Patient monitor human interface design

Hewlett-Packard Journal, Oct, 1991 by Gerhard Tivig, Wilhelm Meier

Hifsim was used widely in exposing the human interface design during shows, demonstration sessions, and marketing training at a time when no finalized Component Monitoring System hardware or software was available. This allowed the design team to get very early feedback on its user interface design.

Usability Testing

Hifsim was a prerequisite for being able to set up the Component Monitoring System usability tests. The purpose of the usability tests was to discover which features of the human interface design were effective and which needed to be improved, and to do this testing early in the design process where changes could still be made in the human interface design.

An extensive usability test session was organized in the Boston area by an independent research institute that specializes in human interface studies and human factors research. Our objective was to conduct an independent evaluation of the monitor's user interface, basically the control structure and the screen layout. The test was conducted on a sample of 13 nurses and anesthesiologists who were asked to perform typical patient monitoring tasks. A second objective was to assess the value of usability tests as an aid to the design process of a monitor's user interface.

A game plan was worked out that included a list of 30 different scenarios commonly performed by the clinical personnel in operating rooms and the ICUs. The test sessions were conducted by a moderator who first read the task scenario and then asked the test subjects to perform it. All sessions were videotaped and members of the Component Monitoring System R&D and marketing teams watched them in a separate room. This way the design engineers got firsthand insights into user reactions to the human interface.

Before each session the moderator gave a very brief explanation about how the monitor works. This demonstration was kept to a minimum to test how easy it would be for a nurse to operate the monitor with almost no previous training. The test subjects were asked before the test what functionality they expected to activate with each hardkey. In this way we got more feedback on how intuitive the Component Monitoring System keypad labeling was.

At the end of each session the test subjects were asked to pretend that they had to train the moderator to do a simple procedure, such as changing the leads on the ECG or adjusting the pressure alarm limits. The purpose was to see if they could recall the procedure they had performed about an hour ago. This was a measure of how well they had learned and how well they understood the operating concept.

The general assessment was that the Component Monitoring System user interface is sound, easy to learn, and effective to use. A significant number of recommendations and problems were found, many of which had not been reported in previous tests with clinical specialists and HP employees. For example, the monitor has a function that allows the user to activate or suspend the monitor's alarming capability. This function was implemented in the prototype as a toggle key. To suspend alarms, the user had to press a softkey labeled Suspend Alarms. The label then changed to Activate Alarms and an ALARMS SUSPENDED message appeared in the upper part of the screen. The subjects frequently overlooked the message. They were therefore confused to see the softkey label changing to Activate Alarms. They were not sure whether the alarms were on or off. Even with extensive explanations, they had trouble understanding the functionality of the toggle softkey. Because this is a critical function that involves patient safety, we separated this function into two separate softkeys.