Postpartum Support-Bot: An Ethically Co-Designed Chatbot for Mothers Experiencing or at Risk of Postpartum Mood and Anxiety Disorders

Overview 

Postpartum mood and anxiety disorders (PPMADs) affect up to one in five women globally and are associated with many adverse effects on child development across cognitive, motor, and mental health domains. As such, simple, cost-effective, and efficacious approaches to providing PPMADs support are urgently needed.  

Recently, one-on-one digital helplines have emerged as an avenue of support for caregivers suffering from PPMADs. In this work, we seek to understand support seekers’ experiences and needs as shared on one such helpline with an ultimate goal of developing a “chatbot” application to provide meaningful support to caregivers struggling with or at risk of PPMADs. This work is done in collaboration with the international non-profit Postpartum Support International.  

Initial Findings 

We conducted a descriptive analysis of the concerns, psychological states, and motivations shared by distressed postpartum support seekers on the PSI helpline. Using a combination of human annotations, dictionary models and unsupervised techniques, we find distressed mothers described interpersonal problems and a lack of support most frequently, with 8.60% reporting severe symptoms, such as suicidal ideation. Our results provide new insights into the experiences of caregivers suffering from postpartum mental health distress, and will be fundamental in designing the next generation of support tools for postpartum parents.  

Additionally, we leveraged human-centered and ethical design principles to obtain a nuanced profile of support-seeker user personas, experiences, and motivations. Working closely with PSI, we used these materials to co-develop chatbot wireframes, a blueprint-like design of the application’s interface.