3 - (SYM 53) Applying Novel Group and Idiographic Modeling of Menstrual Cycle Exacerbation to Adults with Elevated Borderline Personality Disorder Symptoms

Background: Individuals with borderline personality disorder (BPD) suffer from a constellation of rapidly shifting emotional, interpersonal, and behavioral symptoms. Understanding how fluctuations in ovarian hormones across the menstrual cycle may contribute to symptom expression and instability among ovulating people with this disorder is key for accurate assessment of BPD symptoms and effective implementation of behavioral interventions.

Methods: Data presented are findings from an ongoing study examining fluctuation in a wide range of psychological symptoms across two menstrual cycles in ovulating females with clinically significant (3+) BPD symptoms (N=107 at present, expected ~135 when re-analyzed for this presentation). Participants may also meet criteria for most other disorders, and our sample includes a wide range of diagnoses. Participants completed diagnostic interviews at baseline and reported daily emotional and behavioral symptoms across at least two menstrual cycles. Urine LH surge testing was used to confirm ovulation across the study period, and daily levels of metabolites of progesterone (P4) and estrogen (E2) were assessed via cutting edge urine testing technology.

Results: Compliance rates for daily surveys were high (93%). Data were scaled across the cycle using novel methods (Nagpul et al, under review) and generalized additive mixture models (GAMMs). Across the full sample, significant luteal and/or perimenstrual worsening was observed for depression, anger, irritability, anxiety, mood swings, and suicidal ideation, with some differences by symptoms in onset/offset timing. In contrast, positive affect demonstrated ovulatory surges. Individual-level generalized additive models (GAMs) were used to determine frequency of significant mood exacerbation. Rates of exacerbation of specific symptoms based on thresholding for statistical significance and degree of cyclicity range from 15-38%. Descriptives for peak symptom timing and correlations between exacerbation patterns and clinical diagnoses will be presented.

Conclusions: These results not only provide evidence from the largest sample to date on menstrual cycle exacerbation of BPD symptoms, but also demonstrate feasibility of new analytic approaches with potential for a wide range of research and clinical applications. We will discuss implications of these findings for assessment and behavioral skills training for hormone sensitive clients.