High frequency data collection in a participant’s usual environment (ecological momentary assessment, or EMA) offers a promising approach for understanding, and potentially intervening on, the precursors of clinically important states and behaviors in mental health disorders.
UCSD researchers have developed a “mhealth” method and companion software, which employ mobile phones/computers to collect self-reported data from patient and apply real-time statistical analyses within the individual to provide predictive projections to tailor behavior change interventions. The innovation uses (1) decision logic attached to individual elements of the survey to assess the momentary data on precursors to events such as behavioral crises, (2) the employment of statistical learning analyzes patterns of data for decision rules and branching to project values of variables in the future, and (3) the dynamic interaction with the server to determine which responses to send to intervene upon predicted trajectories, allowing the content presented to users to be highly personalized and preventative.
The technology is useful for real-time data collection and associated public health interventions that can be delivered dynamically and outside of the clinical setting for a variety of behavioral assessment and behavioral change purposes, either for research studies, clinical programs, or commercial behavioral assessment/intervention uses.
The technology has been applied to the self-monitoring data collected from 50 patients with bipolar disorder. The results indicate that the method is useful for identifying future clinically important states in some individual patients.
ecological momentary assessment, mental health, bipolar, behavior monitoring