Year: 2023
Type: dissertation
Abstract: Bipolar disorder affects about 1% of the world population and presents a serious problem for affected individuals in their attainment of educational and other societal goals. Although there has been some investigation into bipolar disorder’s mathematical structure using damped oscillator models, so far there has been little attempt to use this mathematical structure to inform individuals of their mood states and predict mood into the future, much as one may forecast the weather. Methodologies that have been tried for these purposes have been prohibitively expensive and/or invasive and subjectively unacceptable to individuals with bipolar disorder, thus hindering their use in a variety of real-world settings such as schools and clinics. For this dissertation an app was developed called Shiny Mood App that allowed 13 participants to log various facets of their mood state (such as energy levels, sleep, and motivation) each day for 12 weeks and outputted a classification of their mood as manic or not manic, and depressed or not depressed using a form of latent class analysis. Weekly questionnaires were also completed by participants, which included standardized measures of mania and depression. Correlations between classifications and questionnaires were assessed; for the initial model, mania classifications were significantly predictive of mania questionnaire scores when compared using a t-test (p = .001), but depression classifications did not predict depression questionnaire scores (p = .42). Additionally, the mean mania classification across weeks (proportion of time spent manic) was a significant predictor of mania severity as measured by questionnaire score (ASRM; β = .21; R2 = .042; p < .001). A regression approach was used to fit sinusoidal waves to mania and depression classification probabilities. Although these findings are interesting, this study was limited by participant burnout and subsequent low participation from some individuals. Further research is needed to establish the feasibility of using an app-based approach.
Author: Keenan, Elliot Gavin
Institution: Stony Brook University
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7067185759
