Recognizing the rapid adoption of voice technology in various scientific fields, scientists of Janssen Research and Development built a voice-enabled software platform to explore the potential of using “people’s number one communication tool” in clinical trials.
The software application, which runs on smartphones, is currently being tested by a group of patients with chronic conditions receiving treatment at Boston’s Brigham and Women's Hospital.
Senior Director of Clinical Innovation at Janssen Research & Development, Bert Hartog, told us that the platform aims to “introduce an additional channel to replace some of the interaction that takes place, and make it more flexible.
Voice-enabled technology is used to collect information from patients, with questions programmed to be ‘asked’ by the application at specific times and answered by the patient verbally; For example, regarding the pain that they are experiencing at the time.
“The answer is recorded and algorithms can either provide immediate follow-up to the patient or capture the recording for future analysis,” said Hartog. The voice-enabled application makes the process “less dependent on the direct involvement between patients and healthcare professionals in a hospital setting,” he said.
Additionally, the platform has the potential to provide patients with real-time answers to questions regarding their disease state, the clinical trial that they are participating in, or their scheduled visits.
Testing of the platform began in spring 2019, and to date, the application has been used by approximately 40 patients with immunological disorders aged from 30 to over 70 years old.
A lower threshold for sharing
The feedback that scientists received during the study was quite encouraging, with patients welcoming the approach, but it also led to ‘unexpected’ findings, according to Hartog, as patients expressed a higher level of comfort sharing some of their feelings with the voice technology than they would with a nurse or doctor.
“That came ‘out of the blue’ and we will need to see how it works in larger groups of people,” Hartog told us.
As he noted, in speaking with a physician or nurse, certain topics may be easier to discuss than others, “and personal feelings may not be the easiest to relay in this type of setting.”
In comparison, Hartog said “a voice technology setting is more objective and neutral, therefore patients expressed a lower threshold in sharing some of what was on their mind.”
The researchers are planning to explore the finding further, since “that may actually help to really understand what is going on in the patient's mind.”
In search of the preferable persona
During the study, researchers of Clinical Innovation at Janssen R&D were also led to another surprising outcome, with patients showing ‘very specific’ preferences as it pertains to the voice ‘persona.’
“Some people were very explicit saying that they prefer a male voice over a female voice, while some others preferred a voice that is rather neutral and not really emotional,” Hartog explained.
As a result, researchers are exploring how they can respond to a patient’s preference and create a “neutral persona that is acceptable and doesn’t upset anyone. Because in a clinical trial you cannot always offer many different voice personas,” he added.
Challenges of personalization
Asked about potential barriers to the use of the application by patients with speaking difficulties, Hartog said it has been recognized as a potential challenge.
“The beauty of voice is that everybody is using it; it is the number one communication tool for people. But everyone's voice is different,” said Hartog, explaining his team’s aim for a better understanding of the limitations.
“We need to understand to what extent we can use voice across a very large, diverse population, without having to go into each possible individual nuance that we may encounter.” That, according to Hartog, would make the implementation in a large international trial setting ‘very complex,’ and may diminish or reduce some of the benefits of the technology.
Over the following months, researchers expect to examine long-term exposure data, with a second study on the voice-enabled application scheduled for the first half of 2020.
The company anticipates it will be ready to provide the technology for use in clinical studies by the end of next year.