«Special Issue Journal of the Association for Information Systems Configurable Politics and Asymmetric Integration: Health e-Infrastructures in India* ...»
Journal of the Association for Information Systems
Configurable Politics and Asymmetric Integration:
Health e-Infrastructures in India*
Sundeep Sahay Margunn Aanestad
Department of Informatics Department of Informatics
University of Oslo University of Oslo
email@example.com firstname.lastname@example.org Eric Monteiro Department of Computer and Information Science The Norwegian University of Science and Technology email@example.com Abstract Information Infrastructures typically evolve in an incremental fashion, through partly planned and unplanned processes. A significant mechanism of growth is when previously unconnected systems are integrated, facilitating the transition from networking to inter-networking. Conversely, failure to integrate systems contributes to the lack of evolution of the infrastructure. Integration seems crucial for evolving infrastructures; however, there is little consensus on what it entails, as can be seen when different connotations of ‘integration” are unpacked. In contrast to the dominant view of integration as a largely technical concern, our focus is on how political and institutional interests are embedded in efforts to achieve integration. More specifically, we explore strategies for institutional integration that take into account uneven distribution of political influence. The paper builds on empirical material from our ongoing
1. Introduction The growth trajectories of stand-alone technological artefacts differ significantly from large-scale, inter-connected — i.e. infrastructure – technology. The field of large technical systems consists of historical accounts of the development of infrastructures including railroads, sewage systems, and highways (Summerton 1994). An exemplary case is Hughes’ (1983) description of the establishment of electrical power supply systems on the east coast of the US in the latter half of the 18th century.
Predominantly, this field has emphasised evolutionary, path-dependent change processes that ultimately lead to stabilisation. ICT-based infrastructure technology, e-Infrastructure, exhibits characteristics similar to those of other infrastructures, e.g., of path-dependencies (Hanseth et al.
2001, 2006, Kallinikos 2004). Far less attention is devoted to (more) radical change (but see Egyedi and Verwater-Lukszo 2005; Hanseth 2001, and Geels 2007 for exceptions). Often these radical change processes arise due to political events, as illustrated by Silva’s (2002) description of how post-war crisis in Guatamala paved the road for the delegation of responsibility to consultants for the design of information systems in hospitals.
A characteristic, arguably defining, aspect of e-Infrastructures is their configurability — both technical (Fleck 1988, 1994, Williams, Stuart and Slack 2005) and interpretive (Orlikowski 1992) which inscribes in them the potential to evolve (or not) over time. These e-Infrastructures, by their very definition, need to adapt, interconnect, co-evolve — in short, integrate with other systems — in ways that are poorly understood in research, in terms of patterns of development, innovation, and use.
Strong business trends favor ambitious, generic software packages (e.g., enterprise resource planning systems, content management publishing systems, or customer relationship management systems) that presuppose the possibility of extensive local adaptation in the form of configuration through calibration of various technical parameters (Pollock et al. 2007). Likewise, the wrapping of complementary components or applications into portfolios or services is propelled by the apparent ease of integration of ICT artefacts (as claimed especially by ICT vendors).
IS literature over the last two decades has emphasized the social and political construction and use of IS applications (Orlikowski 2000; Williams et al. 2005), strongly suggesting that integration should be treated as much more than a technical process. This problem gets magnified in the case of eInfrastructures, which, by their nature of being constituted by a diversity of systems, standards, and uses (for example, the Internet), have a multiplicity of social and political interests inscribed in them.
Thus, it is fair to hold that our current conceptualisation of the full implications of the qualifier “integrated” in connection with information systems leaves much to be desired (see, e.g., Boudreau and Robey 2005). Specifically, we have little empirically based knowledge about the interplay of the political and technical configurations that arise during attempts to integrate multiple ICT systems.
The aim of this paper is to explore the implications of the configurational aspects (both political and technical) of ICTs stemming from trends of integration, which ultimately provide the constitutive element in the formation of modern e-Infrastructures. In our analysis, we try to unpack and analyze these configurational elements, both with respect to the technical systems of the e-Infrastructures and the political manoeuvring that is embedded around and within it. Empirically, the analysis is based within the public health sector of a developing country (India), where political and institutional aspects of integration are salient. We specifically focus on two different sets of actors (large and powerful vs.
small and relatively powerless), the systems they are trying to introduce within the same setting, and the political capacity and credibility they each bring to the table. More specifically, the aim of the paper is to analyze how the interplay of the political and technical configurational aspects of IS applications shape processes of integration. Understanding this interplay provides rich insight into the dynamics of integration in the context of e-Infrastructures, providing is a contribution to the research field, which has largely focused on the technologies and techniques of integration.
The rest of this paper is organized as follows. In section 2, we discuss how integration has been conceptualised in IS research. Information the empirical data from a health information systems project in India is presented in section 3, and the case study in section 4. Analysis and discussion follow in section 5 and we conclude with section 6.
Journal of the Association for Information Systems Vol. 10 Special Issue pp. 399-414 May 2009 Sahay et al./Configurable Politics and Asymmetric Integration
2. Conceptualising Integration in IS Research ”Integration has been the Holy Grail of MIS since the early days of computers in organizations” (Kumar and van Hillegersberg 2000, p. 23). Within the domain of health information systems, too, integration is a normative goal stipulated by international agencies like the World Health Organization and national governments. The current thrust of the World Health Organization is the creation of
national enterprise architectures intended to facilitate extensive integration:
“The enterprise architecture is the next level of elaboration of the [framework] where general lessons, standards, and processes can be aggregated and documented for knowledge sharing. A well-thought-out and collaboratively supported architecture enables systems to be built and implemented using consistent standards for data collection, management, reporting, and use. The components of the enterprise architecture will be adapted from or collaboratively generated with the global disease programs whose buy-in and endorsement is crucial to its success. Investments in health information systems can be aligned and leveraged around such architectures to build stronger core health information systems supporting better local health services management, health policy and ultimately stronger health systems” (Stansfield et al 2008, p.1).
Despite the long-standing and increasing focus on integration, the existing literature within both the IS and health information system domains remains overly optimistic and prescriptive, often touting new and better technical approaches to deal with integration (see e.g Chari and Seshadri 2004;
Grimson et al. 2000). However, there are alternative views to this emphasis on the technical aspects of integration, which we will review further below. Our concern for these alternative perspectives on integration was initially empirically motivated. Through our engagement with the challenges related to health information systems in developing countries, we have witnessed the significant negative effects of fragmentation caused by a multiplicity of technical and manual systems, typically introduced by various donor, governmental, and vendor interests (Heeks 2008). The existence of multiple information systems creates redundancy and additional workload for the already over-burdened health workers responsible for both providing clinical care to large numbers of patients and carrying out various administrative tasks including those related to health information systems (Mosse 2006).
The kind of health information systems we focus on support local collection and processing of data relating to health status, disease incidence, prevalence, services rendered, infrastructure and the catchment population. Due to redundancy between systems, health workers may have to report the same data several times: for example, both in the routine reports (within the districts) and in programspecific reports of so-called “vertical” health programs (e.g., on malaria, tuberculosis and HIV/AIDS).
Poor coordination and linkages between and across health programs adversely influences both health delivery and the quality of the reporting systems (Braa et al. 2004). For example, the HIVpositive, pregnant women who are enrolled under the Mother and Child program for antenatal care services may fail to show up in the Prevention of Mother to Child Transmission program, which comes under the umbrella of the HIV/AIDS program (Shidende 2005). Recognizing these various challenges in the health sector, both international donors and national governments are extolling the virtues of integration, but unfortunately, they often do no more than extol the need to buy more sophisticated technologies to link these multiple systems technically.
The public health domain discussed above presents uneasy similarities with current IS research, where, again, integration is predominantly conceived of as a technical issue, and the emphasis is on different mechanisms and strategies for achieving tighter integration (Hasselbring 2000, Grimson et al. 2000). We elaborate on this trend in more detail below.
Traditional View: Integration as A Remedy Since the 1970s, Western business organizations have struggled with the fragmentation of their collection of information systems (McNurlin and Sprague 2001) and looked for ways to integrate them by defining standards for common services and building shared information repositories, terminologies, and technical platforms. Technically, integration refers to the degree of interoperability and interconnectivity among technical components, and relies on standardization at a certain level.
Journal of the Association for Information Systems Vol. 10 Special Issue pp. 399-414 May 2009 Sahay et al./Configurable Politics and Asymmetric Integration Over the last decades, a rich and expanding repertoire of technical mechanisms for integration has been proposed, from low-level (e.g., database schema integration), to middle-level (e.g., middle-ware like CORBA, Web services), to high-level (e.g., Service-Oriented Architectures (SOA)) solutions (Chari and Seshadri 2004).
The high level of uptake of so-called Enterprise Resource Planning (ERP) systems in large business organizations is a poignant illustration of the need to address integration. SAP, the world’s market leader in ERP systems, points out on its web site that the goal is to have business processes “[integrated] ….across departments and functions.” Yet organizational implementation lags significantly behind these promised returns (Goodhue et al. 1992, Hanseth et al. 2001, Kallinikos 2004, Pollock and Cornford 2004, Singletary 2004). The traditional approach to integration, in short, remains overly optimistic, prescriptive, and programmatic.
The Downsides of Integration There are critics of the one-sidedness of the technically focused position on integration. Goodhue et al. (1992) have emphatically called for a more nuanced approach to analyze this complex issue.
Working out a pragmatically based contingency model, they identify conditions under which the costs (in terms of loss of flexibility, increase in development costs) may outweigh the benefits of integration.
Similarly, and more recently, Singletary (2004) found that practitioners saw the downsides to integration as lock-in with vendors, costs, and project risks (see also Markus 2001). Empirically underpinned case studies (see e.g., Hanseth et al. 2006; Rolland and Monteiro 2006; Perrow 1984) demonstrate in a more detailed way the form and implications of the unintended consequences of integration. As the complexity of systems increases with tighter integration, so does the likelihood for unintended effects of any action taken. Thus, the wished-for integration may not emerge, and increased control over fragmented systems may not be achieved.
Political Ecology of Integration: “Asymmetric Integration” The contributions cited above document the unintended consequences of integration but present little analysis of why and how they occur. We argue that achieving an understanding of these deeper questions involves placing greater emphasis on the political and institutional conditions that envelop and shape the context and processes around the dynamics of integration. A political perspective on IS, in general, and integration, in particular, highlights the importance of gaining, maintaining and expanding the political and institutional legitimacy and support for an IS (Cox and Ghoneim 1998).
For instance, within a business sector, dominant actors can design integration solutions that define and solidify trading relationships and reinforce their dominant role (Webster 1995). The fate of integration initiatives often seems to “hinge on the wider issues of inter-organisational relations” (Spinardi et al. 1997, p. 260).