HE380.0.1 Managed Healthcare Assignment 8

nmbeytrickrtts

Chapter6Option2.pdf

Home >Business & Finance homework help >HE380.0.1 Managed Healthcare Assignment 8

RESEARCH ARTICLE

Exploring the impact of operating

model choice on the governance

of inter-organizational workflow:

the U.S. e-prescribing network

Nelson King

Business Information & Decision Systems,

Olayan School of Business, American University

of Beirut, Beirut, Lebanon

Correspondence: Nelson King, Business Information & Decision Systems, Olayan School of Business, American University of Beirut, Bliss Street, POB 11-0236, Riad El-Solh, Beirut 1107 2020, Lebanon. Tel: þ961 1 350000/3731; Fax: þ961 1 750214

Received: 15 November 2010 Revised: 9 June 2011 2nd Revision: 16 October 2011 3rd Revision: 3 February 2012 4th Revision: 13 May 2012 5th Revision: 9 July 2012 6th Revision: 8 August 2012 7th Revision: 30 August 2012 Accepted: 12 September 2012

Abstract Inter-organizational networks play an increasing role in delivering computer- mediated public services such as healthcare. Many networks govern through an

infomediary (i.e., electronic broker) that brings together disparate member

organizations. These networks can resemble an enterprise where standards and incentives for use are imposed on its partners. This study seeks to extend an

enterprise IT governance (ITG) concept to the U.S. e-prescribing network as it

transitions from a paper-based network to a computer-mediated one. The operating model, proposed by Ross et al (2006), emphasizes choices in stan-

dardization and integration to align strategy with operational processes to

improve enterprise performance. Missing in their work is evidence that macro-

level choices embedded in the operating model directly impact network workflow. A comparative synthesis traces the changes made to the U.S.

e-prescribing operating model to their impact upon the roles and relationships

among network members. Some workflow mis-alignments were traceable to the operating philosophy imposed by healthcare policy-makers. The study

suggests IT alignment in networks may be better achieved through governing

operating models rather than the traditional ITG focus on organizational forms. European Journal of Information Systems (2013) 22, 548–568. doi:10.1057/ejis.2012.47; published online 13 November 2012

Keywords: healthcare information systems; IT governance; system design; operating model; inter-organizational network; workflow

Introduction Information systems embedded in an inter-organizational network (‘network’) require new ways of thinking about IT governance (ITG) as the traditional enterprise is displaced by a collection of private and public entities. While ITG research has begun to concern itself with enterprises whose boundaries extend to its suppliers or customers (Lee, 2009), the locus of ITG remains the enterprise – not the network (Tapia et al, 2008; Croteau & Bergeron, 2009). However, enterprise systems are prone to poor assimilation of new work processes and organizational design (e.g., Robey et al, 2002) despite the presence of enterprise ITG (Ross et al, 2006, p. 65). The dispersed nature of networks and the heterogeneity of ITG within member organizations suggest that assimilation of processes and organiza- tional design will also be problematic in networks.

Recent research about ITG in networks (‘network ITG’) echoes these assimilation concerns. Researchers examined a national payment network

www.palgrave-journals.com/ejis/

(Croteau & Dubsky, 2011), some criminal justice initia- tives (Pardo et al, 2008), emergency management (Marich et al, 2008; Vogt et al, 2011), and various healthcare projects (Wiggins et al, 2006; Sulistyo, 2009; Bygstad & Hanseth, 2010). These researchers point to a limited understanding of network ITG, especially with respect to the alignment of strategy to the roles per- formed by various members of the network (Wiggins et al, 2006; Croteau & Bergeron, 2009). For network members, the alignment must provide sufficient benefit to make adop- tion worthwhile (Tapscott et al, 2000; Davidson & Bryant, 2002; Wiggins et al, 2006; Borman & Ulbrich, 2011).

ITG seeks to align IT functionality with business needs thereby delivering value to the enterprise. ITG employs different methods for alignment ranging from organiza- tional structure to governance processes. Network ITG has similar governance objectives except applied to a more complex arrangement of organizations. This study is motivated by the need for deeper insight into the role of network ITG in the alignment of network strategy to its workflow, permitting proper assimilation to take place. U.S. e-prescribing serves as a case study to explore the extension of the operating model concept (e.g., Ross et al, 2006), currently used in enterprise ITG, to a network. A brief introduction to the U.S. e-prescribing network follows to illustrate that a macro-level choice (e.g., a pre- scriber initiating an electronic transaction) specifies work- flow (i.e., sequence of steps between network members).

Outpatient e-prescribing (‘e-prescribing’) in the United States embodies a transaction-driven, nationwide, public- private healthcare network. The members of this network (i.e., businesses that provide prescription-related health- care services) would encompass upwards of 250,000 locations around the country if e-prescribing were fully deployed by these businesses. Figure 1 shows a simplified representation of the U.S. e-prescribing network. The diagram shows the roles and relationships between key members of the network and illustrates the flow of messages between them.

The processing of an electronic prescription (e-script) is analogous to an order transaction in business (Figure 1). The transaction begins with a physician in a medical practice who generates the e-script (i.e., order entry). The e-script can be transmitted to any participating phar- macy in the network via the network infomediary. Once approved, a pharmacy dispenses the medication (i.e., order fulfillment). Order approval comes from a phar- macy benefit manager (PBM), who determines at the behest of a payer, the eligibility of a patient and determines the reimbursement to the pharmacy for the prescribed medication. The PBM approves the request electronically unless a problem exists. The relationships between mem- bers of the network are more complex than described here with further details in the case analysis section.

This study seeks to shed light on network ITG by building upon the traditional enterprise focus of ITG research. Some concepts used in enterprise ITG may be applicable to networks, especially to align computer- mediated workflow to network strategy. Acceptance and assimilation of this workflow by network actors should lead to adoption. The U.S. e-prescribing network serves as an exemplary case since, to some extent, the central role of the infomediary controls the actions of its members much like an enterprise controls its business units. This exploratory study seeks to address the research question: can the strategic choices embedded in a network operat- ing model be traced directly to its impact on workflow, which then determines the roles and relationships among network actors?

This research found that macro-level choice for policy or strategy, embedded in the operating model, can directly affect organizational coordination and the roles of net- work actors in at least the U.S. e-prescribing network. This finding leads to the contribution of the paper: the operating model can serve as a method for inter- organizational governance without imposing specific ITG methods on all network members, whose capacity to employ ITG may vary considerably. While the impact of changes in the operating model for U.S. e-prescribing are likely network-specific, the methodological approach of examining operating model choices should be applicable to the specific conditions of other networks.

The theoretical framework section begins by describing the configuration of a network. The tenets of ITG are then summarized in the context of governing IT for a network. The rationale for applying the ITG concept of operating model to a network follows. The methodology section describes the challenges of analyzing an entire network rather than its individual members. The study uses comparative synthesis to address the call for long- itudinal analysis of networks (Knoben et al, 2006; Raab & Kenis, 2009). The comparisons seek to show that macro- level choices embedded in the operating model directly impact micro-level outcomes (i.e., assimilation of work- flow). The findings are presented in three sections: macro-description of the network, micro-description of the network workflow and the implications of e-prescribing

Pharmacy Benefit Manager

Pharmacy

Order Entry

Order Approval

Order Fulfillment

Transmit

Infomediary

Medical Practice

Patient Eligibility

Payer

Member Type

Transaction Step

Message

Legend

Certified Vendors

Flow

Figure 1 E-prescribing network – members and messages.

Governance in e-prescribing Nelson King 549

European Journal of Information Systems

operating model choices. The first of these sections con- trasts the pre-existing network (i.e., manual prescribing) with the intended e-prescribing implementation at a macro-level (i.e., operating model). The second section makes the same comparison at a micro-level to show the impact on network roles and relationships. The last section traces specific choices implemented in the U.S. e-prescribing operating model to mis-alignments encoun- tered in practice.

Theoretical framework Studies of network ITG are in their infancy, so this study builds upon existing enterprise ITG concepts that could be extended to a network. Throughout this paper, the term ITG when used alone refers to enterprise IT gover- nance while governance for a network is referred to as network ITG. This section begins by showing that the U.S. e-prescribing network has structural similarities to an enterprise. The conceptual underpinnings of ITG are then described to set a context for ITG in a network setting. Finally, a rationale for choosing an operating model to directly align network strategy to its operations is offered.

Network configurations A description of public networks is first necessary to draw parallels to the enterprise where the bulk of ITG research has focused. A network represents ‘consciously created forms of social organization whose members strive to achieve common goals’ (Raab & Kenis, 2009, p. 205). The effectiveness of a network depends on the extent that control mechanisms are applied in the production of collective goods (Raab & Kenis, 2009). Network researchers in public administration pay attention to configuration (Kenis & Provan, 2009) and governance (Provan & Kenis, 2008). Provan & Kenis (2008) identified three configura- tions for public networks: shared governance, lead organi- zation, and network administrative organization (NAO). The NAO resembles a centralized enterprise in which standards and incentives are imposed from top-down, such as with supply chain partners. The U.S. e-prescribing network, with its privately run infomediary (i.e., Surescripts), is best characterized as an NAO (King, 2012).

IT governance context ITG frameworks have become increasingly important in the past two decades. Brown & Grant (2005) point to two streams in the ITG literature that have developed since the 1990s: forms and contingency analysis. They grouped these forms into the locus of IT decision-making and IT decision-making structures. For example, Kayworth & Sambamurthy (2000) found the perception of operating units towards the imposition of PC/LAN network stan- dards depended on the organizational context and the extent that processes were stipulated. Contingency analysis examines the fit of parameters (e.g., organiza- tional and decision-making structure) to an ITG frame- work (e.g., Strategic Alignment Model, Henderson & Venkatraman, 1993). According to Brown & Grant (2005),

a merging of these two streams has begun, pointing to the work on governance archetypes, such as the IT monarchy (e.g., Weill & Ross, 2005). An archetype addresses a number of IT decisions including IT architecture, infrastructure strategies, and business application needs.

Recent research seeking a more refined definition for ITG places greater emphasis on the ramifications of existing ITG frameworks to the operational levels of the organization. Webb et al (2006) point to the convergence of corporate governance and strategic information systems planning in the 12 ITG definitions they examined. The resulting definition emphasized functional areas of ITG:

IT Governance is the strategic alignment of IT with the

business such that maximum business value is achieved

through the development and maintenance of effective IT

control and accountability, performance management and

risk management. (p. 7)

Emphasizing the alignment between strategy and operations, Simonsson & Ekstedt (2006) synthesized defini- tions from 60 articles and defined ITG as ‘the preparation for, making of, and implementation of IT-related deci- sions regarding goals, processes, people, and technology on a tactical or strategic level’ (p. 24). These two com- posite definitions not only reflect the role of ITG but also point toward alignment to operational considerations: achieving maximum business value at the tactical level (e.g., workflow).

There is growing recognition that the alignment of IT with business goals may not on its own improve enterprise outcomes (e.g., Shpilberg et al, 2007). De Haes & Grembergen (2006) suggest going beyond strategy and drawing information from each ITG level (strategic, managerial, and operational). In a study of structure and processes for 329 independent insurance agencies, Zaheer & Venkatraman (1995) concluded ‘merely adopt- ing an appropriate governance structure may not lead to the required process outcomes of governance’ (p. 388). Perko (2008), in a study of 50 Finnish firms implement- ing a service-oriented architecture, found that ‘while business and IT agree at the strategic level, without a clearly articulated operating model they have difficulties in achieving the same amount of alignment at opera- tional level’ (p. 189). This gap between strategy and operations may be why Simonsson & Ekstedt (2006) include ‘understanding’ as a component of ITG, offering the example ‘modeling complex problems to make them understandable for all stakeholders’ (p. 21). All of these authors suggest that ITG must look beyond strategic alignment to consider how strategy impacts lower level outcomes (e.g., workflow).

Achieving this linkage depends upon more explicit consideration of technology and processes. Ross et al (2006) argue that enterprise architecture provides this linkage. Henderson & Venkatraman (1993) point to information technology infrastructure and processes as one of the strategic pillars of ITG. Yet, according to Romero (2011), few organizations actively incorporate IT

Governance in e-prescribing Nelson King550

European Journal of Information Systems

architecture processes (e.g., linkage to operational level) as part of ITG. This ITG gap in practice, between strategy and operations, poses additional challenges that will likely carry over to any extension of ITG concepts to networked organizations.

In conclusion, the predominant approach to ITG in academia and practice (e.g., ISO/IEC 38500 conceptuali- zation) can be summarized as structure-driven govern- ance (Figure 2, left). There are numerous variants, but this ISO/IEC diagram is a widely accepted representation. Stra- tegic concerns (business needs and pressures) are linked to lower level IT implementation efforts that embed busi- ness processes via ITG, implemented through direction (plan) and assessment monitoring (performance, confor- mance). While the locus and structure of decision-making are prominent in ISO/IEC 38500, explicit linkages between strategy and operations may not be (i.e., Romero, 2011). The operating model conceptualization of ITG (Figure 2, right), proposed by Ross et al (2006), is addressed later in this section.

ITG for networks The ITG community acknowledges that inter-organiza- tional ITG still needs to be developed. Recognizing that governance extends beyond the boundaries of a single organization (Buckl et al, 2011), the IT Governance Institute states ‘In the extended enterprise environment there is no standard pre-existing governance structure’ (IT Governance Institute, 2005, p. 67). Similarly, COBIT (Control OBjectives for Information and related Technol- ogy) does not address a networked organization perspec- tive (Tapia et al, 2008). Referring to the archetypes of Weill & Ross (2005), Croteau & Bergeron (2009, p. 3) point out that ‘these archetypes and views of IT gover- nance have been determined within firms and not in an interorganizational context’. Furthermore, no contingency theory has yet been put forward to develop a set of governance mechanisms that are based on network form, mixture of public and private firms, relational mechan- isms, and the nature of coupling between operational

processes (Barringer & Harrison, 2000; Croteau & Dubsky, 2011).

Networks face several challenges with respect to ITG. The network increases access to information across its membership through the use of shared resources (Klein et al, 2005). However, these members represent a disparate community at many levels: IT infrastructure, IT strategy-making processes, and IT vendor management processes (Croteau & Bergeron, 2009; Hekkala et al, 2010). The notion of inter-firm coupling raises an interesting dilemma for network infrastructure design. Klein et al (2005, p. 178) ask:

Is one of the key goals of a shared or common ICT infra-

structure in interfirm business networks to achieve com-

mon definition and meaning of key information entities

across the network or rather flexibility needed in order to

improve existing work practices?

Given these challenges, finding common ground between enterprise ITG and network ITG can be difficult. Three com- mon factors are considered in this study: locus of decision- making, strategic planning, and organizational coupling.

The locus of decision-making is one common con- sideration. Weill & Ross (2004, p. 2) define ITG as the ‘decision rights and accountability framework to encou- rage desirable behaviour in the use of IT’. However, Mueller et al (2008) tell IT practitioners that these decision rights take place at the ‘lowest levels’ of ITG decomposition (e.g., content of a transaction message). Several inter-organizational studies also point to a broader distribution of decision rights. Croteau & Dubsky (2011, p. 4) state that ‘an appropriate IT governance structure is meant to adequately distribute power and responsibility, and so each organization’s structure will play a role in defining the relationship’. The division of work between organizations is important in networks (Kumar & van Dissel, 1996; Croteau & Dubsky, 2011). According to Kumar & van Dissel (1996, p. 284), ‘Struc- ture, by formalizing the form, process, and content of the relationship, implies a level of agreement about mutual expectations’. Top-down authority to regulate the net- work is limited, so specific coordination mechanisms (e.g., imposed data transfer standards) are used to address the governance problems of control (Kurimoto, 2008). The first challenge for network ITG is to incorporate decision-making made at the lowest levels of the network.

Strategic planning also plays an important role in networks because of the necessity to divide work between organizations. Network stakeholders are often ‘multiple independent entrepreneurs [who] have a sizeable stake in the development and outcome of the network’ (Kurimoto, 2008, p. 2). How are these stakeholder voices heard in large networks? Spil & Salmela (2007) found in their study of 15 networked healthcare organizations that the lack of formal strategic planning, a central capability of ITG, led to ill-defined plans. In addition, information infrastructures with many stakeholders and technologies are, in fact, heterogeneous networks without central

Operating Model Conceptualization

D ire

ctio n

M o n ito

Strategy

Operating Model

IT Engagement

Business Units Implement Process

Adapted: Ross et al (2006)

Traditional IT Governance

Adapted: ISO/IEC 38500

IT Governance

Business Processes

Plan

Performance

Conformance

IT Projects IT Operations

Direct Monitor

Evaluate

Business Pressures

Business Needs

Figure 2 Alternative approach to IT governance.

Governance in e-prescribing Nelson King 551

European Journal of Information Systems

control and thus are likely to fail as seen in the Norwegian health sector (Bygstad & Hanseth, 2010). Bygstad et al (2010) see network control as using a regulatory approach revolving around ‘law, norm, incentives and architec- ture’. Integrated planning would be difficult given the ‘multiple, mostly intangible or conflicting goals’ in public sector organizations (Sethibe et al, 2007) and the size of a public network. The challenge remains to socially embed ties within the network but keep them at ‘arm’s-length’ (Akhlaghpour, 2008). Thus, the need for centralized strategic planning is a second challenge for successfully governing a network such that it reflects the diversity of network members without directly involving all of them.

A third common factor is a governance structure that embeds operational processes within an information system. These processes specify the roles and relation- ships between organizational units and external mem- bers. The degree of coupling between organizations is important (Croteau & Bergeron, 2009), with success tied to increasing inter-organizational participation through the formation of business relationships (Hong, 2002). The key to participation is not its formal aspects (e.g., rules and contracts) but the nature of the relationship and how the organizations are coupled (Hong, 2002). Governance must address the organizations’ mutual dependence, not just the mechanics of information exchange (Kurimoto, 2008; Borman & Ulbrich, 2011). Governance in a network can be seen as ‘establishing and employing power’ to coordi- nate member efforts (Heide, 1994; Croteau & Dubsky, 2011). Yet the broker (i.e., infomediary) generally benefits as the network member who ‘leads the choreography, value realization, and rule making activities of the system’ (Tapscott et al, 2000, p. 19). The challenge for network governance is centralization while being decoupled from the complexity of stakeholder (member) relationships, while at the same time aligning strategy with operational details.

Operating model rationale This paper suggests an ITG conceptualization for the enterprise that may address the governance challenges of a network. Ross et al (2006) recommend that the enterprise define an operating model because critical decisions must be made concerning which strategies will be supported. These choices identify ‘key customer types, core processes, shared data, and technologies to be standardized and integrated’ (Ross et al, 2006, p. 65). Mueller et al (2008) add that these decisions include not just the extent of integration and standardization, but also the definition of strategic limits and core capabilities. The operating model provides a more ‘actionable’ view of a company than strategy alone (Ross et al, 2006). The exercise of defining the operating model presumably forces decision-makers to articulate their thoughts on the interaction between workflow and strategy. Ross et al (2006) subsume the locus of decision-making and control elements of traditional ‘IT Governance’ (triangle Figure 2, left) into the ‘IT Engagement’ portion of their framework

(Figure 2, right). They define IT engagement as ‘the system of governance mechanisms assuring that business and IT projects achieve both local and company-wide objectives’ (Ross et al, 2006, p. 65). The traditional gover- nance roles (i.e., direction and monitor) continue to exist in the Ross et al (2006) framework, but the emphasis shifts to the operating model.

The operating model concept has gained traction in the IT practitioner community through its use by the IBM Red Book (Mueller et al, 2008) and The Open Group (2009). Practitioners like DuMoulin & Probst (2011) consider an operating model ‘an extension and deliver- able of IT Governance’ as it represents the ‘blueprint’ of the IT value chain process architecture. Similarly, Phillipson (2008) says the operating model serves as a ‘shock absorber’ by insulating the implemented system from the external environment while providing the blueprint for an evolving business.

The emerging evidence for the operating model indicates its importance not only for enterprise perfor- mance, but also for ITG. Ross et al (2006) provides macro- level evidence (e.g., firm performance indicators) for the efficacy of the operating model by examining 68 case studies and two surveys that included 180 firms. Perko’s (2008, p. 191) study of 50 Finnish companies that implemented a service-oriented architecture found a similar impact: ‘IT governance needs a solid basis to work on; a clearly articulated operating model was found to have a positive effect on almost all aspects of IT governance’. However, only 41% of the companies they studied had a clearly articulated operating model.

Choosing what to embed in an operating model is a major challenge, and little research has been conducted concerning how strategy is linked to operational pro- cesses. Micro-level details of the actual work supported by an IT infrastructure are generally not available to senior managers who articulate the operating model (Ross et al, 2006; Singh & Woo, 2009). At the same time, the initial interpretations of stakeholders are significantly influ- enced by the scope and adaptability of the system’s functionality, which depend on knowledge of these details (Doherty et al, 2006). The action research of de Vries et al (2010) found significant deficiencies in the way that senior managers built an operating model. These authors emphasize the need to analyze business (model) architecture parameters and find opportunities for data sharing and replicating processes across organizing entities. Significant IT impacts are known to occur at low levels in the organization (Barua et al, 1995).

An operating model conceptualization for network ITG should provide transparency to the whole network. Strategic decisions have to be explicated early on, thereby giving visibility to stakeholders who need to be kept at ‘arm’s-length’ due to their sheer numbers. System developers must also immerse themselves in the depths of the operating environment (Wilkin, 2009). This paper seeks to show that a linkage exists between strategic choices articulated for the operating model and the

Governance in e-prescribing Nelson King552

European Journal of Information Systems

impact on network workflow, specifically to roles and relationships among network members.

Coordination in inter-organizational networks Coordination is the heart of network ITG, particularly organizations coupled through a network transaction, as in e-prescribing. Scholars are revisiting coordination within and between organizations (Sinha & Ven, 2005; Okhuysen & Bechky, 2009; Kartseva et al, 2010). Coordination is ‘managing dependencies between activ- ities’ in which organizations perform interdependent activities to achieve goals (Malone & Crowston, 1994, p. 90). Coordination problems stem from ‘dependencies that constrain how tasks can be performed’ (Malone & Crowston, 1994; Crowston, 1997).

The management of inter-organizational work is known to be difficult due partly to business processes owned and managed by independent business units (Cameron et al, 2005; Fonstad & Robertson, 2006), or companies in the case of a network. More attention must be placed on the mix of structure, process, participant, and maturity of organizational relationships (Simonsson & Ekstedt, 2006; Croteau & Bergeron, 2009). In addition, IT alignment with strategy is still important in an inter-organizational context (Wieringa, 2008; Zarvić, 2008). This means that a micro-level analysis of the roles and relationships within the workflow of a network, and analyzing their connection to strategy via the operating model, are critical concerns in the study of governance (e.g., Kartseva et al, 2010).

Methodology This study seeks to show that macro-level choices (e.g., e-prescriber sends e-script – not someone else) embedded in the e-prescribing operating model have the leverage to alter operational processes at the level of network member. These choices dictate roles and relationships (e.g., e-prescriber, rather than patient, responsible for transmitting e-script to pharmacy) and thereby influence the assimilation of e-prescribing technology. This work originated from a prospective analysis for the IBM Center for The Business of Government to inform U.S. policy- makers about the impact of e-prescribing as the pilot phase ended in 2006. That roles and relationships would be altered when a manual prescribing operating model is supplanted with e-prescribing is not surprising. This study argues these alterations are potential mismatches ‘between desirable behavior and governance’ (Weill & Ross, 2004; Ross et al, 2006). These mismatches reflect choices made for the operating model (by the broker) rather than solely mis-aligned organizational forms or mechanisms at a lower level of ITG (e.g., lack of standards or insufficient incentives).

Network ITG must look at the whole network since its output results from the collective action of network members. Yet most scholars refrain from studying the complexity of collecting whole network data and focus on a dyadic relationship within a network (Provan et al, 2007; Provan & Kenis, 2008; Raab & Kenis, 2009). With few exceptions (e.g., Ross et al, 2005; Hollingworth et al,

2007), e-prescribing studies rarely look at multiple actors within a single class of network member (e.g., medical practice) let alone those they are connected with. There are some examples in the e-prescribing literature that consider a dyadic variable, though not necessarily studying the dyad itself. A recent study of medical practices adopting e-prescribing did include the dyadic variable ‘number of callbacks’ received from a pharmacy (Dainty et al, 2012). This research takes a first step towards applying whole network analysis, prior to widespread adoption, to healthcare information technology by synthesizing the expected change in relationships among all members whose employees ‘touch’ the e-prescribing transaction.

The study of a nation-wide network also faces the challenge of a widely distributed population of members who interact with each other and inside their own organizations. While interviews were initially conducted at a few sites, they could not capture the variability in which work is done in roughly 250,000 sites (medical practices and pharmacies) that are expected to join the network. The research must ‘scale up’ from a traditional ethnographic site (e.g., an organizational unit) (Star, 1999) to study tasks among dispersed actors and their firms in a network. According to Lyytinen & Damsgaard (2011), a single ‘adopter’s behavior neither form an independent observation unit (as most statistical analyses assume) nor can they be analyzed in isolation’ (p. 4). In addition, the e-prescribing network is undergoing a ‘fundamental overhaul’ according to the eHealth Initia- tive (Teich et al, 2004). This perturbation to the network requires longitudinal analysis to monitor its evolv- ing form (Knoben et al, 2006; Raab & Kenis, 2009). The network challenges of collective action, heterogeneous membership, and a disruptive change to the operating model drove the study to a comparative synthesis approach.

Synthesis method background This study follows the analytical tenets of a case descrip- tion (Yin, 2003) using a comparative synthesis approach for data collection. A research synthesis seeks to utilize the wealth of narrowly focused quantitative and/or qualitative studies that surrounds e-prescribing, one of many fields of study in medication management, by connecting them into a ‘reasonable representation’ to explore underlying phenomena (Rousseau et al, 2008). Work practices are highly situated so interpretive methods like ethnography play a prominent role in organizational studies (Bernard et al, 1984) and especially ‘e-infrastructures’ that share information across diverse groups of users (Pollock & Williams, 2010). These identified work practices, the data for this study, describe the interactions among actors for a particular activity within a workflow.

A synthesis tells a systematic story that draws upon all forms of research designs including randomized con- trolled trials and observational studies (Popay et al, 2006). For this study, synthesizing (piecing together) a transaction- based workflow from ‘informants’ (e.g., a peer-reviewed article) is a virtual ethnography of sorts. An informant

Governance in e-prescribing Nelson King 553

European Journal of Information Systems

points to either new informants (e.g., citations) or new topics. These topics might appear as a questionnaire item or an anecdotal observation in interview data, not just the findings discussed by study authors. These infor- mants are drawn from the broader field of medication management rather than e-prescribing alone. Table 1 includes a sample of ‘informants’ that include dyadic relationships among actors in the network and the research method employed. These informants represent the broader field of medication management that encompass e-prescribing.

Synthesis methods are popular in healthcare, especially nursing and evidence-based medicine, due to the large volume of research that exists. Several books on synthesis approaches in these healthcare areas delve into the sub- tleties of applying the various methods available for the benefit of researchers (Paterson et al, 2001; Sandelowski & Barroso, 2007). Synthesis approaches that blend together quantitative and qualitative findings are of particular interest to this study. Paterson et al (2003) found that qualitative investigations of chronic fatigue shared meth- odological assumptions with a parallel body of quantita- tive research that focused on measurable factors rather than meaning and context. Sandelowski et al (2008) synthesized both qualitative and numerous quantita- tive studies on stigmas facing HIV-positive women. In

e-government, Siau & Long (2005, p. 448) started a line of research using meta-synthesis to ‘compare, interpret, trans- late, and synthesize different [e-government] research frameworks’.

This study embraces the comparison of evidence collected using different methods (i.e., a synthesis), but goes further to compare the network during its transition. Most IS practitioners use some form of ‘as-is’ process modeling to determine the impact of an intended system on its adopters (e.g., Curtis et al, 1992). Adoption studies need to consider changes both within an organization and between network members (Cho et al, 2007). Network structures are known to evolve particularly when there is dyadic change at the network level (Knoben et al, 2006).

Comparative synthesis method Figure 3 illustrates the sequence of research steps used in this study. The first step (circled number ‘1’) builds a synthetic model of existing workflow (prior to e-prescrib- ing) from the medication management literature (e.g., medicine, pharmacy, medical informatics). Pre-existing dataflow models for the prescribing artifact are the starting point (Teich et al, 2004; Johnson & FitzHenry, 2006). However, the process of writing a prescription is only the most visible part of a workflow that also includes generating, transmitting, and adjudicating before dispensing

Table 1 Representative medication management relationships literature

Citation Journal Method Sample Relations Finding

Mott & Cline

(2002)

Medical Care Logbook survey

(self-reported)

86 community pharma-

cists, 6380 prescription

orders for generic

substitution

Prescriber-Pharmacist Pharmacist substi-

tuted 84% of

allowable

prescriptions

Lapensee

(2003)

Journal of Managed

Care Pharmacy

Cross-sectional

statistical analysis

One month (22,009)

of Prior Authorization

(PA) requests

Prescriber-Payer-Patient 95% PA authorized

Kajioka et al

(2005)

The American

Journal of

Emergency

Medicine

Quantitative Emergency records

+pharmacy claim data

for 3 months

(403 urgent cases)

Prescriber-Patient-

Pharmacy

65% of high-

urgency

prescriptions

were filled

Linton et al

(2007)

Journal of Managed

Care Pharmacy

Outpatient prescription

fill records analyzed

in NC, TX, CA

TRICARE beneficiaries

age 65 years or older

(N¼300,084)

Patient-Pharmacy 67% use one

dispenser for

4 medications

Kassam et al

(2008)

Pharmacy Practice Semi-structured

face-to-face interviews

20 practicing pharmacists

(1 h each – 9 questions)

Pharmacist-Patient ‘develop partner-

ship with patient’

Brown et al

(2006)

American Journal

of the Medical

Sciences

Grounded Theory 30 pharmacists in

3 focus groups

Prescriber-Pharmacist Pharmacist final

checkpoint for

error detection

Tarn et al

(2008)

Patient Education

and Counseling

Audiotape –

pre/post visit

survey (36 questions)

900 patient encounters –

55 physicians – HMO and

teaching hospital

Physician-Patient 23 s spent

justifying new

drug and use

Governance in e-prescribing Nelson King554

European Journal of Information Systems

a prescription (Bell et al, 2004). Within the workflow there are dyadic acts between network members. For example, the simple task of handing a completed prescrip- tion to a patient involves a dyadic act between network members (medical practice and patient). Known roles and relationships often govern this act (e.g., patient has responsibility to fill the prescription not the physician).

The synthetic workflow describes more than a standard business process. Dyadic acts are dependencies between actors that constrain how tasks are performed (Crowston, 1997). Insights into these dependencies described by informants are of particular interest. For example, physicians in manual prescribing typically leave it to the pharmacist to substitute a brand name drug with a generic when allowed by a payer. Why did manual prescribing forego these opportunities for a dyadic relationship? The synthetic researcher would then seek an explanation. They might find an informant that reports physicians don’t have time to talk costs with patients (Alexander et al, 2005; Beran et al, 2007). Additional informants are found through ‘snowballing’ (e.g., Greenhalgh & Peacock, 2005) backwards by using citations within the article or forwards using ‘cited by’ lists available on search engines like Google Scholar. The synthetic researcher would then uncover that physicians don’t know what drugs cost (Ernst et al, 2000) and find it easier to write the brand name since they don’t have to memorize generic variants (e.g., Kwo et al, 2009). Adding context to the existing prescriber-pharmacist relationship helps anticipate the behavior of this dyad in the intended system.

While new technologies take years to reach mass adoption, the intended usage of a new system can be determined in advance from the statements of those advocating change. The intended model builds upon informants that include advocacy reports (e.g., Teich et al, 2004), feature lists and published standards for vendors (e.g., Bell et al, 2004; Wang et al, 2005), and early pilot studies (e.g., Tamblyn et al, 2006; Bell et al, 2007).

The synthetic model of intended workflow (step 2) is analogous to a ‘to-be’ process model except enhanced to reflect the dyadic relations.

The existing and intended workflow models are not built in isolation. The task boundaries for both models have to match in order to conduct a comparative analysis. The existing model may inform the intended model or vice versa. For example, manual prescribers often delegate the refilling of prescriptions to their medical staff (De Smet & Dautzenberg, 2004). A synthetic researcher aware of this practice would purposefully look for the use of surrogates within the intended design of e-prescribing. One of the e-prescribing pilot studies did in fact stumble upon the unanticipated use of surrogates. They found that surrogates entered 77% of e-scripts for 170 physicians at 47 medical practices (Barich, 2007). While surrogacy was not the intention of system designers, ‘changes in process directly affect organiza- tional form’ (Crowston, 1997, p. 159). The e-prescribing pilot study final report noted that ‘prescribers’ staff played a much more important role in the e-prescribing process than most pilot sites had anticipated’ (Moiduddin et al, 2007, p. ix).

Differences between these models are identified in step 3 using the lens of coordination theory (step 4). As noted earlier, there are several reasons manual prescribers don’t write generic prescriptions. However, e-prescribing pushes the e-prescriber to choose the lowest cost option (e.g., generic) with some systems providing drug pricing information. The researcher flags this task as a ‘misfit’ since they already uncovered (in step 1) that manual prescribers don’t have enough time and would have uncovered in step 2 that e-prescribing likely takes more time (e.g., Devine et al, 2010).

The study sought ‘specific implications’ from this exemplar case (Yin, 1989; Walsham, 1995), rather than generalizations for theory development. Processual ana- lysis directs the researcher to view context and action (Pettigrew, 1997; Azad & King, 2008). The researcher seeks to find changes in coordination that may explain unintended usage or conflicts in roles and relationships. For this study, specific implementations are macro-level decisions of an operating model that can be linked to aligned (or not) workflow. These revelations then lead to explicit research questions for the future (step 6) now that the boundaries for inquiry have been narrowed. These questions can then be addressed by more conven- tional retrospective research designs on representative members of the network (step 7).

Addressing validity A number of safeguards were adapted for this compara- tive synthesis study of a nation-wide e-prescribing net- work. As an interpretive study, the goal was confirmable findings using an internally consistent approach (Gasson, 2003). The reliance on peer-reviewed articles, many with large samples, means other researchers can see the same data even if they will interpret them through their own

Synthetic Model of Existing

Workflow

Medicine

Pharmacy

Medical Informatics

E-prescribing

Health Policy

Coordination Theory

Synthesis of Domain Literature

Add Emerging Literature

Differences Between Models

Specific Implications

Future Research

Synthetic Model of Intended Workflow

Synthetic Analysis

Body of Literature

StepsOutput

Select Lense

Retrospective Study 7

Figure 3 Synthetic comparative framework.

Governance in e-prescribing Nelson King 555

European Journal of Information Systems

topical and theoretical lenses (Sandelowski et al, 2007). Empirical studies, typical in the healthcare informatics literature, focus almost solely on the instantiated artifact and contain limited context (Lomas, 2005; Greenhalgh & Russell, 2006; Niazkhani et al, 2009). To provide context for these studies, additional informants were deliberately sought. Rather than rely upon unreliable retrospec- tive self-reports of individuals about their experiences (Bernard et al, 1984), the workflow models were built almost entirely upon virtual informants (i.e., peer- reviewed articles).

While there were some interviews in 2007–2008, these were only used to identify additional topics to be investigated using peer-reviewed informants. For exam- ple, the sales and product development managers for a pharmacy management vendor didn’t see any interest from their customers (i.e., independent pharmacists) so were reluctant to invest in a software upgrade for e-prescribing. A pharmacy professor confirmed that trans- actions fees were a barrier to independent pharmacies. The owner of an independent pharmacy pointed out that it was more than transaction fees since a software upgrade is expensive and likely requires new computer equipment. Several pharmacists pointed out that speak- ing to the physician was quite rare. It was important for them to know who they were speaking with (i.e., a clinician – not receptionist) – not just a computer message. One pharmacy provided a list of the small per- centage of physicians that printed computer-generated prescriptions. These early interviews suggested that the idealized workflow and adoption context found in the documentation of e-prescribing advocates might not reflect what is done in practice. For example, e-prescribing design documents presumed that an e-script sent by an e-prescriber will be immediately processed by a pharmacy. Yet a pharmacist working within a supermarket chain (store) pharmacy said ‘[I] don’t want to process [an e-script] if I don’t know them because then they’ll [pharmacy technicians] have to reverse’ and ‘often the [e-]script is for a nearby store [of the same chain]’. Such an insight prompted a search for articles about the choices exercised by a patient in filling prescriptions (e.g., Kinnaird et al, 2003).

The workflow equivalent of theoretical saturation (Glaser & Strauss, 1967; Gasson, 2003) was adopted so that each topic had sufficient informants. Keyword searches through electronic databases are known to be insufficient since alternative terms may be used for the same topic (Sandelowski et al, 2007). The previously described backward and forward snowballing technique avoided reliance on sampling the literature (Rousseau et al, 2008). Over 400 published resources were used with 72% from academic journals (breakdown by medication management topics found in Appendix A).

A synthesis approach for data collection may not reach the gold standard for either qualitative or quantitative research, but collectively creates a workable understand- ing for a longitudinal exploratory study of a nation-wide network prior to mass adoption. The study recognizes

that coordination networks ‘span functional boundaries, organizational boundaries and levels of analysis, and academic research should reflect these complexities’ (Gittell & Weiss, 2004, p. 148). The results of this comparative analysis (step 3) of existing and intended workflow (steps 1 and 2) are described in the following sections at both the macro (operating model) and micro- level (workflow). The findings section titled ‘Operating Model Implications’ addresses specific implications (step 5).

E-prescribing: macro-description A reduction in medication errors and lower costs has been the promise of e-prescribing. The Institute of Medicine report, ‘To Err is Human’ (Kohn et al, 1999), pointed to studies showing most medication errors could be prevented with information systems that ‘disseminate knowledge about drugs and make drug and patient information readily accessible’ (p. 40). E-prescribing attempts to computer- mediate what was a predominantly manual process between outpatient prescribers, patients, payers, and pharmacies. The outpatient prescribing network in the United States dispensed 3.9 billion prescriptions in 2009 from a net- work that comprises 600,000þoffice-based prescribers, 62,000þpharmacies, 300þmillion patients, and tens of thousands of payers (Surescripts, 2010). E-prescribing adds to this network the infomediary (network broker) and hundreds of vendors that sell standards-compliant software to medical practices and pharmacies. Appendix B summarizes the upper limits of network membership.

The cost reduction impetus for e-prescribing has been primarily third-party payers who represent state or federal governments, insurance companies, or PBMs (SureScripts, 2006). Over 85% of outpatient prescriptions filled in community pharmacies involve a third-party payer (Cardinal Health, 2008) such that increasing generic utilization among e-prescribers has the potential for reducing costs to payers. The Centers for Medicare & Medicaid Services (CMS), the dominant public payer in the United States with a 27% share of $217 billion spent nation-wide on retail prescription drugs in 2006 (Catlin et al, 2008), mandated e-prescribing to drive down their drug expenditures.

Going beyond the policy and economic rationale for e- prescribing, the remainder of this macro-description section describes the operating model and associated network for manual prescribing followed by e-prescrib- ing. The micro-description section follows with a task- oriented tracing of the prescribing workflow for both the manual and e-prescribing networks. Table 2 summarizes the two operating models.

Manual prescribing operating model The operative phrase guiding manual prescribing is the ‘pharmacist as final checkpoint’. Brown et al (2006) concluded from their focus groups that:

The ambulatory pharmacist is the common link between

the physician and the patient. y As the dispenser of

Governance in e-prescribing Nelson King556

European Journal of Information Systems

prescribed medication, the ambulatory pharmacist is in a

unique position of detecting both patient-reported errors

and errors made by the healthcare provider. The pharmacist

is the final checkpoint in preventing an error from reaching

the patient. (p. 23)

Decoupling between its members characterizes the manual network. The prescriber hands the paper pre- scription to a patient signaling the termination of the visit (Flynn et al, 2003; Hunt et al, 2008). The pre- scriber has no direct involvement in filling a prescription except if a pharmacy cannot resolve the error with the prescriber’s medical staff (Hansen et al, 2006; Barich, 2007). Prescribers delegate the error-checking role to pharmacists who are reluctant to report problems for fear of creating a poor working relationship (Brodsho, 2005; Brown et al, 2006). The pharmacy adjudicates a medica- tion claim with a PBM using long-established electronic connections. The micro-findings section provides a more detailed task-oriented description of this manual workflow.

E-prescribing operating model Process consolidation at the point of prescribing drives e-prescribing. The eHealth Initiative (Teich et al, 2004) states:

y built-in error checking ensures that the primary

prescription inspection point is moved earlier in the process

– specifically, to the prescriber at the point of prescribing –

and lessens dependence on later review in the pharmacy.

This change in approach represents a fundamental overhaul

to our national prescription error prevention system y. (p. 28)

The new operating model is in stark contrast to what has existed before when the pharmacist serves as ‘interceptor, detector, and reporter of medication errors to the physician’ (Brown et al, 2006, p. 22).

E-prescribing spreads the e-script order transaction across different members of the network (Figure 4) includ- ing the medical practice, pharmacy, and patient. Two network integration infomediaries (pre-merger names used in this article), now jointly run by Surescripts after their 2008 merger, are parts of this nation-wide network: National Patient Health Information Network (NPHIN) and the Pharmacy Health Information Exchange (PHIE)

(Bell et al, 2004; King et al, 2007). These infomediaries provide the ‘integration’ called for by Ross et al (2006) in their characterization of the operating model for govern- ance. The PHIE transmits routine prescription informa- tion between pharmacists and e-prescribers utilizing certified e-prescribing or electronic medical record soft- ware applications (shown previously in Figure 1). The NPHIN connects clinicians to patient histories and payer formularies that are unique to each patient to permit point-of-prescribing error checking.

Ross et al (2006) also call for ‘standardization’ of business processes and related systems. E-prescribing standards currently in use have roots in the process models of early vendor systems (Bell et al, 2004) and expert panel recommendations (Wang et al, 2005). Interoperability to support electronic prescribing requires standards (Hammond, 2004; Teich et al, 2005; U.S. National Library of Medicine, 2005). These government- endorsed standards govern information exchange for formulary and benefit information, exchange of medica- tion history, fill status notification, patient instructions, drug terminology, and prior authorization (Leavitt, 2007; Friedman et al, 2009). Surescripts also certifies that a vendor’s software application, over a hundred in 2006,

Table 2 Operating model comparison

Attribute Manual prescribing E-prescribing

Operative Phrase Pharmacist final checkpoint in preventing error Point-of-prescribing primary inspection point

Goals (in practice) Minimize impact on prescriber time

(see micro-findings)

Increase generic utilization (lower cost), reduce errors on e-script

(limited outcomes to date on intended use)

Standardization Adjudication message via long-established

pharmacy-payer proprietary links

Certification of standards-based e-prescribing systems to provide

common functionality excluding pre-existing pharmacy-payer links

Integration Between pharmacy-payer for

adjudication only

Between e-prescribers and payers for e-script generation, transmit

e-script, checking formulary; existing adjudication does not change

Manual Prescribing

Electronic Prescribing (intended)

Order Entry

Decide to Fill Receive

e-Handoff

Order Entry

Decide to Fill

Receive

Artifact

Handoff

Practice

Generate Script

PharmacyPatient

Patient Pharmacy

Deliver

Generate Script

Record Location

Deliver (e-send)

Practice

P A

Figure 4 Role and relationships – prescription generation.

Governance in e-prescribing Nelson King 557

European Journal of Information Systems

meets government standards for functionality (Wang et al, 2005; SureScripts, 2006; Leavitt, 2007).

Changes in network roles and relationships: micro-description Prescription generation and the formulary check, two processes of the e-prescribing workflow, are analyzed at the micro-level to identify mismatches that could limit alignment (e.g., adoption). Given the large number of network actors spread across the United States, ranging from micro-businesses to businesses with hundreds of employees (Anderson, 2007; SK&A Healthcare Informa- tion Solutions, 2007; Hing & Burt, 2008), the process described herein represents a typical one, not necessarily all the variation likely to be encountered in practice.

Prescription generation In manual prescribing, the paper prescription is handed off among network members (top of Figure 4). The prescriber in a medical practice writes a prescription [P] and passes a copy to her medical staff for updating the patient’s record [A] (Johnson & Fitzhenry, 2006). The prescriber passes the paper prescription to the patient member (e.g., patient or guardian) signaling the end of the office visit [B] (Hunt et al, 2008). There is little or no interaction between prescriber and patient regarding the prescription (Wilson et al, 2007; Khan et al, 2008). The patient first decides if the prescription will be filled [D] before choosing to deliver it [E] – typically to a local pharmacy (Linton et al, 2007).

In contrast, the e-prescriber inputs the e-script [P] and completes the record keeping online [A] before pushing the e-script electronically to the selected pharmacy [E] which is required by the network (bottom of Figure 4). The process repeats for each prescribed medication since each e-script contains only one medication order. The e-prescriber must interact with the patient to obtain a pharmacy location [C]. The intended design of e-prescribing appears to bypass the decision role of the patient [D] since the e-prescriber transmits the e-script directly to the pharmacy. Patients are known to choose among multiple dispensers (Linton et al, 2007), yet can no longer freely exercise that choice since an e-script must be sent directly by the e-prescriber. With an e-script, the pharmacy must then determine if the patient is going to pick up the medication as many do not [F] (Kirking et al, 2006).

Whether intended or not, e-prescribing alters many of the roles and relationships of manual prescribing (compare top and bottom of Figure 4). Many of these changes result from integration of information which couple together network members. For example, the e-prescriber now has a direct link with the pharmacy via the infomediary which represents an ‘e-handoff’ that did not exist in manual prescribing. This means the e-prescriber now takes on the role of designating a pharmacy which previously was a patient’s responsibility. E-prescribing also changes the role of the prescription

artifact. A paper prescription allowed information to be distributed around the network at a time chosen by the patient (Luff et al, 1992). An e-script, when pushed from prescriber to pharmacy as in the United States, limits such mobility. Delivery of a paper prescription signified to a pharmacy the patient’s intent to fill it [E]. In contrast, delivery of an e-script only indicates that one has been sent which is a notable change from the past.

Formulary check The adjudication of prescriptions has been computer- mediated for many years. Online pharmacy claims processing became mainstream in the 1990s beginning with separate modem-based adjudication terminals for different payers followed by integration directly into pharmacy management software (Sardinha, 1998; Romza & Black, 1999). This is the point-to-point linkage in manual prescribing to the activity ‘claim approval’ [N] in the top half of Figure 5. The pharmacy is the network member in manual prescribing who interacts with the payer (most often a benefit manager) during prescription adjudication. Adjudication requires the pharmacist to check patient eligibility [K], check the formulary that the prescribed medication is covered by the payer [L], and finally submit a claim for reimbursement [N] and collect any co-payment due.

Eligibility verification [K] is required for non-cash patients who receive a benefit from third-party payers (Arthur Andersen LLP, 1999). The patient’s eligibility for a prescription plan must be verified [N] before the choice of drug can be checked for adherence with the formulary for that patient. This computer-mediated routine is per- formed every time a prescription involves a third-party payer. The pharmacy must call the payer whenever a problem cannot be resolved (e.g., patient name known but not of policyholder). Pharmacists spend 7% of their day resolving third-party (payer) eligibility issues (Arthur Andersen LLP, 1999).

Manual Adjudication

E-Prescribing Pre- Adjudication (intended)

Check Formulary

Verify Eligibility

Dispense

Pharmacy

Claim Approval

Benefit Manager

Check Formulary

Verify Eligibility

Dispense

Pharmacy

Claim Approval

Benefit Manager

Check Formulary

Verify Eligibility

Practice

Pre-Claim Status

Benefit Manager via Surescripts

L L

Figure 5 Role and relationships – formulary check.

Governance in e-prescribing Nelson King558

European Journal of Information Systems

Once the verification is received, the formulary for a particular patient becomes available. The prescribed drug is compared against those drugs eligible for that patient [L]. The payer allows only certain drugs to be on the list. To assist in formulary adherence, the prescription artifact in many U.S. states has a tick box saying ‘dispense brand only’. This protocol permits a pharmacist to substitute a lower cost drug on the formulary (e.g., generic) when such drugs are available without consulting the prescriber (Ross et al, 2005). Pharmacists already change, when a state allows such substitutions, 84% of prescriptions written for brand to generic (Mott & Cline, 2002). This substitution rate is likely the upper limit since pharma- cists will not change some prescriptions to generic for various medical and professional reasons (Al-Gedadi & Hassali, 2008).

If not on the formulary, the pharmacist substitutes an equivalent medication or contacts the prescriber or the medical staff who acts at their behest (Suh, 1999). Pharmacists will often make the changes at their own discretion and the prescriber is generally contacted as a last resort (Chen et al, 2005; Brown et al, 2006). Once approved, the pharmacy then proceeds to dispense the medication. The substitution of a generic by pharmacists without getting explicit approval is an exception to the norm that prescribers do not allow independent pharma- cist action (Ritchey & Raney, 1981).

The formulary check and resolving of issues related to the formulary check consume another 14% of the pharmacist’s day (Arthur Andersen LLP, 1999). ‘Today, pharmacists spend much of their time checking formu- laries, filling out insurance forms and waiting on hold for some untrained, unlicensed functionary from an HMO’, said Mark Griffin, president and chief executive of Lewis Drug and former chairman of the National Association of Chain Drug Stores quoted in Frederick (2003).

E-prescribing at the point of prescribing changes several adjudication roles and relationships (lower half Figure 5). E-prescribing couples the e-prescriber and payer through the need for conducting both an eligibility check [K] and a formulary check [L]. The statuses from these checks made by the e-prescriber do not get passed to the pharmacy which must make the same checks during adjudication. This coupling adds additional tasks to the e-prescriber that are not directly compensated. In addi- tion, any eligibility issues must be addressed by the e-prescriber which in the past was done by the pharmacy.

The e-prescriber is now coupled to the patient. In manual prescribing, there is little or no interaction between prescriber and patient with regards to eligibility for certain kinds of drugs. Most prescribers had limited or no access to a patient’s formulary information (Shrank et al, 2006). E-prescribing presumes (desires) that an e-prescriber takes advantage of a patient’s formulary information to choose a lower-cost drug therapy [L]. Physicians have traditionally had few discussions about costs with patients and those rare patient-initiated discussions lasted for less than 10 s (Tarn et al, 2006;

Beran et al, 2007; Tarn et al, 2008). Yet talking about cost of treatment with patients opens up a new role that prescribers may have tried to avoid in the past since there is no standard for a cost-effective drug therapy (Malone, 2005). Given the disparity in medication co-pay costs (e.g., generic may be free or upwards of $25 depending upon formulary), the differential between brand and generic may be the basis for which a patient is willing to accept generic, not the advice of a physician (Mager & Cox, 2007). The net result of satisfying the intent of e- prescribing is additional time spent with the patient.

Net impact The demarcation of roles in manual prescribing meant that the relationships were clearly defined and limited any coupling. The roles and handoffs developed over the years had minimized interaction with the prescriber. Now that the roles have been altered with e-prescribing, old relationships have changed and new relationships cre- ated. For example, the prescriber in manual prescribing rarely interacted with the pharmacist. Even with a callback from the pharmacy, the medical office staff shielded the prescriber from most calls. With e-prescribing, the authorized sender (i.e., e-prescriber) receives all the messages for clarification, not just the ones that the medical staff could not handle on their own. Practices have had to implement protocols to allow office staff to handle e-renewal requests (Crosson et al, 2011).

A second example is interaction with the patient. Handing the paper prescription to the patient signaled not only the end of the office visit but also turning over responsibility for transmitting the prescription to the patient. The patient had a choice over whether the prescription would be filled or not. E-prescribing now changes the relationship between the e-prescriber and the patient to a more interactive one that includes discussion of drug choice (e.g., generic or brand) and where the prescription is to be filled. From the patient perspective, there is now a necessity to choose a phar- macy location before leaving the examination room, leaving no time to think about whether the prescription will even be filled or not.

E-prescribing has been implemented as intended: pro- cess consolidation at the point of prescribing (Teich et al, 2004). The intended consolidation of roles and tasks to the e-prescriber appears to have had unintended ramifi- cations. Does the e-prescriber want to pay for the privilege of adopting a system that forces them to do more? Does the pharmacy want to pay a transaction fee that can increase costs by at least 3% (Thornton, 2007) for an e-script that introduces uncertainty to their process? The drug-drug interaction (DDI) check process, not discussed in this paper due to space limitations, fur- ther confounds the role boundaries between e-prescriber and pharmacist thereby increasing tensions (Lewis et al, 2010). In sum, the changes in roles and relation- ships may be a source of reluctance in the adoption of e-prescribing.

Governance in e-prescribing Nelson King 559

European Journal of Information Systems

Implications of operating model choice The micro-findings shows a discrepancy between exist- ing prescribing practices and the intended workflow embedded in the e-prescribing software dictated by the operating model. This section shows that these discre- pancies are probably not by chance. An operating model offers more than a guiding vision, but at least implicitly reflects infrastructure design choices. The following exam- ples show how choices made in the operating model do in fact ripple directly to the tasks of network actors thereby impacting their workflow.

It is a high-level integration decision to push or pull a message transaction through a network. The choice to push the e-script to a pharmacy has numerous ramifica- tions. First, a push model that acts as a pass-through (e.g., relay the e-script) requires the e-prescriber to track down all information necessary to transmit each and every e-script during the office visit. This means the patient has little say on the decision to fill a prescription, which in turn introduces uncertainty into the status of an e-script received at the pharmacy. The push model also means that the e-prescriber can no longer efficiently delegate their additional tasks to other professionals (e.g., medical staff or pharmacist) as in the past. The medical practices that have adopted e-prescribing become tightly coupled with patients and pharmacies.

The operating model could be changed to better align itself with the existing roles and relationships familiar to network members. First, the infomediary could allow a post-visit ‘push’ from a patient or ‘pull’ from a pharmacy. The former would give tech-savvy patients, once they’ve decided to fill an e-script, the chance to ‘shop around’ for the best price as they do now. The latter would allow patients to tell their pharmacy to retrieve an e-script. In both cases, the pharmacy has some assurance that the patient intends to fill the e-script. The IT consequences are the need to tag the e-script with the status of various events (e.g., waiting to send) and hold (not relay) it. Such a choice increases centralization of the broker and requires sufficient infrastructure capacity to store mes- sages for subsequent retrieval.

The second example of operating model choice points to the lack of message status accompanying the e-script. Did the e-prescriber update eligibility? Did the e-prescriber check the formulary? Did the e-prescriber check for DDI? The recipient of the e-script, a pharmacy, has no means to determine that these checks were done with the current infrastructure.

Once the e-script is capable of being tagged with additional information, a design choice, then other information could be affixed to the e-script. These would include whether (a) eligibility verified (and for how long), (b) formulary checked by e-prescriber, and (c) details of DDI checked by e-prescriber. If checks (a) and (b) were recognized by the adjudication software, the pharmacy could avoid repeating this step. While helping pharmacy efficiency, e-prescribers may demand compensation for helping others in the network.

The final example speaks to the fundamental change in operating philosophy of e-prescribing – the responsibility for the safety check that moves from the pharmacist at the point of dispensing to the e-prescriber during the office visit. There are numerous questions to be explored, including the willingness of the e-prescriber to take on these responsibilities. If no, the safety check reverts back to the pharmacist who at the moment is legally obligated to do so. So in practice, the efforts of an e-prescriber who takes the time to check for DDI are currently duplicated since the pharmacist can’t be sure if the check was done. Ideally, the e-script should include a report on the DDI check that lists the alerts that had been ignored or overridden by the e-prescriber. The pharmacist would now be in a position to do a value-added quality control check to augment the one done by the e-prescriber. At the moment, the pharmacist has to assume no DDI check has been done.

Discussion The U.S. ambulatory e-prescribing network has been touted as a fundamental overhaul to medication error prevention (Teich et al, 2004). Traditional IT best prac- tices were employed to build the e-prescribing inter- mediary, such as inter-operability standards and certifica- tion of software vendors. Policy-makers also introduced incentives and penalties to encourage the adoption of e-prescribing. Adoption should have been rapid as physicians embraced the idea of error prevention (Lipton et al, 2003; Pizzi et al, 2005). Despite adhering to these best practices, e-prescribing adoption has been slow (Wang et al, 2009). For example, one in four physicians did not send 10 e-scripts by mid-2011 despite the financial penalties being imposed on them for non- adoption (Centers for Medicare & Medicaid Services, 2011). The explanations for slow e-prescribing adoption have been typical of healthcare IT initiatives in general: immature standards, lack of stakeholder involvement, and poor implementation practices (e.g., Friedman et al, 2009).

While great diligence was paid to developing and testing standards (Moiduddin et al, 2007), these can only reflect the choices made in the operating model, which dictates the workflow to be embedded in standards. Policy-makers asked for a standards-based system that represented a ‘fundamental overhaul to error preven- tion’. The e-prescribing infomediary (i.e., Surescripts) implemented this policy through standards that pushes forward error prevention to the point of prescribing. Yet perception of standards depends on organizational con- text (Kayworth & Sambamurthy, 2000). The findings show that workflow expectations of e-prescribing network members are inconsistent with the implemented operat- ing model. Network members have not overcome the tension between forces promoting and opposing change (Robey et al, 2002). Physicians support e-prescribing but are unwilling to do so without compensation for addi- tional time spent on their part (Lipton et al, 2003). Training

Governance in e-prescribing Nelson King560

European Journal of Information Systems

on software procedures are insufficient without addres- sing broader issues of process change (Robey et al, 2002). In sum, people in organizations may reject technology they are in favor of (Leonardi, 2009).

The comparative analysis revealed changes in roles and relationships which trace back to choices made for the e-prescribing operating model. The study points out that sufficient evidence already existed in the medication management literature to warn of a fundamental change in the way organizations and work are structured. Yet these policy-makers and system designers did not appear to consider the social aspects known to plague healthcare systems (Greenhalgh et al, 2009; Niazkhani et al, 2009). The discomfort of network members towards their new roles and relationships may in fact be a root cause for slow adoption due to the unexpected perturbation of the existing manual network (Knoben et al, 2006).

Nested operating models Applying the operating model conceptualization to net- work ITG offers a distinct advantage. First, networks must consider a broader set of stakeholders. Network ITG must deal with multi-sourcing or a ‘co-operative’ approach to governance where network failure rates exceed 50% (Lee, 2009). While the typical enterprise may have to incorporate government policy regulations, the policy- maker typically doesn’t specify the way the system operates as has been done in e-prescribing. While this study treated the operating model for the e-prescribing network as a singular entity, a network operating model may represent just one layer of the network.

A network operating model may in fact be a set of derived operating models. In object-oriented program- ming, a derived object inherits attributes and behavior from pre-existing objects. A similar metaphor is the nested doll principle (i.e., Russian nested dolls) where objects encapsulated within a similar object retain its general characteristics but are unique in its own ways.

Using the case description, the basis for a nested set of operating models becomes apparent. Policy-makers set the tone for how the network will operate, such as error prevention at the point of prescribing. It is the role of the infomediary (network broker) to further articulate the network operating philosophy into an infrastructure.

Standards must be established, oftentimes with the direct involvement of the policy-makers. The infomediary uses the operating model in much the same way as an enterprise. Many vendors, selling applications to network members, consider e-prescribing as a module that they incorporate into their existing software. Vendors must now interpret the standards in the context of their own applications (Bendoly et al, 2007). The attributes inherited from the operating model of e-prescribing (i.e., macro-level) could of course be at odds with the operating model for the vendor’s application. Standards, unless strictly stipulated, are interpreted in a way that best suits the vendor (Wareham et al, 2005). Finally, network members must choose to implement a stan- dards-based e-prescribing application obtained from a certified vendor. Once again, each member must decide the extent to which they align their practices with the network-level operating model. The medical practice could choose the simplest system and use it only to send 10 e-scripts to qualify for the incentives. Other members might choose to transform their medical practices to embrace electronic medical records with an e-prescribing module. Table 3 summarizes the different levels of operating models in the network.

Early articulation of the operating model, at various levels of the network, provides the opportunity to reconcile disparate views of what the proposed standards embody. E-prescribing advocates recognize that gaining the expected benefits for the U.S. healthcare system depends upon all parties pulling together, despite the costs some may incur (Teich et al, 2004). Policy-makers, infomediary, vendors, and network members all have different interests and perception of the impact that e- prescribing has on their own work practices (King, 2011). Second, traditional control mechanisms used for enter- prise ITG are difficult to apply to a network since there is no single organization. Membership of the e-prescribing network ranges from solo physician medical practices to drugstore (pharmacy) chains. Each member has varying levels of ITG capabilities that may or may not be able to respond to the demands of policy-makers and the info- mediary. Networked systems must therefore rely upon standards because they represent one of the few points of leverage over network members (e.g., Kurimoto, 2008).

Table 3 Nesting of operating models

Level Design emphasis Implementation

Network Operating philosophy of network; Policy-regulations ‘point-of-prescribing’

Infomediary Architect an infrastructure (standardization-integration);

Set standards

Push e-script; Extent of standards stipulation;

Certification of vendors

Vendor Interpret-implement standards within operating model

of own applications

Add e-prescribing module

Member Configure standards-based application to own practices Varies: standalone to complete change of practices

Governance in e-prescribing Nelson King 561

European Journal of Information Systems

Thus the operating model plays an even greater role in network ITG since traditional ITG mechanisms may not be applicable across the network.

Contribution An understanding of the linkage between strategy (or policy) and workflow of the U.S. e-prescribing network was gained through the operating model con- ceptualization of Ross et al (2006). The existence of a direct pathway between strategy and low-level network processes (e.g., roles and relationships) suggests that the enterprise-centric ITG concept of operating model can be carried over to the governance of networks for at least those configured as an NAO. The comparative analysis (Figures 4 and 5) shows how a change in process at a policy level directly affects organizational coordination within a network (Crowston, 1997). Such a finding adds to the body of literature that reports on the impact of operating model alignment to improve enterprise perfor- mance (Ross et al, 2006; Perko, 2008; Singh & Woo, 2009; de Vries et al, 2010).

This study contributes the notion of nested operating models that govern various levels of the network. Each level of operating model derives its core characteristics from higher level operating models. The notion of nesting helps explain the variation in practices adopted at lower levels of the network despite the integrated infrastructure, standards, and vendor certification pro- vided by the U.S. e-prescribing infomediary. Given the expanse of most networks and heterogeneity of member ITG capabilities, governing the nested operating models may offer the most leverage to propagate network policy and strategy throughout the network. The network broker may also virtually engage its members earlier in system development as the transparency of the operating model allows these stakeholders to assess the impact on their own workflow. The operating model concept shifts the focus of inter-organizational ITG from how it is done (e.g., form of organizational structure) to what is done operationally – the impact of macro-level choices on workflow.

Future research The importance of the operating model in network governance suggests several research directions. While the impact from the operating model choices made for this case study are likely network-specific, the methodo- logical approach of examining operating model choices can be applied to the specific context of other networks. First, the role of operating model could be explored in other e-prescribing networks. These are typically national entities so a multi-country study would be necessary that includes Finland (e.g., Salmivalli, 2008) and the United Kingdom (e.g., Schade et al, 2006) where a body of research already exists. These networks may make different choices in their operating models (e.g., pull from national database in Finland) to reflect the way healthcare is organized in their countries. In addition, the

behavior of operating model in enterprises (e.g., Ross et al, 2006; de Vries et al, 2010) could be compared with networks to examine differences in the locus of decision- making, approaches to strategic planning and degree of organizational coupling. The third direction would be exploring the broker’s allocation of benefits among network members (Tapscott et al, 2000). These allocations relate to the value added by network services and the cost to create such value. A service value network framework may provide insights on these allocations (Peterson, 2004; Basole & Rouse, 2008). Finally, current exchanges of health information have reached a point of diminish- ing return that necessitate an expanded scope for trans- actions (Brantes et al, 2007). As suggested in the findings, the additional complexity of task status (e.g., formulary check) may need to be passed within an e-script to generate value within the network. This expanded scope deviates from the intended operational philosophy of ‘point of prescribing’ (for e-prescribing) to a more distributed approach found in manual prescribing.

Limitations This exemplary case shows only the linkage between choices embedded in the e-prescribing operating model and the impact on network roles and relationships for a single network. Further work would be necessary to demonstrate that this linkage exists in other network configurations. In addition, the impact of dependencies upon coordination must be considered (Crowston, 1997). Very little collaborative coordination takes place in an essentially sequential e-prescribing transaction compared with an electronic health record.

The challenge for this study was analyzing multi-level network phenomena even though network governance research is in its infancy (Salmivalli et al, 2008; Lee, 2009; Grant & Ulbrich, 2010). The synthetic method of data collection responds to the lament of Rousseau et al (2008, p. 477): ‘y the underuse of research evidence with substantive implications for understanding and work- ing with organizations’. This study also responded to the call for longitudinal studies of networks undergoing a fundamental change (Knoben et al, 2006; Hyysalo, 2010). However, readers may ask whether comparative synthesis was the most appropriate approach for this study.

The intent of the original data collection was to inform decision-makers as e-prescribing moved towards mass adoption in 2007. Four years later, one of the first post- pilot phase studies that addresses some of the tasks examined in this synthesis was published by Grossman et al (2011). Their study was government-funded and this author provided feedback to their proposed protocol during the public comments period (September 2009). This retrospective study interviewed only 75 members (0.03% of pharmacies and medical practices in the network) and confirmed many of the assertions derived from synthesis (eg, pharmacies reluctant to pay transmis- sion fee). Their pinpoint study provided many insights

Governance in e-prescribing Nelson King562

European Journal of Information Systems

but did not fully develop the context that surrounds a whole network analysis (see comparison Table 4).

In the parlance of the synthesis method used in this paper, the Grossman et al (2011) study acts as an informant, just one more piece to build a story of the network. Many more retrospective studies would have to be conducted by these researchers to match the breadth of this synthesis. Since their telephone interviews were already 45 min in length, asking healthcare professionals to spend additional time answering questions about other tasks could be problematic. Even if lengthier inter- views were possible with a sufficient number of infor- mants to address the heterogeneity of the network, there still would be a need for synthesis. So fundamentally, the question comes down to whose data does one synthesize: your own or that of others.

Concluding remarks Inter-organizational networks, especially on the scale of U.S. e-prescribing, introduce a governance dilemma. The network broker (operator of infomediary), implementing the decisions of a policy-maker, appeals to its members to adopt the technology for the ‘common good’ (i.e., reduced medication errors). Yet its members who operate with a degree of autonomy, like operating units in an enterprise, seek some benefit from technology adoption. The policy-maker may also be far removed from the

actual practices of its network members. Thus choices for the operating model are based on incomplete informa- tion which arguably inhibits adoption due to a lack of workflow alignment.

The nested operating model, as the locus of govern- ance, departs from the traditional focus of ITG frame- works that rely upon form and contingency analysis for IT decision-making. The operating model provides trans- parency into workflow alignment which provides the network broker a means to engage its network members. The network governs through the choices embedded in the operating model, rather than an ITG organizational form, since a network broker has few means to control what network members actually do. While an exploratory study, this study offers one window into understanding inter-organizational ITG relationships.

Acknowledgements An early version of this paper, co-authored with Bijan Azad,

was published in the Proceedings of the SIGPrag Workshop

held at ICIS 2010 in Saint Louis. The guidance of the special issue editors was instrumental, along with the feedback of

Bijan Azad and anonymous reviewers. The research was

partially supported by the University Research Board of the American University of Beirut and the IBM Center for the

Business of Government.

About the author

Nelson King is an associate professor at the Olayan School of Business – American University of Beirut. His research interests are in networked information systems, especially inter-organizational collaboration in healthcare. He ob- tained his Ph.D. in industrial and systems engineering from the University of Southern California where he also did post-doctoral research in imaging informatics. Before joining academia, he spent over 20 years as a systems engineer. Nelson’s healthcare work has been published in

journals such as the European Journal of Information Systems, Communications of the Association for Information Systems, E-Service Journal, and International Journal of Organizational Design and Engineering. Some of his earlier work has been published in MIS Quarterly and IEEE Transactions on Engineering Management. He serves as associate editor for Communications of the Association for Information Systems, Health Systems, Information Systems Management, and Inter- national Journal of Organisational Design and Engineering.

Table 4 Synthesis vs traditional methods

This study Grossman et al (2011)

Method Prospective comparative synthesis Retrospective qualitative

Network coverage Tasks: Generation, Transmission,

Adjudication for whole network

Transmission between 75 medical practices

and pharmacies

Sample Not quantified (see Table 1 for examples): mixture

of quantitative and qualitative research from

400+ articles collectively involving tens of thousands

of informants and hundreds of organizations

114 telephone interviews including 24 physician

practices, 48 community pharmacies, and three

mail-order pharmacies

Research focus Network roles and relationships – downstream

consequences of policies

Facilitators and barriers of technology use

Governance in e-prescribing Nelson King 563

European Journal of Information Systems

References AKHLAGHPOUR S (2008) Impacts of social embeddedness on IOS plan-

ning effectiveness: a multilevel perspective. Annual Conference of the Administrative Sciences Association of Canada 29(4), 196–216.

AL-GEDADI NA and HASSALI MA (2008) Pharmacists views on generic medicines: a review of the literature. Journal of Generic Medicines 5(3), 209–218.

ALEXANDER GC, CASALINO LP and MELTZER DO (2005) Physician strategies to reduce patients’ out-of-pocket prescription costs. Archives of Internal Medicine 165(6), 633–636.

ANDERSON MR (2007) AC Group’s 2007 Annual Report: The Digital Medical Office of the Future Computer Systems for the Physician’s Office. AC Group, Houston, TX.

ARTHUR ANDERSEN LLP: (1999). Pharmacy Activity Cost and Producti- vity Study. National Association of Chain Drug Stores Education Foundation, Chicago, IL.

AZAD B and KING N (2008) Enacting computer workarounds within a medication dispensing system. European Journal of Information Systems 17(3), 264–278.

BARICH DP (2007) Pilot Testing of Electronic Prescribing Standards. Centers for Medicare & Medicaid Services, Washington, DC.

BARRINGER BR and HARRISON JS (2000) Walking a tightrope: creating value through interorganizational relationships. Journal of Management 26(3), 367–403.

BARUA A, KRIEBEL CH and MUKHOPADHYAY T (1995) Information technol- ogies and business value: an analytic and empirical investigation. Information Systems Research 6(1), 3–23.

BASOLE RC and ROUSE WB (2008) Complexity of service value networks: conceptualization and empirical investigation. IBM Systems Journal 47(1), 53–70.

BELL DS, CRETIN S, MARKEN RS and LANDMAN AB (2004) A conceptual framework for evaluating outpatient electronic prescribing systems based on their functional capabilities. Journal of American Medical Informatics Association 11(1), 60–70.

BELL DS et al (2007) Pilot Testing of Electronic Prescribing Standards. Agency for Healthcare Research and Quality, Washington, DC.

BENDOLY E, CITURS A and KONSYNSKI B (2007) Internal infrastructural impacts on RFID perceptions and commitment: knowledge, opera- tional procedures, and information-processing standards. Decision Sciences 38(3), 423–449.

BERAN MS, LAOURI M, SUTTORP M and BROOK R (2007) Medication costs: the role physicians play with their senior patients. Journal of the American Geriatrics Society 55(1), 102–107.

BERNARD HR, KILLWORTH P, KRONENFELD D and SAILER L (1984) The problem of informant accuracy: the validity of retrospective data. Annual Review of Anthropology 13, 495–517.

BORMAN M and ULBRICH F (2011) Managing dependencies in inter- organizational collaboration: the case of shared services for Application hosting collaboration in Australia. In 2011 44th Hawaii Inter- national Conference on System Sciences (SPRAGUE RH, JR, Ed), pp 1–10, 4–7 January 2011.

BRANTES FD, EMERY DW, OVERHAGE JM, GLASER J and MARCHIBRODA J (2007) The potential of hies as infomediaries. Journal of Healthcare Information Management 21(1), 69–75.

BRODSHO KC (2005) Patient expectations and access to prescription medication are threatened by pharmacist conscience clauses. Minnesota Journal of Law, Science & Technology 7(1), 327–336.

BROWN AE and GRANT GG (2005) Framing the frameworks: a review of IT governance research. Communications of the Association for Information Systems 15(1), 696–712.

BROWN C, BAILEY J, LEE J, GARRETT P and RUDMAN W (2006) The pharmacist- physician relationship in the detection of ambulatory medication errors. American Journal of the Medical Sciences 331(1), 22–24.

BUCKL S, MATTHES F and SCHWEDA CM (2011) Socio-technical dependency and rationale models for the enterprise architecture management function. In Advanced Information Systems Engineering Workshops (SALINESI C and PASTOR O, Eds), pp 528–540, Springer, Berlin, Heidelberg.

BYGSTAD B and HANSETH O (2010) IT governance through regulatory modalities. Health care information infrastructure and the ‘blue fox’ project. In Scandinavian Information Systems Research (Kautz K and Nielsen PA, Eds), pp 0–64, Springer, Berlin, Heidelberg.

CAMERON B, ORLOV LM and BRIGHT S (2005) Morphing IT Governance from Function to Process. Forrester Research, Cambridge, MA.

CARDINAL HEALTH (2008) 2008 NCPA Digest: Benchmarking Analysis. National Community Pharmacists Association, Alexandria, Virginia.

CATLIN A, COWAN C, HARTMAN M and HEFFLER S and THE NATIONAL HEALTH EXPENDITURE ACCOUNTS T (2008) National health spending in 2006: a year of change for prescription drugs. Health Affairs 27(1), 14–29.

CENTERS FOR MEDICARE & MEDICAID SERVICES (2011) Medicare Program; Proposed Changes to the Electronic Prescribing (erx) Incentive. Depart- ment of Health and Human Services, Washington, DC.

CHEN Y-F, NEIL KE, AVERY AJ, DEWEY ME and JOHNSON C (2005) Prescribing errors and other problems reported by community pharmacists. Therapeutics and Clinical Risk Management 1(4), 333–342.

CHO S, MATHIASSEN L and ROBEY D (2007) Dialectics of resilience: a multi-level analysis of a telehealth innovation. Journal of Information Technology 22(1), 24–24.

COMMISSION ON SYSTEMIC INTEROPERABILITY (2005) Ending the Document Game. U.S. National Library of Medicine, Washington, DC.

CROSSON JC, ETZ RS, WU S, STRAUS SG, EISENMAN D and BELL DS (2011) Meaningful use of electronic prescribing in 5 exemplar primary care practices. Annals of Family Medicine 9(5), 392–397.

CROTEAU AM and BERGERON F (2009) Interorganizational governance of information technology. In 42nd Hawaii International Conference on System Sciences (HICSS) (SPRAGUE RH, JR, Ed), pp 1–8, 5–8 January, Waikoloa, Big Island, Hawaii, IEEE Computer Society.

CROTEAU AM and DUBSKY J (2011) Uncovering modes of interorganiza- tional governance of IT. In Hawaii International Conference on System Sciences (SPRAGUE RH, JR, Ed), pp 1–10, 4–7 January 2011.

CROWSTON K (1997) A coordination theory approach to organizational process design. Organization Science 8(2), 157–175.

CURTIS B, KELLNER MI and OVER J (1992) Process modeling. Communica- tions of the ACM 35(9), 75–90.

DAINTY KN, ADHIKARI NKJ, KISS A, QUAN S and ZWARENSTEIN M (2012) Electronic prescribing in an ambulatory care setting: a cluster random- ized trial. Journal of Evaluation in Clinical Practice 18(4), 761–767.

DAVIDSON SJ and BRYANT DM (2002) B2b exchanges: lessons from the trading pit. Journal of International Law 5(10), 1–10.

DE HAES S and GREMBERGEN WV (2006) Information technology gover- nance best practices in Belgian organisations. In Proceedings of the 39th Hawaii International Conference on System Sciences (SPRAGUE, JR, RH Ed).

DE SMET PAGM and DAUTZENBERG M (2004) Repeat prescribing: scale, problems and quality management in ambulatory care patients. Drugs 64(16), 1779–1800.

DE VRIES M, VAN DER MERWE A, GERBER A and KOTZE P (2010) Refining the operating model concept to enable systematic growth in operating maturity. In Sustaining South Africa’s Global Competitiveness after 2010: 24 Annual SAIIE Conference, pp 32–46, 6–8 October, Muldersdrift.

DEVINE EB, HOLLINGWORTH W, HANSEN RN, LAWLESS NM, WILSON-NORTON JL, MARTIN DP, BLOUGH DK and SULLIVAN SD (2010) Electronic prescribing at the point of care: a time–motion study in the primary care setting. Health Services Research 45(1), 152–171.

DOHERTY NF, COOMBS CR and LOAN-CLARKE J (2006) A re-conceptualiza- tion of the interpretive flexibility of information technologies: redress- ing the balance between the social and the technical. European Journal of Information Systems 15(6), 569–582.

DUMOULIN T and PROBST J (2011) The strategic role of an IT operating model. The Hitch Hiker’s Guide to the ITIL Galaxy and Beyond, Troy’s Blog, Pink Elephant, http://blogs.pinkelephant.com/index.php?/troy/ comments/the_strategic_role_of_an_it_operating_model/.

ERNST ME, KELLY MW, HOEHNS JD, SWEGLE JM, BUYS LM, LOGEMANN CD, FORD JK, KAUTZMAN HA, SOROFMAN BA and PRETORIUS RW (2000) Prescription medication costs: a study of physician familiarity. Archives of Family Medicine 9(10), 1002–1007.

FLYNN EA, BARKER KN and CARNAHAN BJ (2003) National observational study of prescription dispensing accuracy and safety in 50 pharmacies. Journal of American Pharmaceutical Association 43(2), 191–200.

FONSTAD NO and ROBERTSON D (2006) Transforming a company, project by project: the IT engagement model. MIS Quarterly Executive 5(1), 1–14.

FREDERICK J (2003) Pharmacy in 2003: a search for recognition. Drug Store News 25(10), 55–56.

Governance in e-prescribing Nelson King564

European Journal of Information Systems

FRIEDMAN MA, SCHUETH A and BELL DS (2009) Interoperable electronic prescribing in the United States: a progress report. Health Affairs 28(2), 393–403.

GASSON S (2003) Rigor in grounded theory research: an interpretive perspective on generating theory from qualitative field studies. In Handbook for Information Systems Research (WHITMAN ME and WOSZCZYNSKI AB, Eds), pp 79–102, Idea Group Publishing, Hershey, PA.

GITTELL JH and WEISS L (2004) Coordination networks within and across organizations: a multi-level framework. Journal of Management Studies 41(1), 127–153.

GLASER BG and STRAUSS AL (1967) The Discovery of Grounded Theory. Aldine, Chicago.

GRANT GG and ULBRICH F (2010) Governance of IT-shared services: moving between hierarchical and networked service provision. In International Conference on Information Resources Management (CONF- IRM) [WWW document] http://aisel.aisnet.org/confirm2010/32.

GREENHALGH T and PEACOCK R (2005) Effectiveness and efficiency of search methods in systematic reviews of complex evidence: audit of primary sources. BMJ 331(7524), 1064–1065.

GREENHALGH T, POTTS HWW, WONG G, BARK P and SWINGLEHURST D (2009) Tensions and paradoxes in electronic patient record research: a systematic literature review using the meta-narrative method. Milbank Quarterly 87(4), 729–788.

GREENHALGH T and RUSSELL J (2006) Reframing evidence synthesis as rhetor- ical action in the policy making drama. Healthcare Policy 1(2), 34–42.

GROSSMAN JM, CROSS DA, BOUKUS ER and COHEN GR (2011) Transmitting and processing electronic prescriptions: experiences of physician practices and pharmacies. Journal of the American Medical Informatics Association 19(3), 1–7.

HAMMOND WE (2004) Perspective: the role of standards in electronic prescribing. Health Affairs Web Exclusive, w4.325-w324-327, http:// content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.325v1.

HANSEN LB, FERNALD D, ARAYA-GUERRA R, WESTFALL JM, WEST D and PACE W (2006) Pharmacy clarification of prescriptions ordered in primary care: a report from the applied strategies for improving patient safety (ASIPS) collaborative. Journal of the American Board of Family Medicine 19(1), 24–30.z

HEIDE JB (1994) Interorganizational governance in marketing channels. The Journal of Marketing 58(1), 71–85.

HEKKALA R, URQUHART C, NEWMAN M and HEISKANEN A (2010) Under- standing governance issues in an inter-organizational IS project. In 21st Australasian Conference on Information Systems (ACIS) 2010 Proceedings, 1–3 December.

HENDERSON JC and VENKATRAMAN N (1993) Strategic alignment: lever- aging information technology for transforming organizations. IBM Systems Journal 32(1), 472–484.

HING E and BURT CW (2008) Characteristics of office-based physicians and their medical practices: United States, 2005–2006. Vital and Health Statistics 13(166), 1–34.

HOLLINGWORTH W et al (2007) The impact of e-prescribing on pre- scriber and staff time in ambulatory care clinics: a time- motion study. Journal of the American Medical Informatics Association 14(6), 722–730.

HONG IB (2002) A new framework for interorganizational systems based on the linkage of participants’ roles. Information & Management 39(4), 261–270.

HUNT JS, SIEMIENCUK J and KOCH C (2008) Physicians perspectives on physician drug dispensing using a qualitative focus group methodology. Journal of Pharmaceutical Finance, Economics & Policy 16(4), 57–66.

HYYSALO S (2010) Health Technology Development and Use. Routledge, New York.

IT GOVERNANCE INSTITUTE (2005) Governance of the Extended Enterprise. Wiley, Hoboken.

JOHNSON KB and FITZHENRY F (2006) Activity diagrams for integrating electronic prescribing tools into clinical workflow (case report). Journal of American Medical Informatics Association 13(4), 391–395.

KAJIOKA EH, ITOMAN EM, LI ML, TAIRA DA, LI GG and YAMAMOTO LG (2005) Pediatric prescription pick-up rates after ED visits. The American Journal of Emergency Medicine 23(4), 454–458.

KARTSEVA V, HULSTIJN J, GORDIJN J and TAN Y-H (2010) Control patterns in a health-care network. European Journal of Information Systems 19(3), 320–343.

KASSAM R, VOLUME-SMITH C and ALBON S (2008) Informed shared decision making: an exploratory study in pharmacy. Pharmacy Practice 6(2), 57–67.

KAYWORTH TR and SAMBAMURTHY V (2000) Facilitating localized exploita- tion and enterprise-wide integration in the use of IT infrastructures: the role of PC/LAN infrastructure standards. SIGMIS Database 31(4), 54–80.

KENIS P and PROVAN KG (2009) Towards an exogenous theory of public network performance. Public Administration 87(3), 440–456.

KHAN S, SYLVESTER R, SCOTT D and PITTS B (2008) Physicians’ opinions about responsibility for patient out-of-pocket costs and formulary prescribing in two midwestern states. Journal of Managed Care Pharmacy 14(8), 780–789.

KING N (2011) A benefits dependency network as the bridge between requirements and business objectives: an ODE perspective. Interna- tional Journal of Organisational Design and Engineering 1(3), 185–208.

KING N (2012) Whole networks versus inter-organizational systems: exploring common ground for US e-prescribing. In Eighteenth Americas Conference on Information Systems, (JESSUP L and VALACICH J, co-chairs) Seattle.

KING N, CHRISTIE T and ALAMI KM (2007) Process implications of e-prescribing information integration models: United States versus a Middle East approach. E-service Journal 5(3), 15–38.

KINNAIRD D, COX T and WILSON JP (2003) Unclaimed prescriptions in a clinic with computerized prescriber order entry. American Journal of Health-System Pharmacy 60(14), 1468–1470.

KIRKING DM, LEE JA, ELLIS JJ, BRIESACHER B and MCKERCHER PL (2006) Patient-reported under use of prescription medications: a comparison of nine surveys. Medical Care Research and Review 63(4), 427–446.

KLEIN S, POULYMENAKOU A, RIEMER K, PAPAKIRIAKOPOULOS D, GOGOLIN M and NIKAS A (2005) IOIS and interfirm networks – interdependencies and managerial challenges. In Inter-organizational Information Systems in the Internet Age (EOM SB, Ed), pp 170–212, Idea Group Publishing, London, UK.

KNOBEN J, OERLEMANS LAG and RUTTEN RPJH (2006) Radical changes in inter-organizational network structures: the longitudinal gap. Techno- logical Forecasting and Social Change 73(4), 390–404.

KOHN LT, CORRIGAN JM and DONALDSON MS (Eds) (1999) To Err is Human: Building a Safer Health System. National Academy Press, Washington, DC.

KUMAR K and VAN DISSEL HG (1996) Sustainable collaboration: managing conflict and cooperation in interorganizational systems. MIS Quarterly 20(3), 279–300.

KURIMOTO A (2008) Structure and governance of networks: cases of franchising and co-operative chains. In Strategy and Governance of Networks (HENDRIKSE G, TUUNANEN M, WINDSPERGER J and CLIQUET G, Eds), pp 63–82, Physica-Verlag, Heidelberg.

KWO EC, KAMAT P and STEINMAN MA (2009) Physician use of brand versus generic drug names in 1993–1994 and 2003–2004. The Annals of Pharmacotherapy 43(3), 459–468.

LAPENSEE K (2003) Analysis of a prescription drug prior authorization program in a medicaid health maintenance organization. Journal of Managed Care Pharmacy 9(1), 36–44.

LEAVITT MO (2007) Pilot Testing of Initial Electronic Prescribing Standards – Cooperative Agreements Required Under Section 1860d-(4) (e) of the Social Security Act as Amended by the Medicare Prescription Drug, Improvement, and Modernization Action (MMA) of 2003, Report to Congress, Washington, DC.

LEE LL (2009) IT Governance in a Networked World: Multi-sourcing Strategies and Social Capital for Corporate Computing. IGI Global, Hershey, PA.

LEONARDI PM (2009) Why do people reject new technologies and stymie organizational changes of which they are in favor? Exploring misalignments between social interactions and materiality. Human Communication Research 35(3), 407–441.

LEWIS L, ISBELL MG and KOSCHMANN M (2010) Collaborative tensions: practitioners’ experiences of interorganizational relationships. Com- munication Monographs 77(4), 460–479.

LINTON A, GARBER M, FAGAN NK and PETERSON M (2007) Factors associated with choice of pharmacy setting among DoD health care bene- ficiaries aged 65 years or older. Journal of Managed Care Pharmacy 13(8), 677–686.

LIPTON H, MILLER R and WIMBUSH J (2003) Electronic prescribing: ready for prime time? Journal of Health Information Management 17(4), 72–79.

Governance in e-prescribing Nelson King 565

European Journal of Information Systems

LOMAS J (2005) Using research to inform healthcare managers and policy makers questions: from summative to interpretive synthesis. Healthcare Policy 1(1), 55–71.

LUFF P, HEATH C and GREATBACH D (1992) Tasks-in-interaction: paper and screen based documentation in collaborative activity. In 1992 ACM Conference on Computer-supported Cooperative Work, (MANTEL M and BAECKER R, chairman.) Toronto, Ontario, Canada. ACM, New York.

LYYTINEN K and DAMSGAARD J (2011) Inter-organizational information systems adoption – a configuration analysis approach. European Journal of Information Systems 20(5), 496–509.

MAGER DE and COX ER (2007) Relationship between generic and preferred-brand prescription copayment differentials and generic fill rate. American Journal of Managed Care 13(6 (Part 2)), 347–352.

MALONE DC (2005) The role of pharmacoeconomic modeling in evidence-based and value-based formulary guidelines. Journal of Managed Care Pharmacy 11(4), S7–S10.

MALONE TW and CROWSTON K (1994) The interdisciplinary study of coordination. ACM Computing Surveys 26(1), 87–119.

MARICH MJ, HORAN TA and SCHOOLEY BL (2008) Understanding IT governance within the San Mateo county emergency medical service agency. In 5th International ISCRAM Conference (FIEDRICH F and WALLE BVD, Eds), May, ISCRAM Association, Washington, DC.

MOIDUDDIN A et al (2007) Findings from the Evaluation of E-prescribing Pilot Sites. Agency for Healthcare Research and Quality (AHRQ), Washington, DC.

MOTT DA and CLINE RR (2002) Exploring generic drug use behavior the role of prescribers and pharmacists in the opportunity for generic drug use and generic substitution. Medical Care 40(8), 662–674.

MUELLER L, MAGEE M, MAROUNEK P and PHILLIPSON A (2008) IBM IT Governance Approach: Business Performance Through IT Execution. IBM Redbooks, Raleigh, NC.

NIAZKHANI Z, PIRNEJAD H, BERG M and AARTS J (2009) The impact of computerized provider order entry systems on inpatient clinical workflow: a literature review. Journal of the American Medical Informatics Association 16(4), 539–549.

OKHUYSEN GA and BECHKY BA (2009) Coordination in organizations: an integrative perspective. The Academy of Management Annals 3(1), 463–502.

PARDO TA, GIL-GARCIA JR and BURKE GB (2008) Governance structures in cross-boundary information sharing: lessons from state and local criminal justice initiatives. In Hawaii International Conference on System Sciences, Proceedings of the 41st Annual (SPRAGUE RH, JR, Ed), p 211, 7–10 January 2008.

PATERSON B, CANAM C, JOACHIM G and THORNE S (2003) Embedded assumptions in qualitative studies of fatigue. Western Journal of Nursing Research 25(2), 119–133.

PATERSON BL, CANAM C, THORNE SE and JILLINGS C (2001) Meta-study of Qualitative Health Research: A Practical Guide to Meta-analysis and Meta-synthesis. Sage, Thousand Oaks, California.

PERKO J (2008) IT Governance and Enterprise Architecture as Prerequisites for Assimilation of Service-oriented Architecture. Tampere University of Technology, Tampere, Finland.

PETERSON RR (2004) Integration strategies and tactics for information technology governance. In Strategies for Information Technology Governance (VAN GREMBERGEN W, Ed), pp 37–80, Idea Group Publishing, London, UK.

PETTIGREW AM (1997) What is processual analysis? Scandinavian Journal of Management 13(4), 337–348.

PHILLIPSON A (2008) Governing your value stream: a foundation for lower IT costs and increased returns. The Rational Edge, http://www.ibm .com/developerworks/rational/library/edge/08/may08/phillipson/.

PIZZI LT, SAH D-C, BARONE J and NASH DB (2005) Factors related to physicians’ adoption of electronic prescribing: results from a national survey. American Journal of Medical Quality 20(1), 22–32.

POLLOCK N and WILLIAMS R (2010) E-infrastructures: how do we know and understand them? Strategic ethnography and the biography of artefacts. Computer Supported Cooperative Work (CSCW) 19(6), 521–556.

POPAY J et al (2006) Guidance on the Conduct of Narrative Synthesis in Systematic Reviews: A Product from the ESRC Methods Programme, Manchester.

PROVAN KG, FISH A and SYDOW J (2007) Interorganizational networks at the network level: a review of the empirical literature on whole networks. Journal of Management 33(3), 479–516.

PROVAN KG and KENIS P (2008) Modes of network governance: structure, management, and effectiveness. Journal of Public Administration Research and Theory 18(2), 229–252.

RAAB J and KENIS P (2009) Heading toward a society of networks. Journal of Management Inquiry 18(3), 198–210.

RITCHEY FJ and RANEY MR (1981) Medical role-task boundary maintenance: physicians’ opinions on clinical pharmacy. Medical Care 19(1), 90–103.

ROBEY D, ROSS JW and BOUDREAU MC (2002) Learning to implement enterprise systems: an exploratory study of the dialectics of change. Journal of Management Information Systems 19(1), 17–46.

ROMERO S (Ed.) (2011) Eliminating ‘Us and them’. Apress, New York. ROMZA JH and BLACK GE (1999) Pharmacy data and information systems.

In Managed Care Pharmacy Practice (NAVARRO R, Ed), pp 125–144, Aspen Publishers, New York.

ROSS JW, WEILL P and ROBERTSON DC (2006) Enterprise Architecture as Strategy. Harvard Business School Press, Cambridge, MA.

ROSS SM, PAPSHEV D, MURPHY EL, STERNBERG DJ, TAYLOR J and BARG R (2005) Effects of electronic prescribing on formulary compliance and generic drug utilization in the ambulatory care setting: a retrospective analysis of administrative claims data. The Journal of Managed Care Pharmacy 11(5), 410–415.

ROUSSEAU DM, MANNING J and DENYER D (2008) Chapter 11: evidence in management and organizational science: assembling the field’s full weight of scientific knowledge through syntheses. The Academy of Management Annals 2(1), 475–515.

SALMIVALLI L (2008) Governing the implementation of a complex inter-organizational information system network. Turku School of Economics.

SALMIVALLI LAT, SALMELA H and KESTILA T (2008) Building inter-organiza- tional cooperative network for IT collaboration. In Proceedings of the 41st Annual Hawaii International Conference on System Sciences, 7–10 January, IEEE Computer Society, Waikoloa, HI.

SANDELOWSKI M and BARROSO J (2007) Handbook for Synthesizing Qualitative Research. Springer, New York.

SANDELOWSKI M, VOILS CI and BARROSO J (2007) Comparability work and the management of difference in research synthesis studies. Social Science & Medicine 64(1), 236–247.

SANDELOWSKI M, VOILS CI, BARROSO J and LEE E-J (2008) Distorted into clarity: a methodological case study illustrating the paradox of systematic review. Research in Nursing & Health 31(5), 454–465.

SARDINHA C (1998) Electronic prescribing: the next revolution in pharmacy? Journal of Managed Care Pharmacy 4(1), 35–39.

SCHADE CP, SULLIVAN FM, DE LUSIGNAN S and MADELEY J (2006) E-prescribing, efficiency, quality: lessons from the computerization of UK family practice. Journal of American Medical Informatics Association 13(5), 470–475.

SETHIBE T, CAMPBELL J and MCDONALD C (2007) IT governance in public and private sector organisations: examining the differences and defining future research directions. In 18th Australasian Conference on Information Systems IT Governance in the Public Sector, 5–7 December, Toowoomba.

SHPILBERG D, BEREZ S, PURYEAR R and SHAH S (2007) Avoiding the alignment trap in information technology. MIT Sloan Management Review 49(1), 51–58.

SHRANK WH et al (2006) The effect of pharmacy benefit design on patient-physician communication about costs. Journal of General Internal Medicine 21(4), 334–339.

SIAU K and LONG Y (2005) Synthesizing e-government stage models – a meta-synthesis based on meta-ethnography approach. Industrial Management & Data Systems 105(4), 443–458.

SIMONSSON M and EKSTEDT M (2006) Getting the priorities right: literature vs practice on IT governance. In Portland International Center for Management of Engineering and Technology Conference: Technology Management for the Global Future (KOCAOGLU DF, ANDERSON TR and DAIM TU, Eds), pp 18–26, 9–13 July, Portland International Center for Management of Engineering and Technology (PICMET), Istanbul, Turkey.

SINGH S and WOO C (2009) Investigating business-IT alignment through multi-disciplinary goal concepts. Requirements Engineering 14(3), 177–207.

Governance in e-prescribing Nelson King566

European Journal of Information Systems

SINHA KK and VEN AHVD (2005) Designing work within and between organizations. Organization Science 16(4), 389–408.

SK&A HEALTHCARE INFORMATION SOLUTIONS (2007) Consumer Access to Pharmacies in the United States, 2007. Pharmaceutical Care Manage- ment Association (PCMA), Washington, DC.

SPIL T and SALMELA HTT (2007) Inter organizational evaluation of SISP: what new criteria are needed? In ECIS 2007 Proceedings, [WWW document] http://aisel.aisnet.org/ecis2007/106.

STAR SL (1999) The ethnography of infrastructure. American Behavioral Scientist 43(3), 377–391.

SUH D-C (1999) Trends of generic substitution in community pharma- cies. Pharmacy World & Science 21(6), 260–265.

SULISTYO P (2009) Inter-organizational strategic information systems planning (iosisp) in network perspective: how network governance influences iosisp effectiveness evaluation of healthcare, government and logistic/transport industry sectors in the Netherlands. Electrical Engineering, Mathematics and Computer Science, University of Twente.

SURESCRIPTS (2006) Statewide Technology Assessment: ‘Get Connected’ Campaign. Final Report, 21 December.

SURESCRIPTS (2010) 2009 Progress Report on E-prescribing. Alexandria, VA. TAMBLYN R et al (2006) The development and evaluation of an integrated

electronic prescribing and drug management system for primary care. Journal of the American Medical Informatics Association 13(2), 148–159.

TAPIA RS, DANEVA M, VAN ECK P and WIERINGA R (2008) Towards a business-IT aligned maturity model for collaborative networked organizations. In 2008 12th Enterprise Distributed Object Computing Conference Workshops (VAN SINDEREN M, ALMEIDA JPA, PIRES LF and STEEN M, Eds), pp 276–287, 16 September.

TAPSCOTT D, LOWY A and TICOLL D (2000) Digital Capital: Harnessing the Power of Business Webs. Harvard Business Press, Cambridge, MA.

TARN DM, PATERNITI DA, HERITAGE J, HAYS RD, KRAVITZ RL and WENGER NS (2006) Physician communication about the cost and acquisition of newly prescribed medications. American Journal of Managed Care 12(11), 657–664.

TARN DM et al (2008) How much time does it take to prescribe a new medication? Patient Education and Counseling 72(2), 311–319.

TEICH J, BORDENICK JC and MARCHIBRODA J (2004) Electronic Prescribing: Toward Maximum Value and Rapid Adoption. eHealth Initiative, Washington, DC.

TEICH JM, OSHEROFF JA, PIFER EA, SITTIG DF and JENDERS RA (2005) Clinical decision support in electronic prescribing: recommendations and an action plan: report of the joint clinical decision support workgroup. Journal of the American Medical Informatics Association 12(4), 365–376.

THE OPEN GROUP (2009) TOGAF version 9.0 ‘Enterprise edition’, The Open Group, Reading, UK.

THORNTON G (2007) Cost of Dispensing Study: National Study to Determine the Cost of Dispensing Prescriptions in Community Retail Pharmacies. Coalition for Community Pharmacy Action, Chicago.

VOGT M, HERTWECK D and HALES K (2011) Strategic ICT alignment in uncertain environments: an empirical study in emergency manage- ment organizations. In Hawaii International Conference on System Sciences, (SPRAGUE RH, JR, Ed), pp 1–11, 4–7 January 2011.

WALSHAM G (1995) Interpretive case studies in IS research: nature and method. European Journal of Information Systems 4(1), 74–81.

WANG CJ, MARKEN RS, MEILI RC, STRAUS JB, LANDMAN AB and BELL DS (2005) Functional characteristics of commercial ambulatory electronic prescribing systems: a field study. Journal of the American Medical Informatics Association 12(3), 346–356.

WANG CJ et al (2009) Perceptions of standards-based electronic prescribing systems as implemented in outpatient primary care: a physician survey. Journal of American Medical Association 16(4), 493–502.

WAREHAM J, RAI A and PICKERING G (2005) Standardization in vertical industries: an institutional analysis of XML-based standards infustion in electricity markets. Electronic Markets 15(4), 323–334.

WEBB P, POLLARD C and RIDLEY G (2006) Attempting to define IT governance: wisdom or folly? In Proceedings of the 39th Annual Hawaii International Conference on System Sciences (SPRAGUE RH, JR, Ed), p 194a, 4–7 January.

WEILL P and ROSS J (2005) A matrixed approach to designing IT governance. MIT Sloan Management Review 46(2), 26–34.

WEILL P and ROSS JW (2004) IT Governance: How Top Performers Manage IT Decision Rights for Superior Results. Harvard Business School Press, Boston, MA.

WIERINGA R (2008) Operational business-IT alignment in value webs. In Information Systems and E-business Technologies (KASCHEK R, KOP C, STEINBERGER C and FLIEDL G, Eds), pp 371–378, Springer, Berlin.

WIGGINS C, BEACHBOARD J, TRIMMER K and PUMPHREY L (2006) Entrepre- neurial IT governance: electronic medical records in rural healthcare. International Journal of Healthcare Information Systems and Informatics 1(4), 40–54.

WILKIN C (2009) The influence of net benefits on collective, innova- tive, configural system use: a case study of small-to-medium enter- prises. The Electronic Journal Information Systems Evaluation 12(2), 187–198.

WILSON I et al (2007) Physician–patient communication about prescrip- tion medication nonadherence: a 50-state study of America’s seniors. Journal of General Internal Medicine 22(1), 6–12.

YIN RK (1989) Case Study Research. Sage Publications, Beverly Hills. YIN RKB (2003) Case Study Research: Design and Methods. Sage

Publications, Beverley, CA. ZAHEER A and VENKATRAMAN N (1995) Relational governance as an

interorganizational strategy: an empirical test of the role of trust in economic exchange. Strategic Management Journal 16(5), 373–392.

ZARVIĆ N (2008) Considering Structural Properties of Inter-organizational Network Fragments During Business-IT Alignment. Centre for Telematics and Information Technology, The Netherlands.

Governance in e-prescribing Nelson King 567

European Journal of Information Systems

Appendix A

Appendix B

Table A1 Breakdown of academic medication management citations

Topic Medicine Pharmacy Informatics Other Total

Formulary adherence & generic utilization 15 (0) 25 (1) 1 (1) 2 (0) 44 (2)

Medication costs 23 (1) 32 (1) 3 (3) 5 (3) 64 (8)

Drug interaction alerts 7 (1) 30 (1) 12 (2) 2 (0) 55 (4)

Physician–pharmacist relationships 5 (0) 22 (1) 2 (0) 2 (0) 29 (1)

Medication errors 11 (1) 37 (3) 4 (2) 5 (2) 57 (8)

E-prescribing 9 (9) 3 (3) 15 (15) 8 (8) 35 (35)

Other topics 14 (2) 38 (4) 60 (6) 20 (2) 129 (14)

Total (April 2011) 84 (14) 187 (14) 97 (29) 44 (15) 413 (72)

Note: Articles with e-prescribing in title indicated in (parentheses).

Table B1 U.S. e-prescribing network membership

Member Type Primary network role E-connections Entities

Medical Practice (includes prescriber,

medical staff)

Private Order Entry Pharmacy, Intermediary via App

Vendor, Payer, Policy–maker

B200,000

Pharmacy (includes staff, pharmacist) Private Order Fulfillment Patient, Medical Practice, PBM,

Payer, Intermediary via App Vendor

B55,000

Pharmacy Benefit Manager (PBM) Private Order Approval: Eligibility

and Reimburse pharmacies

Pharmacy B50

Payer and/or insurance company Private, Public Provide medication health

benefit to employees

Patient, Pharmacy, Medical Practice B100,000

Policy-maker New! Public Drive adoption Medical Practice, Intermediary 1

Intermediary New! Private Set standards; certification;

Transfer e-scripts

Patient, Medical Practice, PBM, Payer,

App Vendor, Policy-maker

Application vendors New! Private Implement standards – sell

to medical practices and

pharmacies

Medical Practice, Pharmacy B200

Note: Bold members (e-connections column) are those who belonged to the manual network before joining the e-prescribing network. When pilot studies were conducted in 2006 (Moiduddin et al, 2007), there were 206,157 medical practices (Anderson, 2007) and 58,355 pharmacy locations (SK&A Healthcare Information Solutions, 2007).

Governance in e-prescribing Nelson King568

European Journal of Information Systems

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.