Review
Design of decision support interventions for medication prescribing

https://doi.org/10.1016/j.ijmedinf.2013.02.003Get rights and content

Abstract

Objective

Describe optimal design attributes of clinical decision support (CDS) interventions for medication prescribing, emphasizing perceptual, cognitive and functional characteristics that improve human–computer interaction (HCI) and patient safety.

Methods

Findings from published reports on success, failures and lessons learned during implementation of CDS systems were reviewed and interpreted with regard to HCI and software usability principles. We then formulated design recommendations for CDS alerts that would reduce unnecessary workflow interruptions and allow clinicians to make informed decisions quickly, accurately and without extraneous cognitive and interactive effort.

Results

Excessive alerting that tends to distract clinicians rather than provide effective CDS can be reduced by designing only high severity alerts as interruptive dialog boxes and less severe warnings without explicit response requirement, by curating system knowledge bases to suppress warnings with low clinical utility and by integrating contextual patient data into the decision logic. Recommended design principles include parsimonious and consistent use of color and language, minimalist approach to the layout of information and controls, the use of font attributes to convey hierarchy and visual prominence of important data over supporting information, the inclusion of relevant patient data in the context of the alert and allowing clinicians to respond with one or two clicks.

Conclusion

Although HCI and usability principles are well established and robust, CDS and EHR system interfaces rarely conform to the best known design conventions and are seldom conceived and designed well enough to be truly versatile and dependable tools. These relatively novel interventions still require careful monitoring, research and analysis of its track record to mature. Clarity and specificity of alert content and optimal perceptual and cognitive attributes, for example, are essential for providing effective decision support to clinicians.

Highlights

► Alerts should be tiered by severity, have concise text, justification, clear response options. ► Prioritize concurrent alerts, use controlled color sets, consistent terminology. ► Format text to visually associate drug categories, show clinical context data. ► Maintain manageable pick lists, allow multiple entry options and custom order sets.

Introduction

Clinical decision support (CDS) systems can safely and effectively support medication prescribing when they deliver relevant, unambiguous and actionable advice well integrated into patient care [1], [2], [3]. Many contemporary installations, however, have poor interface design, use verbose or unclear language, non-standard terminology, alerts may be temporally misalignment with corresponding clinical tasks and their important human–computer interaction (HCI) attributes may be inadequate, making the receiving and responding to decision support interventions difficult.

There is a recognized and pressing need for high-performing CDS. Aside from an array of successes at specific sites in individual domains, few systems have substantially delivered on the promise to improve healthcare processes and outcomes [4]. The challenges of designing effective but potentially work-disruptive alerts and notifications are manifold and often require the reconciliation of contradictory goals, such as the need for succinctness with the need to adequately support complex medical decisions.

Designers and developers of health information technology (HIT) need a cohesive, widely accepted and reliable set of industry standards, recommendations and best practices to substantially increase the usability, effectiveness and safety of electronic health records (EHRs) and CDS systems. Such guidelines must be rooted in empirical evidence from biomedical informatics and HCI research, follow recognized usability principles and be informed by decades of software design and evaluation experience from other safety-critical domains.

This report describes design recommendations for CDS interventions that are activated during medication prescribing, such as alerts to drug and allergy interactions. We reviewed published reports on the successes, failures and lessons learned from CDS implementation in large hospitals and small clinics and interpreted the findings with regard to HCI principles and software usability. Emerging themes and specific suggestions were then formulated into a set of design recommendations for CDS interventions that would improve their effectiveness, safety and human interaction by, for example, reducing unnecessary workflow interruptions or allowing clinicians to make informed decisions quickly, accurately and without extraneous cognitive and interactive effort. A related methodological review of design approaches that are applicable to a wider range of decision support and EHR systems can be found in a recent JBI article [5].

This targeted review was focused on articles containing references to design features of CDS and therefore was not exhaustive. The recommendations, however, are not limited to specific CDS and EHR systems as they are partially derived from and reconciled with existing general usability principles. They are organized in the following sections according to specific design goals, with high-level principles and examples of their specific application.

Section snippets

Background

There is somewhat scant but increasingly more reported evidence of medical errors, adverse drug events, near misses and other patient safety problems that can be at least in part attributed to failures in human interaction with poorly designed EHR and CDS interfaces. Published reports include descriptions of decreased cognitive performance [6], medication prescribing errors [7], [8], [9], [10], [11], [12], unsafe workarounds [13], [14] and poor handling of safety alerts [15].

A common unintended

Methods

Published studies concerning EHR and CDS systems report primarily on their implementation or the effect they have on the process of care and usually lack sufficiently detailed description of the interface, its design structure or observations about the interactive behavior of clinicians using them. However, they often do contain statements about design features and their usefulness within the context of care and indirect or anecdotal accounts of interface design quality and its effect on

Reducing excessive alerting

Several design and approaches may help reduce the number of disruptive alerts of low clinical value. The degree of alert intrusiveness can be adjusted according to their level of importance, allowing only the most severe warnings to interrupt work [37]. Rules that trigger alerts can also be filtered and prioritized to suppress low-severity warnings by using more sophisticated algorithms that integrate patient context and provider-specific data into the decision logic [4]. Methods and strategies

Alert content, language and typography

Concise and clear recommendations are the most effective [49]. Verbose language may be difficult to interpret, the clinical consequences may become unclear and the clinician may not see the magnitude of the possible risk [50]. The message needs to have a succinct explanation of the interaction and its consequence, using recognizable and accepted terms, must be easily interpretable and generally shorter than ten words [4]. Triggering medical logic needs to be apparent and outlined in a few words

Visual and perceptual characteristics

Visual indicators such as color, font, or screen placement are powerful means of communicating to clinicians perceptually message importance before they read it, whether they need to pay attention to it immediately or if they can safely defer for a more convenient time. CDS is almost always integrated with EHR and prescribing systems. It is important that meaningful color schemes are consistently applied to all visual aspects of the entire system, not just to alerts and reminders. For example,

Discussion

Human factors and usability characteristics have been at the center of device and software design in high-risk domains for decades and safety has invariably improved as a result [93]. Healthcare has been incorporating best practices and proven design principles into IT development at a much slower pace than is necessary to maintain a high level of function and safety for increasingly more complex systems [29] and HIT is therefore often considered as having low reliability [30]. Basic HCI

Conclusion

This report suggests methods and practices to improve the visual and interactive design characteristics of CDS interventions used in medication ordering. Good performance of CDS and its benefit to clinicians can be significantly reduced by poor interface design, incorrect implementation and inadequate data maintenance and may even become a disruptive factor contributing to medical error [17]. Specificity and clarity of alerts and quick ways of responding to suggestions are key to changing the

Authors’ contributions

All authors contributed to the conceptual design of the study, data collection planning and initial drafting of the manuscript. Drs. Bell and Middleton provided substantial help in revising and critiquing the manuscript.

Conflict of interest

None of the authors report any conflict of interests with respect to this study.

Summary points

Already known:

  • Excessive alerting that leads to high override rates is distractive to clinicians and contributes to the risk of inadvertent dismissal of a serious warning.

  • The performance level of many EHR systems with decision support may be substantially increased by designing their human interfaces according to established principles that emphasize user-centered design.

Contribution of this review:

  • A

Acknowledgements

This work was funded by the United States Department of Health and Human Services Contract No. HHSP23320095649WC, Task order HHSP23337009T, Office of the National Coordinator ARRA Contract entitled “Advancing Clinical Decision Support.” Many thanks to Andrew Bondarenko for creating graphical examples.

References (102)

  • W.L. Galanter et al.

    A trial of automated safety alerts for inpatient digoxin use with computerized physician order entry

    J. Am. Med. Inform. Assoc.

    (2004)
  • J. Zhang et al.

    TURF: Toward a unified framework of EHR usability

    J. Biomed. Inform.

    (2011)
  • D.W. Bates et al.

    Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality

    J. Am. Med. Inform. Assoc.

    (2003)
  • H. van der Sijs et al.

    Overriding of drug safety alerts in computerized physician order entry

    J. Am. Med. Inform. Assoc.

    (2006)
  • K.B. Johnson et al.

    Showing your work: impact of annotating electronic prescriptions with decision support results

    J. Biomed. Inform.

    (2010)
  • Y. Ko et al.

    Practitioners’ views on computerized drug–drug interaction alerts in the VA system

    J. Am. Med. Inform. Assoc.

    (2007)
  • H. van der Sijs et al.

    Turning off frequently overridden drug alerts: limited opportunities for doing it safely

    J. Am. Med. Inform. Assoc.

    (2008)
  • G.J. Kuperman et al.

    Using commercial knowledge bases for clinical decision support

    J. Am. Med. Inform. Assoc.

    (2006)
  • D.F. Sittig et al.

    The state of the art in clinical knowledge management: an inventory of tools and techniques

    Int. J. Med. Inf.

    (2010)
  • B. Dean et al.

    Causes of prescribing errors in hospital inpatients: a prospective study

    Lancet

    (2002)
  • R.A. Miller et al.

    The anatomy of decision support during inpatient care provider order entry (CPOE): empirical observations from a decade of CPOE experience at Vanderbilt

    J. Biomed. Inform.

    (2005)
  • R.M. Gardner et al.

    Using computer technology to detect, measure, and prevent adverse drug events

    J. Am. Med. Inform. Assoc.

    (2004)
  • J. Judge et al.

    Prescribers’ responses to alerts during medication ordering in the long term care setting

    J. Am. Med. Inform. Assoc.

    (2006)
  • R. Filik et al.

    Drug name confusion: evaluating the effectiveness of capital (“Tall Man”) letters using eye movement data

    Soc. Sci. Med.

    (2004)
  • B. Chaudhry et al.

    Systematic review: impact of health information technology on quality, efficiency, and costs of medical care

    Ann. Intern. Med.

    (2006)
  • D.W. Bates et al.

    Improving safety with information technology

    N. Engl. J. Med.

    (2003)
  • R.J. Holden

    Cognitive performance-altering effects of electronic medical records: an application of the human factors paradigm for patient safety

    Cogn. Technol. Work

    (2011)
  • D.W. Bates et al.

    The impact of computerized physician order entry on medication error prevention

    J. Am. Med. Inform. Assoc.

    (1999)
  • J.R. Nebeker et al.

    High rates of adverse drug events in a highly computerized hospital

    Arch. Intern. Med.

    (2005)
  • R. Koppel et al.

    Role of computerized physician order entry systems in facilitating medication errors

    JAMA

    (2005)
  • S. Eslami et al.

    Errors associated with applying decision support by suggesting default doses for aminoglycosides

    Drug Saf.

    (2006)
  • T.B. Wetterneck et al.

    Factors contributing to an increase in duplicate medication order errors after CPOE implementation

    J. Am. Med. Inform. Assoc.

    (2011)
  • C.J. McDonald

    Computerization can create safety hazards: a bar-coding near miss

    Ann. Intern. Med.

    (2006)
  • R. Kaushal et al.

    Effects of computerized physician order entry and clinical decision support systems on medication safety: a systematic review

    Arch. Intern. Med.

    (2003)
  • S.N. Weingart et al.

    Physicians’ decisions to override computerized drug alerts in primary care

    Arch. Intern. Med.

    (2003)
  • R. Wipfli et al.

    Alerts in clinical information systems: building frameworks and prototypes

    Stud. Health Technol. Inform.

    (2010)
  • D. Magnus et al.

    GPs’ views on computerized drug interaction alerts: questionnaire survey

    J. Clin. Pharm. Ther.

    (2002)
  • R.M. McFall et al.

    Quantifying the information value of clinical assessments with signal detection theory

    Annu. Rev. Psychol.

    (1999)
  • M.-S. Ong et al.

    Evaluating the effectiveness of clinical alerts: a signal detection approach

  • M. Greenberg et al.

    Clinical decision support and malpractice risk

    JAMA

    (2011)
  • W.R. Hogan et al.

    Accuracy of data in computer-based patient records

    J. Am. Med. Inform. Assoc.

    (1997)
  • U.S. Army

    Manpower and Personnel Integration (MANPRINT) in the Materiel Acquisition Process

    (2001)
  • J.M. O’Hara et al.

    Human Factors Engineering Program Review Model

    (2004)
  • V. Ahlstrom et al.

    Human Factors Design Standard: Human Interface Guidelines

    (2003)
  • B.-T. Karsh et al.

    Health information technology: fallacies and sober realities

    J. Am. Med. Inform. Assoc.

    (2010)
  • R.L. Wears et al.

    “Safeware”: safety-critical computing and health care information technology

  • International Standards Organization

    Ergonomic Requirements for Office Work With Visual Display Terminals: Guidance on Usability. Presentation of Information

    (1998)
  • ANSI/HFES

    Human Factors Engineering of Software User Interfaces

    (2008)
  • Microsoft Corporation

    Windows User Experience Interaction Guidelines

    (2010)
  • Microsoft Health Common User Interface. http://www.mscui.net. Accessed February 28,...
  • Cited by (89)

    View all citing articles on Scopus
    View full text