Considerations for sociotechnical design: experiences with an electronic patient record in a clinical context

https://doi.org/10.1016/S1386-5056(98)00143-9Get rights and content

Abstract

This paper argues that we should understand the process of IT design as the development of sociotechnical configurations. Drawing upon our experiences with an electronic patient record (EPR) on an Intensive Care Unit (ICU), we depict medical work practices as natural systems. Several considerations for design are developed. First, the EPR should not be overly structured with rationalistic and prefixed notions of the organization and content of medical work. Implementing structure is crucial, however, this should be derived from detailed, empirical knowledge of the practice involved. Second, it is crucial to ensure that the usage of the system will yield immediate benefits for primary users: the systems should support work, not generate it. Third, designing IT should include being aware of the socio–political nature of seemingly ‘neutral’ tools as EPRs.

Introduction

It is more and more acknowledged that a sociological understanding of the complex practices in which information technologies are to function is crucial. Without such knowledge, it is argued, the hope for adequately functioning systems might remain an illusion 1, 2, 3, 4, 5. This knowledge cannot be gathered with traditional methods of eliciting ‘system requirements’ (such as interviews or structured meetings with some end users), nor can its dynamic complexity be captured in abstracted workflow-charts or entity/relationship models 6, 7. To allow for proper design and implementation, it is imperative to conceptualize and study these practices as natural systems [8]. This implies taking a bottom-up approach, and studying a practice in all its complexity rather than depicting it in an abstract, and often overly rational, model. It implies, also, not drawing on preconceived, formal depictions, but on empirical, often qualitative studies, to gather insight into the everyday sociocultural processes that constitute these practices 9, 10.

This awareness also implies that ‘design’ is not limited to the construction of a technical artifact: throughout the whole process of development and implementation, ‘computerization [is about] the development of socio–technical configurations’ [8]. In other words, since a technical system is thoroughly intertwined with the work practice in which it functions, every change in IT will have widespread and often unpredictable consequences for that work practice.

The importance of such a perspective for discussions about the (im-) possibilities and potential designs of the electronic patient record (EPR) has been argued elsewhere [1]. Based on a sociological analysis of medical work (including the multi-faceted role of the medical record), that study drew four implications for design. It rephrased, first what a ‘complete’ medical record should look like, and it argued for a notion of medical ‘data’ as ‘self-evident’ only within the context of their use (this issue is further elaborated in [11]). Thirdly, it argued that computer design could be more oriented toward creating tools that support medical work as a social, interactive process. Finally, the paper argued that these issues required intense user-participation in design.

In this paper, we illustrate and further develop these design-considerations drawing on the experience with the development of an EPR on an Intensive Care Unit (ICU) in a Dutch research hospital. This EPR is a commercial application specifically designed for ICUs. It is particularly strong in the gathering and display of quantitative information, such as laboratory results (automatically filtered from the hospital information system) and information from electronic monitors (often also automatically gathered, but to be validated by a nurse). Two nurses and an anesthesiologist (trained and aided by the vendor; co-authors of this paper) developed a tailored version of the EPR specifically for this ICU, based on the existing forms of the (paper) patient record. The record is unique to the extent in which it manages to fulfil some of the ideal-typed features of an EPR [12]. It is used by both nurses and physicians, end-users enter their own data, and it has all but completely replaced the paper-based record.

Data were gathered through hands-on experience, interviews with both end-users and developers, and participant observation. Although the illustrations given here are all based on one case study, the sociological arguments that are invoked draw on a well established literature. The considerations for design that are presented here, therefore, are not limited to any single case.

Section snippets

Considerations and illustrations

There is a strong tendency to thoroughly structure the EPR: to reduce the amount of free text as much as possible, to use pre-set entries with pre-set options for input, to implement care maps and protocols, and to build in workflows. Different but interrelated considerations lie at the root of this tendency. First, a highly structured record facilitates information retrieval. Much of the impetus for the development of EPRs has come from ‘secondary’ users (hospital management, governments,

Conclusions

Starting with the notion that we should see medical work practices as natural systems, and that we should understand the process of IT design as the development of sociotechnical configurations, we have drawn upon our experiences with an EPR on an ICU to illustrate several considerations for design. These considerations can be summarized in three main points:

(1) The EPR should not be overly structured with rationalistic and pre-fixed notions of the organization and content of medical work.

Acknowledgements

An earlier version of this paper was published in the Proceedings of the Medical Informatics Europe’97 Conference. (C. Pappas, N. Maglaveras, J.-R. Scherrer (Eds.) Medical Informatics Europe’97, IOS Press, Amsterdam, 1997).

References (38)

  • W.J. Clancey

    The learning process in the epistemology of medical information

    Methods Inf. Med.

    (1995)
  • R. Kling (Ed.), Computerization and controversy: value conflicts and social choices, 2nd ed., Academic Press, New York,...
  • B. Kaplan, Organizational evaluation of medical information resources, in: C.P. Friedman, J.C. Wyatt (Eds.), Evaluation...
  • B. Kaplan, Fitting System Design to Work Practice: Using Observation in Evaluating A Clinical Imaging System,...
  • M. Berg, E. Goorman, The Contextual Nature of Medical Information, Int. J. Med. Inf., 1998 (in...
  • R.S. Dick, E.B. Steen (Eds.), The Computer-Based Patient Record: An Essential Technology for Health Care, National...
  • M.J. Ball, M.F. Collen (Eds.), Aspects of the Computer-Based Patient Record, Springer, Berlin,...
  • P.E. Agre

    Surveillance and capture: two models of privacy

    Inf. Soc.

    (1994)
  • S.J. Tanenbaum

    What physicians know

    NEJM

    (1993)
  • Cited by (152)

    • The experiences and perceptions of users of an electronic patient record system in a pediatric hospital setting: a systematc review

      2022, International Journal of Medical Informatics
      Citation Excerpt :

      However, with these promises comes unanticipated consequences. The focus is often on the technological factors, at the expense of the human/social influences that affect the performance of technical systems: implementing structure is essential [49]. According to Berg et al. (1998), the system should yield immediate benefits for primary users, supporting rather than creating additional work [49].

    • Sociotechnical analysis of nurses’ use of personal mobile phones at work

      2016, International Journal of Medical Informatics
    View all citing articles on Scopus
    View full text