Summary

This PhD dissertation contributes with a sociological perspective on the creation of clinical overview in daily clinical practice among physicians at hospitals. The issue about creating clinical overview originated from a change in media, when one of the five Danish regions decided to change from a paper-based patient record to an Electronic Patient Record (EPR). The change in media had the effect that some physicians found it difficult to create overview over patient information and the patient’s illness history. This challenge caused the design of the present PhD study, which investigate the creation of overview in daily clinical practice and further investigates how respectively a paper-based patient record and an EPR supports the creation of clinical overview.

That the change from paper-based patient record to EPR can cause difficulties, interruptions and challenges is not surprising, giving the central position of the patient record in health care work. The patient record is a critical prerequisite for the physicians in conducting their daily clinical work, because it is their primary information source when searching for information about patients and this information is critical in different ways in when treating patients. The patient record is multi- functional and works as both a cognitive and a coordinating artefact. A multiplicity of reasons for implementing EPR systems in the health care sector exist, but some of the most mentioned reasons are a reduction in medical errors, an improvement in efficiency and productivity and improved quality in patient care. The development, design and implementation of EPR systems that perform well in daily clinical practice have proven to be difficult and even more difficult to develop systems that perform across different organisational setups.

One of the reasons for this difficulty in developing well functional EPR systems could be the lack of knowledge about the specificities of practices for which the systems are suppose to support. The material realities and practical struggles involved in these are often not described or rendered visible in detail. Studies have been conducted on the differences on using respectively a paper-based patient record and an EPR, but few studies have focused on the creation of overview in daily clinical practice in connection with this transition. Therefore, the aim of this PhD dissertation will be to investigate the creation of overview in daily clinical practice among physicians and analyse the processes involved in this creation. The primary focus will be on practice and the work that is accomplished in daily clinical practice among physicians at hospitals. By this focus, the dissertation will render visible the material reality in which the creation of clinical overview takes place.

Besides this, the dissertation tries to conceptualise the creation of clinical overview which will enable a nuanced understanding of the creation of overview and provide a language to be used in the investigation of clinical overview. Considering the change from paper-based patient record to EPR, the dissertation also investigates how the two artefacts support the creation of clinical overview.

The present dissertation is primarily based on ethnographic inspired observational studies and semi-structured interviews. The observations were mainly conducted at a medical ward for the treatment of liver and bowl diseases, where I observed the physicians in their daily clinical work, without any prior judgment about which situations to observe. Observations and interviews were conducted while the ward still used a paper-based patient record and then repeated six month after the implementation of a new ERP system. In addition to this investigation of the creation of overview among physicians at a medical ward, the dissertation also consists of an investigation of the physicians’ use of an overview interface provided by the EPR system. This investigation is conducted by means of numeric information about the use of the overview interface across the entire region and interviews with physicians both using and not using the overview interface, distributed across different departments and different hospitals in the region.

The theoretical framework for the dissertation is primarily constituted by Actor-network Theory, Narrative Theory and the theory about Distributed Cognition. Among other things, Actor-Network Theory introduces a focus on materiality and contingency, Narrative Theory provides an understanding of the structures involved in the process of creating clinical overview and Distributed Cognition render visible that cognition is not just something that takes place in the mind of individuals but is distributed in a socio-technical system.

In an iterative analysis process, the theoretical framework and the empirical data have contributed to a developed and nuanced understanding of clinical overview and conceptualised the creation of clinical overview as Distributed Plot-Making. This concept describes clinical overview as being organised around a plot which induces meaning to the overview and arrange information in various causal relations. The creation of plots takes place within a distributed socio-technical system in which information is propagated and computations take place by means of different artefacts and humans and are therefore not restricted to the minds of individuals. From this perspective, the creation of clinical overview is very much local, contingent, context dependent and material.

Besides the conceptualising of clinical overview as Distributed Plot-Making, five conceptual pairs have been developed in the dissertation, which stretch the creation of clinical overview in five axis, describing various aspects of the situated creation of overview. The five conceptual pairs are providing a language making it possible to talk about physicians’ experiences with and practice concerning the creation of clinical overview. Likewise, they provide an analytical framework making it possible to understand the creation of clinical overview, even though the pairs may not exhaust all aspect of creating an overview.

When investigating the two different socio-technical systems including respectively a paper-based patient record and an EPR, it is evident that they both have advantages and drawbacks concerning their ability to support the creation of clinical overview. When looking at the advantages, paper processes a high degree of micro-mobility, which enables it to be placed wherever and next to whatever one wishes and it can be carried where to it is needed. Likewise, paper is very flexible concerning the possibilities to mark up pages in personal ways and add personal notes to it. On the other hand, the EPR provides quick and easy access to lot of information and once information is entered in the EPR they stay in place. Information in the EPR is often updated due to real time entry, which avoid that information has to travel physically between different departments or different hospitals. Another advantage in the creation of clinical overview, when using an EPR, is that the record is never lost. If you have access to a computer, you have access to the record. Besides this, the search function and the graphical presentations in the EPR are new possibilities introduces to the physicians’ creation of clinical overview, which have the potential to support the creation process in various ways. The two socio-technical systems have different advantages in supporting the creation of clinical overview. Following this we need to understand the two technologies (paper and EPR) not as competing technologies, but instead investigate how the two technologies support the creation of clinical overview together.

The physicians’ personal notes on paper and dialogue with colleagues and patients are two central elements in the process of creating clinical overview and these elements are by and large performed identical in the two socio-technical systems.

A specific way of trying to support the creation of overview is by designing overview interfaces in the EPR systems. The EPR system under investigation in the present dissertation has such an overview interface, which can be configured differently according to wishes from the different departments. The overview interface is not used to any great extent, when looking across the entire region. The departments using the overview interface most, is the departments which uses almost the same kind of information whenever they engage with their patients. In this investigation, these departments are primarily anaesthesiological departments and outpatient clinical. On the contrary, departments and physicians treating patients with complex patient histories do not use the overview interface to any great extent. They found that the overview interface was not able to contain all the information they need and they found it difficult to predefine which information they needed in the overview process. Besides this, many of the physicians found it difficult to trust the information in the overview interface. This mistrust is very much correlated with an underlying discussion of standardisation of documentation practice and the importance of understanding how different parts of the EPR system are connected.

If the goal is to support the creation of clinical overview among physicians at hospitals, these results indicate that you need to think besides the presentation of information. If the physicians are supposed to create an overview, information should be presented in ways that enables them to interpret and make sense of this information. At the same time, to support the creation of overview we need to think in terms of both different artefacts and different ways of organising work in all of the distributed socio-technical system.