Am I Stressed?
Wouldn’t it be great if your computer knew when you were stressed? I am sure you can think of a few use cases where this could be helpful. My current favorite one is that my smartphone figures out that I am stressed and offers to play a clip of my twin daughters laughing uncontrollably at each other from when they were ten months old. What’s also interesting about that scenario is that even if I don’t act on my phone’s suggestion and play the clip, I know that just thinking about that moment will reduce my stress levels. I will come back to why this is important later in the post when I discuss clinically proven stress reduction techniques.
One non-trivial hurdle in realizing this vision is that there isn’t much consensus in the expert community on exactly how to measure stress. Actually, in my literature review, I couldn’t really find clear consensus on which expert community is best equipped to answer this question. Think about that: other than yourself, whose diagnosis do you trust the most when it comes to evaluating your stress levels? Is it your primary care physician, psychologist, psychiatrist, yoga instructor, massage therapist, life coach, best friend or your significant other? I got quite a wide variety of answers when I brought it up with my colleagues at Georgia Tech. Each conversation I had on this topic was an interesting one and I would encourage you to pose this question to people in your network and post your findings in the comments section below. One common theme that did emerge in my interviews was that most of my colleagues mentioned at least one nonprofessional, usually a family member or friend, whose judgment they would trust on this subject.
The reason I have been studying stress reduction literature for the last year is not purely based out of personal curiosity. Here at IMTC, we partner with Georgia Tech faculty and external clients on technology driven research projects. Last fall, our health tech team was tasked by Sharecare Inc. to independently conduct a clinical trial with the primary goal of evaluating the stress reduction capabilities of their novel voice analysis platform. Amongst other functionality, their technology utilizes artificial intelligence to sense stress indicators present in voice activity and is currently available in the form of a beta Android smartphone app in the Google Play store. The technology does not listen to, record, store or analyze the actual content of the call; and the user can opt-out at any time. In its current form, the app presents the user with a visual representation of stress intensity (ranging from calm, productive, uneasy, intense and very intense) and a short phrase describing the stress type (irritation, impatience, anxiety, worry, nervousness and discomfort) after every incoming or outgoing call. This is also followed by a one or two sentence summary report of their stress. For instance, a summary report I received for a call with Comcast customer support regarding my impending move to a new house read: “You appear to be trying to suppress a sense of concern. Your attitude seems to be associated with outcomes, timing, etc.” Needless to say I was intrigued.
After the necessary paperwork regarding non-disclosure was in place, in preparation for the clinical trial, Sharecare shared some under-the-hood data from the app with our research team. Specifically, we looked at a log of usage data from 98,603 calls made by 1657 users. Interesting findings included the detection of a downward trend for the stress intensity score generated by the app for users who made their first 100 incoming and outgoing calls as well as for users who made their first 100 outgoing calls. The trend line for stress intensity for the first 100 incoming calls alone was almost flat. These differences in the slope of the stress intensity score between all calls (incoming + outgoing), incoming calls and outgoing calls are thought-provoking when you look at them from the lens of Mindfulness Based Stress Reduction research, but more on that later. As part of this project we also got a chance to look at some scores that are generated by the app behind the scenes but are not displayed to the end user as of yet. One such score was behavioral drivers which consisted of three dimensions: confidence, emotionality, and effort to make contact. Our team discovered statistically significant changes in the makeup of the behavioral driver score for incoming and outgoing calls as well..
So how does any of this help you find out if you are stressed? The short answer is that if you enroll in our clinical trial we will provide you with the current state of the art tools to scientifically measure your stress over a 10-week period. This will include validated questionnaires, symptom checklists and lab tests for physiological bio-markers of stress in the various phases of the trial. Some participants will also receive access to clinically proven stress reduction therapies like Mindfulness Based Stress Reduction (MBSR) for comparison. This program was developed at University of Massachusetts Medical Center and has been consistently documented as highly effective in stress management. It is based on the ability to achieve a mental state of mindfulness which is broadly defined as the act of paying attention in a particular way: on purpose and in the present moment. One of our research questions, based on the differences in the app generated stress scores between call types, is if the utilization of the Sharecare app in a particular manner can reproduce the mental state associated with MBSR. This question is an important one because not only will it help us in realizing the vision of our computers knowing when we are stressed but also lead us towards answering what they should do about it. For example, would it be effective for my phone to offer me a video clip of my daughters to reduce my stress or would just my awareness of the fact that it could do that already get me to a stress free mental state?
If thinking about all of this is giving you stress, then please keep calm and participate in our clinical trial. You can apply to do so by contacting us through our Contact IMTC form and entering the phrase “Clinical Trial” in the Subject field.