Change point analysis for longitudinal physiological data: detection of cardio-respiratory changes preceding panic...

Area of Science:

• Psychophysiology
• Time Series Analysis
• Biomedical Data Science

Background:

• Existing statistical methods like ANOVA and mixed models are inadequate for detecting unknown changes in longitudinal psychophysiological data.
• Change point (CP) analysis is a statistical technique designed to identify distinct shifts in time series data.
• Preliminary studies show promise for CP analysis in analyzing functional magnetic resonance imaging (fMRI) data.

Purpose of the Study:

• To apply CP analysis to ambulatory, peripheral physiological data for detecting discrete changes preceding naturally occurring panic attacks (PAs).
• To evaluate the efficacy of CP analysis in identifying physiological alterations specific to the pre-panic attack period.

Main Methods:

• Utilized change point (CP) analysis on longitudinal, ambulatory psychophysiological data.
• Applied the CP method to identify discrete changes in cardio-respiratory data leading up to panic attacks.
• Compared physiological changes during pre-panic attack periods with matched non-panic attack control periods.

Main Results:

• The CP method successfully identified significant cardio-respiratory changes that preceded the onset of reported panic attacks.
• These detected physiological changes were specific to the pre-panic attack period and were not observed during control periods.
• The identified patterns of change aligned with established respiratory theories of panic, suggesting a link between aberrant respiration and panic etiology.

Conclusions:

• Change point analysis is an effective tool for detecting specific physiological changes associated with panic attacks in time series data.
• The findings support the role of aberrant respiration in the etiology of panic attacks.
• CP analysis offers a valuable statistical approach for understanding the dynamics of psychophysiological responses during panic episodes.