Feasibility and Acceptability of an Artificial Intelligence (AI-) Enabled Distress Monitoring Tool for Adolescents and Young Adults with Cancer: A Clinial Study

Background. Adolescents and young adults (AYAs) diagnosed with cancer are an age-defined population, with studies reporting up to 45% of the population experiencing psychological distress. Although it is essential to screen and monitor for psychological distress throughout AYAs’ cancer journeys, many cancer centers fail to effectively implement distress screening protocols largely due to busy clinical workflow and survey fatigue. Recent advances in mobile technology and speech science, including artificial intelligence, have enabled flexible and engaging methods to monitor psychological distress. However, patient-centered research focusing on these methods’ feasibility and acceptability remains lacking. Therefore, in this project, we aim to evaluate the feasibility and acceptability of an artificial intelligence (AI)-enabled and speech-based mobile application to monitor psychological distress among AYAs diagnosed with cancer. Methods. In this study, we use a single-arm prospective cohort pre-test-post-test design with a stratified sampling strategy. We recruited 60 AYAs diagnosed with cancer and monitored their psychological distress using an AI-enabled speech-based distress monitoring tool over a six-month period. Primary outcomes of the study include both feasibility and acceptability endpoints. The primary feasibility endpoint of this study is defined by the number of participants completing four out of six monthly distress assessments; the acceptability endpoint is defined both quantitatively using the acceptability of intervention measure and qualitatively using semi-structured interviews. Given the nature of the study, statistical analyses remained at the descriptive level, and we present interim analysis findings in this abstract. Results. Fifty-five study participants enrolled in the study for an average of 2.35 months (SD = 1.40), including 6 participants enrolled for 5 months and 6 for 4 months. To date, 26 participants engaged with the app at least once, leaving 29 participants who did not have an engagement with the app by the end of study month 5. AYAs with cancer reported an average of 1.509 engagements with the application. Notably, among participants who had least one engagement with the app (n = 26), they reported an average engagement with the app of 6.077 times, suggesting 2.59 times of engagement with the app per month. For those participants who used the AI app at least one time, the average time engaging the app was 3.04 times (SD = 2.849), ranging from 1 to 11 times. Importantly, out of the 26 participants who engaged with the app for at least once, 25 completed both AI screening and electronically administered PHQ-9 and GAD-7 during their engagement. Conclusions/Implications. Findings from the interim analyses suggested that an AI-enabled mobile app is feasible and acceptable among AYAs with cancer for distress monitoring. Notably, AYAs with cancer who engaged with the mobile app at least once are more likely to keep using the tool, evidenced by significantly higher instances of utilization. A more rigorous design, i.e., a randomized controlled trial, should be a next step to formally evaluate the tool’s superiority in distress monitoring; and to assess the clinical referral rate for psychosocial care among patients using the mobile app versus traditional methods.