Info Systems J

(2004)

14

, 343–362

© 2004 Blackwell Publishing Ltd

343

Blackwell Science, LtdOxford, UKISJInformation Systems Journal1365-2575Blackwell Publishing Ltd, 200414

343362Original Article

Power and information systems implementationB Doolin
Power and resistance in the implementation
of a medical management information system

Bill Doolin

Auckland University of Technology, New Zealand, email: bill.doolin@aut.ac.nz

Abstract.

This paper presents an analysis of the implementation of a large infor-
mation system in a New Zealand hospital. The system was intended to monitor
and scrutinize clinical activity, and management hoped to influence clinical behav-
iour through the increased visibility afforded by the system. However, doctors were
not the passive subjects of a computerized control system, and some were able to
resist the application of the information produced by challenging its validity or
using it to argue for more resources. Nevertheless, the interrelatedness of the
information system with the organizational forms and practices within which it was
used and which it helped shape reinforced the concepts, norms and values asso-
ciated with the new management and economic discourse prevalent in the hospi-
tal, and helped to produce more defined accountabilities for doctors. With time, the
role of the information system was reinterpreted, and in the face of continued
resistance by doctors, it was relegated to a less significant role. The potential for
engendering self-disciplining behaviour in organizational participants through the
pervasive, everyday use of the information the system produced likewise
diminished.
Keywords:



information systems implementation, management control, power,
health care

INTRODUCTION

Early work in the information systems field has shown that while the introduction of information
systems in organizations may be articulated in relation to concepts of efficiency and rationality,
in practice they function in diverse ways related to the social and political processes that exist
in the organization (e.g. Bariff & Galbraith, 1978; Burchell

et al

., 1980; Kling & Scacchi, 1980;
Markus, 1981). In particular, information systems can influence which organizational actions
and their consequences become relatively more visible. They become ‘mechanisms around
which interests are negotiated, counter claims articulated and political processes explicated’
(Burchell

et al

., 1980, p. 17). More recently, attention has focused on the role of information
systems in facilitating power and control in various organizational settings (e.g. Attewell, 1987;
Zuboff, 1988; Gray, 2001).

B Doolin

© 2004 Blackwell Publishing Ltd,

Information Systems Journal

14

, 343–362

344
Much of this work is underpinned by conventional notions of power, which treat information
as a resource to be deployed in the balance of power between organizational actors (cf. Jas-
person

et al

., 2002). Such a notion of power underplays the constitutive effects of information
systems and the way in which they mediate and reinforce particular understandings of orga-
nizational reality. A different understanding of power, influenced by the work of Michel Foucault,
has been proposed as a basis for explaining the role of information systems in facilitating a cal-
culative form of control through computer-based surveillance and monitoring by authors such
as Orlikowski (1991), Bloomfield & Coombs (1992) and Ball & Wilson (2000).
This paper presents a case study of the implementation of a computer-based information
system intended to monitor the activity and performance of hospital doctors. It uses the Fou-
cauldian perspective on power and information systems mentioned above to provide a local-
ized account of power and control in a New Zealand public health setting. In doing so, the paper
builds on previous studies of this nature in three ways. First, it provides a ‘fine-grained’ analysis
(cf. Ball & Wilson, 2000) capable of exploring the detailed implications of computer-based sur-
veillance systems as a means of management control. Second, it draws on longitudinal data
spanning a 6-year period, providing a valuable opportunity to evaluate the long-term power
effects of such information systems. Third, it provides an empirical example of the usefulness
of the Foucauldian perspective on power in understanding the deployment and operation of a
particular type of management information system.
The structure of the paper is as follows. The first part of the paper reviews the literature on
power and information systems, and articulates an understanding of these based on the
relational conception of power drawn from Foucault. The method used in conducting the
research presented here is then discussed. The main part of the paper applies the theoreti-
cal understanding developed earlier to the implementation of a ‘casemix’ information system
in a New Zealand public hospital. This information system was intended by hospital manage-
ment to provide a window on clinical activity, explicitly linking clinical resource usage with its
financial implications in an attempt to control the clinical behaviour of hospital doctors. The
paper ends with some conclusions about the contribution of this work and possibilities for
future research.

INFORMATION



SYSTEMS

,

POWER



AND



CONTROL

Conventional conceptions of power assume that power exists as a capacity that can be pos-
sessed, and exercised over others in a mechanical or causal manner. Power is seen primarily
as something that denies, represses or coerces (Lukes, 1974; Clegg, 1989; Bloomfield &
Coombs, 1992). Such a zero-sum notion of power implies that shifts in organizational power
are the result of corresponding changes in the organizational distribution of resources, such as
information, which confer power on their possessors. This understanding of power can be seen
in early studies of information systems in organizations (e.g. Pettigrew, 1972; Bariff & Galbraith,
1978; Markus, 1981), and continues to provide the dominant view of power in this field (e.g.
Pfeffer, 1994; Gray, 2001; Jasperson

et al

., 2002).