Mental health chatbots can help treat depression symptoms, research shows

Even though conversational programs could not deliver personalized advice, they were found to be “encouraging, nurturing, and motivating.”

A team of clinician-scientists from the Nanyang Technological University (NTU) in Singapore have analyzed nine popular mental health chatbots, five of which had at least half a million downloads from leading app stores.

The research covered four free-to-use apps – Marvin, Serenity, Woebot, and 7 Cups – and five apps that required a one-time fee or a subscription to use, namely, Happify, InnerHour, Wooper, Wysa, and Tomo.

All were found to have a “coach-like” quality that effectively engaged people with depression and helped treat its symptoms by guiding users toward mood-boosting activities and exercises commonly used by psychologists and counselors.

While the analysis showed chatbots could engage in empathetic conversations with users, they could not deliver personalized advice – possibly to avoid breaching user anonymity, according to the paper’s first author Dr. Laura Martinengo.

“However, these chatbots could still be a useful alternative for individuals in need, especially those who are not able to access medical help. For some people, it’s easier to talk to a machine than a human being,” Martinengo said.

Consulting a healthcare provider would be an “ideal” option, researchers said. But chatbots and other digital health tools could assist in providing “timely care” to individuals who may be unwilling or unable to seek professional help, according to Professor Josip Car, who led the study.

“There are still a lot of stigmas surrounding mental health disorders, and the COVID-19 pandemic has significantly increased the number of people affected by mental health issues,” Car said.

According to the World Health Organization, depression affects 264 million people globally and is left undiagnosed and untreated in half of all cases.

Although earlier research shows the potential of conversational programs to help people, the NTU team’s focus was the quality and effectiveness of the user-chatbot dialogues, which it evaluated by running scripted conversations.

For that purpose, the team created nine different personas with depressive symptoms. These scripted users reflected different cultures, ages, and genders.

The study was published in the Journal of Affective Disorders.