoS or Scopus. Consequently, the process resulted in retaining 16 factors which were used to construct the framework of the study (see Figure 2).

3. Results and discussion

In this section, the results of the data gathered from the SLR and the expert evaluation are analysed and deliberated. The data collected from IT, SMEs, and Healthcare experts were analyzed using SPSS and tested using the one-sample t-test (see Table 2). This test is used to compare the mean of the population (X) to the hypothesized value (X_mean) = 4, indicated by the high importance value in the 5-point Likert scale (1 = very low importance to 5 = very high importance). Therefore, the testing value set for the factors is as follows:

1. Included: if the mean of the proposed factor is >4, the factor is considered important i.e., influences the BI adoption decision.

2. Excluded: if the mean of the proposed factor is <4, the factor is considered unimportant, i.e., does not influence the BI adoption decision.

Table 2. Result of the experts’ evaluation

			t	Sig. (2-tailed)	95% Confidence Interval of the Difference
Factors	N	Mean			Lower	Upper	Decision
Perceived security	15	3.933	−0.269	.792	−.599	.466	Excluded
Relative advantage	15	4.733	6.205	.000	.480	.987	Included
Compatibility	15	4.600	4.583	.000	.319	.881	Included
System quality	15	3.867	−0.564	.582	−.640	.374	Excluded
Trialability	15	4.667	5.292	.000	.396	.937	Included
Observability	15	4.467	3.500	.004	.181	.753	Included
Social influence,	15	3.800	−0.899	.384	−.677	.277	Excluded
Top-mgt. support,	15	5.000	3.055	.009	.119	.681	Included
Facilitating conditions,	15	4.400	6.205	.000	.480	.987	Excluded
Government policy	15	4.733	4.583	.000	.319	.881	Included
Organizational readiness	15	4.600	4.583	.000	.319	.881	Excluded
Organization size	15	4.600	2.646	.019	.063	.604	Included
Complexity	15	4.333	4.000	.001	.247	.819	Included
Data quality	15	3.800	4.000	.001	.247	.819	Excluded
Vendor support	15	4.533	3.055	.009	.119	.681	Included
Absorptive capacity	15	4.400	3.500	.004	.181	.753	Included
Manager’s innovativeness,	15	4.467	4.583	.000	.319	.881	Included
Manager’s IT knowledge,	15	4.600	2.646	.019	.063	.604	Included
IT infrastructure,	15	4.333	4.000	.001	.247	.819	Excluded
Organizational resource availability	15	4.800	4.000	.001	.247	.819	Included
Competitive pressure	15	4.533	4.000	.001	.247	.819	Included
Government support	15	4.533	4.000	.001	.247	.819	Included

In Table 2, experts ranked 19 of the 22 factors as important and suggested removing three overlapping factors (facilitating condition, organizational readiness, and technology resources lumped into “organizational resource availability”) and suggested the addition of one new factor as significant in influencing BIS adoption: perceived security, relative advantage, compatibility, system quality, trialability, observability, social influence, top-management support, facilitating conditions, government policy, organizational readiness, organization size, complexity, data quality, vendor support, absorptive capacity, manager’s innovativeness, manager’s IT knowledge, IT infrastructure, manager’s attitude toward IT (added by experts), organizational resource availability, competitive pressure, and government support.

4. Theories, models and frameworks used in BIS studies

According to the articles included in this review, there are a several theories, frameworks, and models used to explore the major determinants of BIS adoption. Consequently, it is discovered that a total of nine theories, frameworks and models were commonly used in BIS adoption studies. Majority of these identified studies have used TOE framework, diffusion of innovation (DOI) theory, or institutional theory. while some which used IS adoption model for small businesses, unified theory of acceptance and use of technology (UTAUT) theory of planned behavior (TPB), resource-based view (RBV) and technology acceptance model (TAM). DOI, institutional theory, RBV and TOE were used for BIS adoption at firm level, while, UTAUT, TPB, IS adoption model for small businesses and TAM were used for BIS adoption at an individual level.

Therefore, based on the review of adoption theories and models, as shown in Figure 2, the 16 identified factors were classified into four dimensions using the technology-organization-environment (TOE) framework (Tornatzky & Fleischer, 1990) and the “IS adoption model for small businesses” (Thong, 1999) as follows: technology (5 factors); organization (5 factors); environment (3 factors), and manager (3 factors). These factors were considered to be important in the adoption of BI in the healthcare sector, as detailed below.

5. Conceptual framework construction

This part provides an explanation of the four classified dimensions and the definition of the constructs required for the construction of the conceptual framework.

5.1. Technology characteristics

A critical review of IT adoption literature discloses that technological contexts of innovation are the main emphasis of numerous IT adoption studies (Oliveira & Martins, 2011). The technological aspect of adoption defines the characteristics of external and internal technologies that might influence organizations or individuals (Khayer, Talukder et al., 2020). According to Tornatzky and Fleischer (1990), the technological context comprises both the external as well as internal technologies that are pertinent to the organizations. The external technologies of the organizations are those that are readily available in the market but are not presently adopted by a given firm; these technologies can affect innovation “by demarcating the limits of what is possible as well as by showing firms ways in which technology can enable them to evolve and adapt” (Baker, 2012, p. 232). The internal technologies of the organizations encompass the current equipment and practices of the organization, which are important in the decision to adopt innovation because they set a broad perimeter on the scope and pace of technological transformation that a firm can accept (Baker, 2012).

Based on the SLR and experts’ evaluation, the factors screened were exactly the Rogers’ five technological factors (Relative advantage, Complexity, Compatibility, Trialability, and Observability - see Table 3 for the definitions). According to Banapour et al. (2020) and Pipitwanichakarn and Wongtada (2019), the factors in Rogers (2003) theory of diffusion of innovation (DOI) are the most commonly studied factors used to examine the impact of technological factors on technology adoption by SMEs. Many studies used these technological characteristics as a criteria for determining the level of IT adoption in a business (Karunagaran et al., 2019; Khayer, Talukder et al., 2020; Ma & Lee, 2019; S. Z. Ahmad et al., 2019). Accordingly, S. Z. Ahmad et al. (2019) employed these attributes to examine social media adoption by SMEs in UAE. Also, Hiran and Henten (2020) used these elements to study the adoption of cloud computing in the Ethiopian higher education sector. Further, C. Yoon et al. (2020) used these same elements to investigate the factors affecting the adoption of the smart farm in Korea. These studies found these attributes important for technology adoption. This is also in line with Teixeira et al. (2018), who conducted a theoretical analysis of 41 articles focused on 45 factors related to organizational digital marketing adoption and found that relative advantage, compatibility and complexity, observability, and trialability are relevant factors associated with innovation adoption. In addition, the results of the meta-analysis of Hameed and Counsell (2014) confirm that these elements are strong factors of IT innovation adoption.

Table 3. Definitions of technological characteristics

Constructs	Description	References
Relative advantage	Is the level to which an idea is considered to be greater than the idea it replaces	Rogers, 2003,
Complexity	Is the level to which an invention is considered as impossible to grasp and use. It is the extent to which new technology such as BI is perceived as relatively intricate to comprehend and use.	(Rogers, 1995).
Compatibility	Compatibility refers to the extent to which an idea is considered to be aligned with values, past perceptions and future implementation needs. Thus, it is the extent to which IT system is in harmony with existing values and experiences of a given organisation.	(Rogers, 1995).
Observability	Is the opportunity to track both innovation and its consequences. It designates how much people can see the influence of technology adoption prior to the adoption.	(McCann et al., 2014)
Trialability	Trialability is the degree to which prospective users have the chance to test an innovation. It is the organizational or individual ability to experiment, use, and practice new technologies or services prior to procuring or utilizing them.	(Rogers, 1995).

This technical dimension gives recommendations for best practice that allow organizations to handle electronic information and data protection over time and the system is implemented by technological changes. Therefore, a large range of variables affecting device implementation, which has many related benefits, form the technology component. If introduced as part of business continuity and as an autonomous method, business analytics is more likely to be embraced as necessary technological and infrastructure support for information processing in the enterprise (An & Wang, 2010; Sittig et al., 2014).

5.2. Organizational characteristics

According to Tornatzky and Fleischer (1990), organizational processes and structure can facilitate and/or inhibit innovation adoption. In the technology adoption context, characteristics of an organization exert a major role in the organizational adoption decision. Organizational characteristics refer to the internal considerations and characteristics of the organization (Clohessy et al., 2019; S. Z. Ahmad et al., 2019). A review of relevant literature indicates several organizational characteristics that may impact technological innovation adoption. These characteristics include all the features of the organization comprising the number of employees, revenue, degree of centralization and formalization, and managerial structure and its resources, including staff and their relationships and networks (Tornatzky & Fleischer, 1990). Given the characteristics of SMEs and based on the results above, the organizational factors that this study will focus on are: presence of champion, absorptive capacity, management support, organizational size, and organizational resource availability (see Table 4 for the definitions).

Table 4. Definitions of Organizational characteristics

Constructs	Description	References
Absorptive capacity	It is a firm’s ability to identify, assimilate and apply the importance of new knowledge. It is the organizational ability to absorb and utilize knowledge from their environment. It acts an interfunctional harmonization role in assimilating and utilizing knowledge from the market, which could profit the organization.	(Alexiou et al., 2019). (Najafi-Tavani et al., 2016).
Organizational resource availability	“Resources are material and non-material capabilities that facilitate and support production and productivity” They are stocks of accessible factors that are possessed or regulated by organization.	(Imran et al., 2019, p. 2)
Top management support	It refers to the degree to which top management recognises the significance of the IS position and the extent to which it is engaged in IS activities	(Ragu-Nathan et al., 2004).
Organizational Size	Organizational size is determined by many parameters such as target market size, number of employees, and capital investment made in an organization. Firm size is one more vital factor that can influence technology adoption. Adopting new innovations encompasses a huge investment that is equivalent to the size of a given organization.	Anand and Kulshreshtha (2007) (Rogers, 2003).
Presence of Champion	The project champion refers to a management-level person who identifies the worth of an idea for the organization, and spearheads authority and resources for such an idea all the way through its development and implementation stage. Champions were influential in inspiring others to support the new innovation and surmounting barriers to its adoption and subsequent execution.	(Urquhart et al., 2019). (Meyer, 2000).

5.3. Environmental characteristics

Based on an analysis of literature on technology innovation, environmental characteristics are generally treated as a vital determining factor of innovation adoption (Damanpour & Schneider, 2006). It is essential to assess the effect of environmental factors prior to adopting a technology as they impact the success of any innovation adoption (Damanpour & Schneider, 2006). The environmental characteristics in the TOE framework integrates the availability of technology service providers, the structure of the industry, and the flexible organizational environment (Awa et al., 2016). The environmental context aids in offering better insight of the power of external environmental pressures on organizational technology adoption (Gutierrez et al., 2015; Taylor, 2019). Environmental characteristics have long been recognized as a motivating force for innovation, as cited in several published innovation studies. Environmental variables include external circumstances under which the company works (S. Z. Ahmad et al., 2019). In general, this study examines three environmental innovation characteristics for BI adoption by small and medium-sized enterprises: vendor support, competitive pressure, and government support (see Table 5 for the definitions).

Table 5. Definitions of Environmental characteristics

Constructs	Description	References
Competitive pressure	As the business environment is ever-changing, competition from other actors in the market upsurges, business organizations are more prone and compelled to pursue ways of realizing viable competitive advantage via innovative technologies. Every organization meticulously monitors the action of other actors to accomplish competitive advantages. New IT adoption is commonly acknowledged as a strategic requirement for organizational survival in today’s great competitive and endlessly ever-changing business environment	(Cruz-Jesus et al., 2019). (Khayer, Jahan et al., 2020).
Vendor Support	The role of the vendor in terms of the provision of support to their customers and its users has been the subject of deliberation for a long period. Vendor support can be offered either by vendors, or by consultants. This support includes training of the users, help during implementation and maintenance, as well as software updates	(Bhatiasevi & Naglis, 2020) (Branco et al., 2019) (Chatzoglou et al., 2017).
Government Support	Government support denotes financial and administrative support given to the organizations by the government in the process of hosting and employing new IT. It also signifies the extent to which organizations are shaped by government actions for stimulating the industry. It is about the availability of relevant government policies and initiatives for stimulating the adoption of technologies by organizations	(C. Yoon et al., 2020) (Park & Kim, 2019) (Chau et al., 2020).

5.4. Owner-manager characteristics

Characteristics of the owner-managers are another force that drives firms to adopt technological innovation. The literature reveals that the process of how potential adopters perceive innovation is one of the main determinants of adoption in many diffusion models (Ghobakhloo & Hong Tang 2013; Nguyen & Waring 2013). The owner-manager is the exclusive decision-maker having a direct influence on decision processes ranging from day-to-day functions to impending investments (Mazzarol & Reboud, 2020). They are individuals who are “frequently found to be working alone, with limited resources and high levels of uncertainty. They are often forced to depend on others, external to their firm, for assistance due to the absence of sufficient resources to bring such capacity ‘in-house’” (2020, p. 140). They are generally the principal shareholder with a position in the management board of the organization (Joe et al., 2019). Several studies on SMEs have advocated that the role of owner-managers is fundamental to the organization as their decisions affect both present and future activities of the organization. Therefore, based on SLR and experts’ evaluation, three constructs under the characteristics of owner-managers are examined in this study: Owner-managers’ innovativeness, attitude toward IT, and IT knowledge (see Table 6 for the definitions).

Table 6. Definitions of Managers’ characteristics

Constructs	Description	References
CEO’s/Owner-managers’ innovativeness	This refers to the degree to which a CEO is prepared to consciously implement innovative strategic strategies and technology to strengthen the organization. It refers to the enthusiasm of the CEO for the introduction of new technologies. Innovative CEOs play a vibrant role in influencing SMEs to adopt technology because they are more likely to generate new thoughts and ideas, assume high risks by employing new technology that may reinvigorate the business structure, move organization to a competitive position and upsurge its efficiency.	(Chau et al., 2020). (Hameed & Counsell, 2012) (Hameed & Counsell, 2012)
Owner-managers’ IT knowledge	It refers to the strength of the IT awareness of managers about new innovative technology, which determines how new innovative technologies could be implemented early/late. It refers to the intensity of managers’ IT knowledge concerning the new innovative technology, which defines how early/late new innovative technologies could be embraced. Managers’ IT knowledge includes experience, skills, as well as training concerning IT usage.	(Alrousan et al., 2020) (Chau et al., 2020) (Kannabiran & Dharmalingam, 2012).
CEOs/Owner-managers’ Attitude toward IT	Attitude toward IT refers to the extent to which a positive or negative personal assessment of IT is hold by the individual. A lack of positive perceptions of worth in IT adoption, slanted toward negative perceptions concerning costs and a shortage of resources, lead to a failure to realize the strategic fit of these new IT applications with their business model. Organizations with affirmative attitudes toward IT adopt and use IT more efficiently.	(Jones et al., 2014).

5.5. Proposed conceptual framework

The aim of a conceptual framework is to identify and define the related concepts and the relationships between them (Aziz et al., 2018). A methodological structure that can be used to help and enhance the implementation of BIS is proposed in this paper. It is relevant to the principles, scientific analysis, and substantial hypotheses used in this study to discover and systemize the information introduced (Morioka & Carvalho, 2016; Ngulube, 2018). The conceptual structure thus describes the factors that affect the adoption of BIS to facilitate organizational continuity, such as technology, organization, environmental, and CEO factors, and reflects on promoting the awareness and desire of users to adopt BIS. The proposed conceptual structure contributes to the exploration of factors which can be used to determine the implementation of BIS in healthcare SMEs.

To this end, this analysis integrates “Diffusion of Innovation” theory (Rogers, 2003), “technology–organisation—environment (TOE) framework” (Tornatzky & Fleischer, 1990) and the “IS adoption model for small businesses” (Thong, 1999). with a range of factors chosen in one framework by the experts to better understand the problems impacting users in the healthcare SMEs’ field with regard to the adoption of this new system (see Figure 2). This framework demonstrates the influences of the five technological factors (relative advantage, complexity, compatibility, trialability, and observability); five organizational factors (presence of champion, absorptive capacity, management support, organizational size, and organizational resource availability); three environmental factors (vendor support, competitive pressure, and government support); and CEO factors (owner-managers’ innovativeness, attitude toward IT, and IT knowledge) on healthcare SMEs in adopting BIS. To describe the adoption outcomes in organizations, these variables can be integrated and they can be classified into the framework of TOE and IS adoption model for small businesses. But, for classifying variables, the frameworks are taxonomies, and it is not a reflection of an interconnected philosophical framework or a well-developed theory. The variables can vary from one context to the next within the frameworks, and thus for enrichment, certain other variables have to be integrated into it. This is the justification for using TOE and IS adoption model for small business as the foundation of this paper.

In Figure 3, the suggested explanatory structure consists of four characteristics: technology characteristics, organizational characteristics, environment characteristics, and CEO’s characteristics. The goal of this framework is to establish a conceptual model of BIS acceptance with a succinct forecast and a simple interpretation of the key constructs and determinants. The framework is built with the combination of most possible models and theories such as the TOE, DOI, and IS framework for small businesses. Previously published literature has verified the regular convergence of the TOE paradigm with other frameworks to investigate the acceptance of innovations. Initially, the TOE did not have a solid model with determined variables affecting the technological acceptance decision of the companies. It only provided a taxonomy for the determinants within their respective context for categorization (Gangwar et al., 2015). In other words, the TOE framework has ambiguous major constructs and is too broad. Thus, this framework is required to be reinforced by incorporating it with the models having unblemished constructs. Therefore, scholars have encouraged the need for combining TOE and other models so that the predictive strength of the resulting model can be advanced and some of their specific shortcomings can be surmounted. Further, in the CEO’s context, TOE does not have clear constructs for innovation acceptance. As a result, to address this void, the IS framework for SMEs has been incorporated.

Figure 3. Conceptual framework

6. Theoretical implications

The theoretical framework for the development of BIS is intended to provide an accurate description of potential determinants and a clear prediction of the successful adoption of BIS in organizations. It may affect the decisions of practitioners prior to integrating and adopting the BIS in their organization. The proposed framework would significantly contribute to the development of the BIS acceptance/adoption theory that is almost non-existent in current literature (Ahmad, Miskon, Alkanhal et al., 2020). A few developed models and theories have been proposed for the adoption of the BIS at the organizational level as well as for the acceptance of the BIS at the individual level. This is one of the first theoretical structures that has implemented the acceptance construct at the level of the individual as well as TOE frameworks with the goal of implementing the BIS at the organizational level. In the literature, all contexts have been extensively discussed and independently researched, but limited researchers have dwelled on the value of user acceptance to complete the implementation of BIS at the company level. Further, not only will this conceptual model and theoretical framework contribute to the body of knowledge in the BIS field, but it will also open new research horizons. In addition, the new conceptual model may be generalized or refined to produce new models or theories.

7. Practical implications

A systematic literature analysis of selected research from business and scholarly publications is the foundation of the present article. Next, a list of the most possible determinants is presented in the results. In different circumstances of each organization, determinants were used for the adoption and recognition of BIS in organizations, but the same determinants could lose their meaning in other business scenarios. Therefore, the collection of determinants according to the specifications of the organization and sector is necessary for producing better returns from BIS programmes. Further, prior studies indicate that the BIS raises market value generally in sectors such as education, telecommunications, insurance, research, supply chain, and retail chain businesses, etc. Hence, this analysis would make a considerable contribution to decision-making processes prior to the incorporation of the BIS into businesses. To be specific, this research empowers market analysts and policymakers to achieve a better understanding of the various aspects of an effective implementation of BIS determinants. Various determinants are illustrated here; strategic interest may be drawn, such as, manager’s innovativeness and management support. Managers need to understand the key steps that contribute to the successful implementation of BIS in organizations in order to pursue these study results. In terms of environmental determinants, companies should also be mindful of the essential challenges and uncertainties associated with BIS implementation. This study will direct BI vendors and cloud service providers to pay attention to complexity and compatibility issues in solving BIS investments, especially for small to medium-sized organizations in developed countries.

8. Research limitations and future research guidelines

Like many other studies, this study is not without some drawbacks. Firstly, owing to the nature of the published research, the determinants are studied initially from a theoretical background. The determinants are taken from the studies defined for the SMEs in the healthcare sector. For different sectors, there could be different determinants that affect the decisions of the companies in various circumstances to adopt the BIS. Secondly, the key purpose of this review was to explore the significant determinants that could affect the decision of the organizations to conceptually implement the BIS. The study was carried out on the theoretical basis findings of previously obtained quantitative. There is a shortage of case studies needed to show in a realistic manner that the current research results are more relevant. Practical implementation is expected by industry experts and practitioners in order to further expand the current study findings. Thirdly, owing to human intervention, the complexity of word definitions and textual interpretation methods, biases cannot be eliminated, while utmost attempts are taken to ensure that the findings are more reliable and minimize the biases. Therefore, from a theoretical viewpoint, the findings might represent typical phenomena. Very few researches discussed the approval of BIS at the individual level. Individual level theories and frameworks such as decision theory, motivation theory, stakeholder theory, and social cognitive theory for implementing the BIS should be considered and suggested by researchers. Before the acceptance decision of BIS in organizations, RBV and critical Success Factor (CSF) theories should also be considered. These will help to define the significant resources and skills that form the strategic edge and lead to the success of BIS projects. Fourthly, this study explored the determinants comprehensively with analytical lenses that were most important for the BIS research field. However, there was no association between the determinants and their effects on the consequences of success, such as the efficacy and reliability of the decision-making process. In addition, a meta-analysis may be used by research to broaden the awareness in this area. Fifthly, the framework developed in this SLR needs to be tested using different statistical techniques such as PLS-SEM and/or MCDM. This would provide a clear picture of the most significant factor(s) that influence the adoption decision. Finally, the findings of this study can be used by researchers and practitioners to better understand and properly align their efforts to address the practical challenges of adopting BIS. Researchers can use this literature review to get a summary of existing studies, define new research issues, and place and coordinate their own work. More specifically, they can use this classification system to explain and analyze the context of the organizations they are researching. Further, the framework presented here can be used by practitioners to consider how they can exploit BIS in their own context and to recognize the realistic obstacles they might encounter while doing so.

9. Conclusion

A systematic literature review was undertaken to explore the most possible determinants and theories that impact the adoption and recognition of BIS in organizations. A total of 63 studies conducted in the last decade (2011–2020) were identified and reported. After the analysis of the results, a total of 22 determinants was identified and grouped into four contexts namely technology, organization, environment, and CEO’s characteristics. In comparison, the researches listed were carried out in the healthcare industry. The findings also revealed that most of the BIS studies on the current topic of research were done for banks and several companies in developing countries. Therefore, from the viewpoint of the implementation of BIS, researchers should also pay heed to developing countries. Finally, more experiments are proposed to be carried out with respect to the uncovering of determinants that could impact the BIS projects by applying some other models and theories, including refining current theories and models. It is also noted that there is sufficient room for the current theoretical structure and conceptual model to be tested and checked with quantitative, qualitative, and mixed process testing methods to develop more refined models in the future.

Additional information

Funding

This work was supported by Fundamental Research Grant Scheme (FRGS) (Grant Number: 14449) from Kementerian Pengajian Tinggi (KPM) & Universiti Utara, Malaysia.

Notes on contributors

Isyaku Salisu

Dr. Salisu Isyaku is a lecturer in the Department of Business Administration, Umaru Musa Yar’adua University, Katsina. He received his Ph.D. in Entrepreneurship from University Utara Malaysia (UUM). He obtained his B.Sc. M.Sc. at Bayero University, Kano. He is currently pursuing another Ph.D. in Decision Sciences/Operation Research at UUM. His research interest is entrepreneurship and business analytics.

Mazlan Bin Mohd Sappri

Dr. Mazlan Bin Mohd Sappri is a lecturer in Decision Science, School of Quantitative Science, UUM. He holds a bachelor’s and master’s degree in information technology from UUM and UKM in 2001 and 2004 respectively. He also obtained his Ph.D. in Computer Science at USM. His work is mainly in Management Information System, Performance Metrics

Mohd Faizal Bin Omar

Dr. Mohd Faizal Bin Omar is an Associate Professor of Decision Sciences, School of Quantitative Sciences, UUM. He received his PhD at Queensland University of Technology, Australia. Currently, he is a Director of Innovation and Commercialisation Centre, UUM and he is also a Research Fellow at Centre for Testing, Measurement and Appraisal (CeTMA)