AI tool improves treatment for people living with depression and anxiety, study shows

Smartphone app boosts communication between clinician and patient and provides ongoing insights into mood and medication

An AI-powered smart-phone app that enables doctors to monitor their patients’ symptoms, and their response to medication between appointments, shows promise as a new tool in the treatment of depression and anxiety, according to a recent research study.

Funded by Brain Canada/Bell Let’s Talk, researchers at McGill’s The Douglas Hospital Research Centre in Montreal joined a nine-centre international feasibility study in 2022 to determine if psychiatric care can be personalized and managed through the use of new technology.

They recruited seven clinicians and 12 patients to test an app, developed by a startup in Quebec, that enabled patients to provide ongoing information to their physicians to improve management of their illness and medications.

Participants at five mood disorder clinics in Canada and four in the U.S. were randomized to assess the effectiveness and safety of the clinician decision support system as part of the study called AID-ME (Artificial Intelligence in Depression – Medication Enhancement.)

“Right now, in providing care for patients with depression and anxiety, doctors are engaging in trial-and-error to find the right medication,” says Dr. Manuela Ferrari, Associate Professor in the Department of Psychiatry at McGill University and a researcher at the Douglas Hospital Research Centre, the study leader. “This is frustrating, takes a lot of time, and it can be debilitating for the patient.”

Using a closed artificial intelligence (AI) system, the app’s algorithm asked the patients questions about their mood based on a standardized questionnaire, and helped predict how someone was responding to medication. Physicians were able to review the responses and gather more information than a during a typical patient visit to a clinic.

“Technology was not making the decision for the clinician, just providing insight on what was happening,” Dr. Ferrari says. This allowed physicians to more quickly decide whether a patient was being prescribed the right drug for their illness, and whether someone was at risk and needed immediate intervention.

Of the 74 patients who took part in the feasibility study, 61 (42 active and 19 active-control) completed the study. Of those, remission from disease was almost 30 per cent in the group that used the app, compared to none of the controls. According to a recent publication by the research team, “this is the first effective and safe longitudinal use of an artificial intelligence-powered clinician decision support system to improve major depressive disorder outcomes.”

Dr. Ferrari says the feasibility study wrapped up in late 2023, researchers analysed the data and recently published their findings. “We know that the technology was effectively used and welcomed by clinicians and clients, and did not compromise the health-care experience, and was not harmful to patients.”

Data gathered from the feasibility study will be offered in open science and made available to other companies developing similar technologies.

Dr. Ferrari says developing and approving new technologies for patient diagnosis and treatment is a lengthy process and requires significant investment, but the study shows that an app that provides an ongoing check-in between clinicians and patients is effective and should be integrated into future treatment protocols.

We need to extend care outside the walls of traditional health-care settings, and outside these one-time interactions. We need to be present where the person is. And this is where technology can help us.

Dr. Ferrari