#Gardener of Mind
Title: Using Data Analytics and Iterative Program Improvement to Achieve Better Mental Healthcare Outcomes in Defined Populations
PIs:
C. B. Taylor, PAU, Stanford
Michelle Newman, Penn State
Rahul Banerjee, BITS
Note: an R21 is an NIH (U.S.A. National Institute of Health) exploratory/developmental grant mechanism. Obtaining this grant would allow us longer-term funding to develop the same platform together and think about how it can be scaled to not only Indian university students but also other locations and populations.
Aims:
(1) To develop a platform that can be used to develop and evaluate technology- based programs for mental health prevention and treatment.
(2) To determine the feasibility of using personal program experience and data analytics to achieve better engagement and outcomes in defined populations (e.g., universities)
Hypotheses:
Data analytic-driven Individualized Program Development
for improving Engagement and Aims/Outcomes Methodology (IDEA) will
significantly improve user engagement and mental health outcomes
Meaning:
Using individual program usage and acceptability/likeability data
along with mental health outcome data to inform program development will
significantly improve overall user engagement and ultimately mental health
outcomes
Introduction:
-
Anxiety disorders and depression are common and disabling problems among college students
-
While many online interventions are available, they reach a very small percentage when disseminated in a defined university population, and have very low engagement rates. Furthermore,the process of developing interventions is glacial: programs are “fixed”, evaluated in long-term controlled studies, and calculated effect sizes (i.e., how much better off those who use the intervention are compared to those who don’t use it) are often unable to be replicated.
-
Modern intervention programs should be developed with:
a. Attention to user interface and experience
b. Iterative program developed based on user needs, interests, and acceptability/likeability, across components of the program, in addition to ultimate mental health outcomes gathered via ecological momentary assessment (EMA – this just means surveys that are conducted “in the moment”)
c. This process is related but different from adaptive or sequential designs (must explain how it is different)
- To achieve this goal (3), it is important to:
a. Have a software program designed for immediate, accessible data collection
b. Have a flexible authoring system
c. Have an intervention program authoring team committed to the on-going
content adaptation in sequence with program adaptation