Digital Health Platform Reduces Bipolar Disorder Symptoms in More Than Half of Users

First data of its kind reveals that digital self-monitoring combined with suggestions for personalized behavioral changes can relieve troubling psychological symptoms

BOSTON, Sept. 7, 2022 /PRNewswire/ -- juli, the AI-powered chronic condition platform, released the results of a first-of-its-kind analysis demonstrating a 55 percent reduction in the number of people with bipolar disorder who experience symptoms of mania after only eight weeks using the juli app.

Bipolar disorder is a lifelong, recurrent, episodic illness with high rates of hospitalization, suicide and comorbidity. It is characterized by periods of extreme low mood (depression) and elevated mood (mania or hypomania). Some 4.4% of U.S. adults or more than 11 million people are affected by bipolar disorder at some point in their lives, resulting in an estimated economic burden of at least  $195 billion per year.

Despite the prevalence and severity of the disorder, an analysis of bipolar disorder apps published in the International Journal of Bipolar Disorder in 2020, before juli was available, found that there are few evidence-based clinically usable apps available to help with its management.

As of July 2022, 2,987 people with bipolar disorder had downloaded the juli app. Of those, 77% (2,300) confirmed that they had been diagnosed with bipolar disorder by a psychiatrist. More than half indicated they had experienced bipolar disorder for more than five years and 65 percent were seeing a physician regularly.

After using juli for eight weeks:

• 55.5 percent of users who completed baseline and week 8 questionnaires experienced a reduction in mania or hypomania.
• Users' depression symptoms were also reduced by 18.7 percent over the same time period, a change that previous studies have shown to be consistent with feeling better.

"This is a very positive result for a population that has generally been underserved by digital health to date," said Joseph Hayes, MD, PhD, Associate Professor of Psychiatry and Principal Research Fellow at University College London (UCL) and co-founder of juli. "In my clinical work with people with bipolar disorder I've met many who have kept a record of their behavior and its impact on their condition for years, and it seemed to help them to prevent relapse. juli automates that process in a more clinically validated fashion as well as revealing patterns related to such things as weather and pollution that may be non-intuitive to our users."

juli effectively engages consumers to power their own health while offering their healthcare providers insights from sub-episodic health data. It helps people manage not only bipolar disorder but chronic conditions like depression, migraine, asthma, hypertension and chronic pain.

The juli platform does this by aggregating and analyzing data from electronic medical records, smartphones, wearable devices and the environment, as well as patient responses to 3-5 daily questions and bi-weekly clinically validated, disease-specific questionnaires. juli then suggests personalized micro-behavioral changes to help alleviate symptoms of the conditions, and adjusts suggestions based on how users respond.

The current analysis was generated by gathering data from two standard screening and diagnostic questionnaires^[1] commonly used in outpatient settings that are designed to assess the presence and severity of manic symptoms and depression (known as ASMR and PHQ-8). The analysis covers a six-month period and includes results from 208 juli users who were previously diagnosed with bipolar disorder by a physician and who completed baseline and week-8 surveys.

"We believe juli works because it does more than help people track their mood -- it identifies and helps them understand unseen correlations between their symptoms and their behavior based on the information they share, and it alerts their care teams if their health state is deteriorating," said Bettina Hein, juli's CEO and co-Founder. "This analysis of bipolar users is just one of many examples of how juli is helping our users to understand the triggers and levers that help them keep their conditions stable."

• A more detailed report on the analysis is available for download at  https://www.juli.co/press

Predicting mood state with juli data

The research team at juli has also developed machine learning models for predicting users' ASRM and PHQ-8 score from the data collected by the app in the week before the assessment. Both models perform well enough to potentially be clinically useful.

• Predicting ASRM: The machine learning model for predicting whether someone is manic has an ROC curve AUC of 0.76 (a research assessment of correlation showing good discrimination). This model prioritizes data on age, patient reported energy levels and step count, but includes multiple additional variables.
• Predicting PHQ-8: The machine learning model for predicting the severity of depression has an ROC curve AUC of 0.80 (excellent discrimination). This model prioritizes data on age, patient reported mood, days using the app, and step count, but also includes multiple variables.

juli is the leading consumer-grade digital health solution targeting healthcare enterprises and institutions. juli harnesses AI to provide people with chronic conditions and their caregivers powerful insights into behavioral patterns that impact their health, pinpointing triggers that make a complex, multifactorial condition worse and identifying changes that can help avoid or alleviate a flare-up.

For example, juli may use location-based air quality measures to alert users who tend to have depressive episodes when air pollution is bad to go to the gym that day rather than exercise outdoors. Or if the user's phone or wearable indicates they're spending more time in bed while not sleeping, juli may offer sleep hygiene suggestions such as cooling down the bedroom or barring screens from the sleeping area.

juli is currently conducting a randomized controlled trial in collaboration with University College London (UCL) to validate its impact on symptoms and well-being. The study includes patients with either depression or asthma; people interested in participating can enroll at https://www.juli.co/clinical-trials.

juli is an AI-driven next-gen chronic condition management platform for multi-factorial and comorbid conditions, currently covering asthma, migraine, chronic pain, hypertension, depression and bipolar disorder.

Patients and their care teams that use juli can track their conditions easily, understand triggers and get recommendations on how to get better or avoid the next episode. This empowers patients to take an active role in managing their condition and improving their outcomes, helping providers, health systems, insurers and employers lower costs. For more information visit https://www.juli.co/.

^[1] The Altman Self-Rating Mania Scale (ASRM), is a tool commonly used in the outpatient setting to measure symptoms of mania/hypomania, and the Patient Health Questionnaire (PHQ-8) measures depressive symptoms.

Todd Stein
[email protected]
510-417-0612

SOURCE juli