Impact of a computerized physician order entry system on nurse–physician collaboration in the medication process

https://doi.org/10.1016/j.ijmedinf.2008.04.001Get rights and content

Abstract

Background

Due to their efficiency and safety potential, computerized physician order entry (CPOE) systems are gaining considerable attention in in-patient settings. However, recent studies have shown that these systems may undermine the efficiency and safety of the medication process by impeding nurse–physician collaboration.

Objective

To evaluate the effects of a CPOE system on the mechanisms whereby nurses and physicians maintain their collaboration in the medication process.

Setting and methodology

Six internal medicine wards at the Erasmus Medical Centre were included in this study. A questionnaire was used to record nurses’ attitudes towards the effectiveness of the former paper-based system. A similar questionnaire was used to evaluate nurses’ attitudes with respect to a CPOE system that replaced the paper-based system. The data were complemented and triangulated through interviews with physicians and nurses.

Results

Response rates for the analyzed questions in the pre- and post-implementation questionnaires were 54.3% (76/140) and 52.14% (73/140). The CPOE system had a mixed impact on medication work: while it improved the main non-supportive features of the paper-based system, it lacked its main supportive features. The interviews revealed more detailed supportive and non-supportive features of the two systems. A comparison of supportive features of the paper-based system with non-supportive features of the CPOE system showed that synchronisation and feedback mechanisms in nurse–physician collaborations have been impaired after the CPOE system was introduced.

Conclusion

This study contributes to an understanding of the affected mechanisms in nurse–physician collaboration using a CPOE system. It provides recommendations for repairing the impaired mechanisms and for redesigning the CPOE system and thus for better supporting these structures.

Introduction

Computerized physician order entry (CPOE) systems are being pushed as a substitution for paper-based medication systems, especially because of the promise that they would increase the efficiency and safety of the medication process. Physicians and pharmacists, in particular, are increasingly expected to work with these tools, especially in in-patient settings [1]. Many recent studies have shown that these systems improve the medication process and reduce medication errors [2], [3], [4], [5], [6], [7]. One of the main reasons that CPOE systems are believed to improve the medication process is that they support better data communication between care providers [5], [8], [9], [10].

However, there have also been concerns in the literature about the potential disadvantages of these systems [11], [12]. Studies have suggested that CPOE systems may undermine nurse–physician communication and collaboration in the medication process [13], [14], [15], [16]. Likewise, it is suggested that CPOE systems may jeopardise patient safety and the efficiency of the medication process through hidden side-effects that cannot be easily discerned by conventional research methods [12], [13], [17], [18].

Recent socio-technical studies have shown that one important reason for the unintended negative effects of CPOE systems is that they change the nurse–physician communication mode from synchronous to asynchronous [13], [14]. This in turn negatively affects nurse–physician collaborative medication work [19], [20]. Understanding the mechanisms whereby the nurse–physician collaborative work is affected by a CPOE system might therefore offer clues about how to manage the side effects of the changes that have taken place and/or how to adapt the system appropriately. Thus far, however, these mechanisms have not been sufficiently evaluated in the literature.

This study evaluated the medication work support of a CPOE system comparing it with that of a paper-based system. By analysing the reasons the two systems were considered to support or not support the medication process, we sought to answer two following questions: Which mechanisms in nurse–physician communication are affected by the switch from a paper based to a CPOE medication system? How do the affected mechanisms impact nurse–physician collaborative medication work? Both quantitative and qualitative study methods were used to determine what nurses and physicians consider to be supportive or non-supportive features of either system.

Section snippets

Theoretical background

Classic medication work in in-patient settings is a dynamic process of highly collaborative tasks. It consists of various phases – prescription, transcription, procurement, dispensing, administration and monitoring – and involves different hospital care providers, in particular nurses and physicians. In this collaborative ensemble, medication tasks are integrated through applying mechanisms that collectively can be called articulation work. Articulation work is necessary to assure that

Study context

The study was conducted in a Dutch tertiary academic medical centre with 1237 beds. Before implementation of a CPOE system, a paper-based medication system, named TIMED, was used by both physicians and nurses in the internal medicine wards. In this paper-based system physicians wrote their prescriptions on the pre-printed forms. Nurses then translated the prescriptions into suitable administration times and dosage forms according to ward routines. They registered data on an administration form

Methods

Qualitative and quantitative methods were used to evaluate the effect of changing from a paper-based prescription system to an electronic one on nurse–physician communication and collaboration. For two reasons we focused our study more on nurses. First, considering the different stages of the medication process, it can be realized that nurses play a considerable role in almost all phases of the medication process. Second, because of their wide spread presence throughout the medication process,

Quantitative results

Overall response rates were 54.3% (76/140) for the pre-implementation survey and 52.14% (73/140) for the post-implementation survey. Two of the pre-implementation questionnaires did not contain answers for questions analyzed in this study, therefore, they excluded from the analysis process. Demographics of the respondents in both surveys are presented in Table 1. The majority of the respondents were female, practicing nurses and between 24 and 33 years old.

The analysis of the first question

Analysis

Nurses rated the two systems with respect to the options ‘No’ and ‘Unsure’ very differently, which caused the Mann–Whitney U-test to show a significant difference between the two groups. More nurses in the paper-based system than in the CPOE system believed that their medication system did not support their medication work. In contrast, more nurses in the CPOE system than in the paper-based system were unsure as to whether their medication system supported their medication work. These

Discussion

Many of the CPOE system's non-supportive features are produced since changing the prescription system induced problems in nurse–physician interoperability. Comparison of the CPOE system's non-supportive features with the supportive features of the paper-based system demonstrates that two important mechanisms in nurse–physician communication are damaged: synchronisation and feedback. Despite the clarity and completeness of prescription labels, damaged feedback mechanisms made it hard for nurses

Conclusion

Our study demonstrated that both the paper-based and CPOE systems supported the medication work of nurses and physicians. However, the notion of support came from different perspectives. The CPOE system improved the main non-supportive features of the paper-based system, but it could not replace some of its important supportive features. In our study, many of the CPOE system's non-supportive features were listed because the system damaged the synchronisation and feedback mechanisms between

Conflict of interest

The third author was the project leader for implementation of the CPOE system. The reported analysis and interpretations in this paper, however, have been free from any financial/personal relation to the project and represent the authors’ perspectives.

Acknowledgements

The authors would like to thank Liselotte van der Meule (from the project team) and Linda van Wijnen (from nursing staff) for their kind cooperation during this research. We also acknowledge our gratitude to the nurses and physicians who kindly participated in interviews and/or filled in the questionnaires.

Authors contribution: HP, ZN, HS, MB and RB were involved in design and implementation of the study. HP and ZN were responsible for conducting the interviews while HS was responsible for

References (31)

  • D.W. Bates et al.

    The impact of computerized physician order entry on medication error prevention

    J. Am. Med. Inform. Assoc.

    (1999)
  • R. Kaushal et al.

    Information technology and medication safety: what is the benefit?

    Qual. Saf. Health Care

    (2002)
  • D.M. Benjamin

    Reducing medication errors and increasing patient safety: case studies in clinical pharmacology

    J. Clin. Pharmacol.

    (2003)
  • S.H. Cassiani et al.

    Electronic medical prescription at a university hospital: writing failures and users’ opinions

    Rev. Esc. Enferm. USP

    (2003)
  • R. Shulman et al.

    Medication errors: a prospective cohort study of hand-written and computerised physician order entry in the intensive care unit

    Crit. Care

    (2005)
  • Cited by (65)

    View all citing articles on Scopus
    View full text