Etiometry T3 Heuristic Evaluation
TimeThursday, April 152:16pm - 2:17pm EDT
DescriptionPROBLEM STATEMENT: Etiometry has developed physiological monitoring data aggregation and display software, called T3. This software presents data gathered from a variety of sources such as pulse oximeters and EKGs, and two novel risk indexes to clinicians. In particular, T3 predicts and displays the probability that a patient is experiencing inadequate oxygen delivery or inadequate carbon dioxide ventilation by computing the Inadequate Delivery of Oxygen Index (IDO2) or Inadequate Carbon Dioxide Ventilation Index (IVCO2), respectively. It is important to ensure that potential users understand how to appropriately interpret and interact with Etiometry’s T3 Software to support clinical decision-making. T3 pulls in data from many different sources; it needs to present this data, and the risk indexes that it computes, in an easy to use and easy to understand interactive display.
OBJECTIVE: Conduct a heuristic evaluation to identify potential usability issues with T3 software, to help ensure and/or improve usability.
METHODOLOGY: Three students with prior experience conducting heuristic evaluations applied the 14 medical device usability heuristics developed by Zheng et al (2003) to assess usability of T3 software.
RESULTS: The heuristic evaluation revealed both strengths and weaknesses of T3. These were reported back to Etiometry for consideration as they continue to update their software. In particular, some of the positive features include:
1. The day/night toggle reduces light emitted by the screen making it easier for patients to sleep.
2. The search bar sorted all the information regardless of entered text. It was efficient and helpful.
3. On the Census Overview table, the columns had up and down arrows indicating the sort preference. The selected one would highlight when clicked. This is like other table software making it easy for most users to understand immediately.
Some of the recommendations for improvement include:
4. The red and green graphs make it extremely difficult for users who are red/green colorblind.
5. The help manual needs a picture with labeled buttons and descriptions that a user can reference.
6. There needs to be a visual distinction between interactive text and informational text.
Next steps include obtaining clinical expert input to help prioritize the suggested changes based upon this evaluation.