Text message responsivity in a 2-way short message service pilot intervention with adolescent and young adult survivors of cancer

Background: Text message interventions hold promise for adolescents and young adults (AYAs) with chronic health conditions, including childhood cancer survivors; however, engagement is often suboptimal. Limited research has studied mobile health intervention outcomes beyond efficacy. Understanding responsivity to different types of text messages (ie, when a participant texts back) can provide practical, actionable information to optimize engagement in future projects. Objective: Within a 2-way text messaging study in AYAs who recently completed treatment for cancer, we sought to evaluate text message responsivity across different types of text messages. Methods: AYAs who recently completed treatment for cancer (n=26; mean age=16 years; 62% female, 16/26 participants) received 2-way text messages about survivorship health topics over a 16-week period. Using participants’ text message log data, we coded responsivity to text messages and evaluated trends in responsivity to unprompted text messages and prompted text messages of varying content (eg, medication reminders, appointment reminders, and texts about personal experiences as a cancer survivor). Results: Across prompted and unprompted text messages, responsivity rapidly decreased (P ≤.001 and =.01, respectively) and plateaued by the third week of the intervention. However, participants were more responsive to prompted text messages (mean responsivity=46% by week 16) than unprompted messages (mean responsivity=10% by week 16). They also demonstrated stable responsivity to certain prompted content: medication reminders, appointment reminders, goal motivation, goal progress, and patient experience texts. Conclusions: Our methodology of evaluating text message responsivity revealed important patterns of engagement in a 2-way text message intervention for AYA cancer survivors.