Clinical Informatics

Evaluating a Clinical Decision Support System: A basic field study approach


  1. Introduction

  2. Evaluation framework and techniques

  3. Evaluation processes

    1. Basic usability
      1. Effectiveness, efficiency and utility
      2. Safety
      3. Learnability and memorability

    2. Field study results
      1. Hardware design
      2. Software design

  4. Redesign possibilities

  5. Strengths and weaknesses of the field study approach

  6. Conclusion

  7. References


This small study sets out to investigate the human-computer-interactions of a clinical decision support system (CDSS) being used in actual clinical practise. This CDSS has been in development for over 10 years and was originally conceived as a rules-based expert system to improve safety in prescribing.

Initially, it was only used in specialised units within the hospital, where its use and development can be closely monitored and developed. It has been role-out on to all the medical and surgical wards in the hospital in 2007.

From its humble beginnings, ts role has been extended to include the support of patient tracking, patient transfers, patient discharge, test ordering and viewing of test results, including pathology reports and x-rays.

It has an integrated messaging module, which supports communication between its main users consisting of nurses, pharmacists and doctors. The messaging system also functions as an early warning system when an abnormal result has been detected.

It is an interesting system to study as it has been developed to support clinical activity from the very beginning. There has been constant input throughout it’s’ development from key users.

As such, it can be considered a good example of iterative user-based design via the cycle of identifying user requirements, design, evaluate and redesign.

This study seeks to evaluate common generic usability issues that arises from the implementation of CDSS in the clinical setting, and to offer suggestions on how these issues may be addressed in order to improve the end user-experience in the clinical setting.

It is not meant to be a technical review, nor as a mean to endorse or otherwise any product. As such, the identity of the CDSS is kept anonymous by intention.

First impressions

The user interface of the CDSS is based on the "Windows" concept. Common tasks are nested in "folders", which then have further sub-folders and more specific tasks in the same activity group. The folders are labeled by tabs which are clearly visible in the main screen and easily accessible with a mouse click.

There is consistency in the way the common tasks are organised within the "folders" and the use of "folders" also maps to the real world usage of folders in the wards. The "windows" graphical user interface would be a familiar interface to most users.

The lay-out is clear and minimalist in nature. Common task tabs are clearly visible and there is a good sense of predictability while using the system. The tabs have real affordance as a way to change folders. Feedback is clear and unambiguous, especially when a prescribing rule has been violated or disallowed.

Unless the initial patient details, medical and drug history are registered in the system, the other tabs are greyed out and cannot be used. Certain tabs are only accessible to specific users (e.g. "prescriptions" to doctors and "pharmacy" to pharmacists). Hence, it has in-built constraints.

From the above, it can be seen that the system has been designed using the Design Principles(1) of:

  1. Visibility

  2. Feedback

  3. Constraints

  4. Mapping

  5. Consistency

  6. Affordance and

  7. Predictability.

The accompanying hardware had been selected to enable its use at the point of care. Hence, portability, wireless networking and durability are key elements to the system.

There are two form factors; a tablet personal computer or a mobile desk-top unit with a rechargeable power pack. Data entry on the tablet is via a specialised stylus with a magnetic tip which attracts the mouse cursor on proximity to the screen while the mobile desk-top unit uses standard mouse and keyboard.

Access to the system is controlled via a unique username and password. The level of access and features accessible are determined by the users clinical role and an audit trail is created each time a user logs-in with their credentials.

The system automatically logs out the last user after a pre-determined period of inactivity in an effort to enhance system security and protect patient confidentiality.


This website is certified by Health On the Net Foundation. Click to verify. This site complies with the HONcode standard for trustworthy health information:
verify here.


arrow right

Page Updated: 17 May, 2017

About Us | Site Map | Privacy Policy | Contact Us | Disclaimer | Acknowledgement | Top of Page |