“I Don’t Like Being Told Just What to Do; I Need to Know Why”: Patient Expectations of Machine Learning-Driven Just-in-Time-Adaptive Interventions (JITAIs) for Prenatal Stress Management

Just-in-time-adaptive interventions (JITAIs) have advanced the ways in which technology supports behavior change, providing tailored interventions at the most appropriate time. With machine learning (ML), these interventions are becoming increasingly sophisticated. However, significant efforts are required to fully understand and address individual expectations of ML-driven JITAIs. In this work, we recruited 20 pregnant people to explore their interest in and expectations of a JITAI designed for prenatal stress prediction and prevention, an area of global public health significance. Using design sessions that combined reflection and design feedback activities, we found that participants were interested in using ML-driven JITAI designs that (1) fit into their daily life by offering flexible engagement, (2) supported them in building a mental model of the underlying JITAI functionality, (3) differentiated between non-adherence and non-compliance, and (4) illustrated apparent reciprocal learning. We use these results to propose designs for future iterations of prenatal stress interventions, particularly those leveraging increasingly complex computational technologies.