Understanding vulvodynia: Using machine-learning-based modeling to identify factors associated with resolution and persistence of vulvar pain

Vulvodynia affects 8-15% of women and is one among the identified chronic overlapping pain conditions (COPCs). The condition is poorly understood by clinical providers; people seeking care for vulvodynia see a mean of 3 providers before receiving a diagnosis. Vulvodynia is different than many chronic pain conditions because it is often highly associated with sexual functioning and results in a conflict between the desire for intimacy and pain. Similar to other COPCs, vulvodynia is associated with higher levels of depression, however it is unclear whether the presence of depression influences progression or persistence of symptoms. People with vulvodynia feel shame and stigma around their diagnosis, and have difficulty discussing it with partners, providers, and friends. In our experience, patients with vulvodynia want to know whether and when their pain will go away and whether the presence of other comorbid conditions like depression will impact the trajectory of their pain, however there are limited longitudinal data on which to base answers to those questions. We will prospectively collect data on multiple components of the vulvodynia pain experience and persistence of symptoms over 2 years. A cohort of 200 women ages 18-45 with vulvodynia will be followed, with monthly surveys and three separate in-person assessments over two years including at each timepoint: 1) Standardized questionnaires including HEAL Common Data Elements, assessment of treatment modalities, and physical exam, 2) collection of vaginal swabs for microbiome and inflammation, 3) Quantitative Sensory Testing (QST) to assess central sensitization, and 4) 2 weeks of ecologic momentary assessment (EMA) and passive activity data collection from a mobile device using the open source mindLAMP platform. We will then use statistical models and machine learning to 1) define the variability over time of pain and associated predictors, 2) identify how depressive symptoms impact associations between predictors and pain, and 3) identify predictors of pain resolution. We will train the mathematical model using data from the first 100 participants and will validate the model with data from the second 100 participants. The model will also be used to test what modifiable factors have the largest impact on pain trajectories. Collection of both symptomatology and a wide range of phenotypic patient characteristics in a longitudinal manner will for the first time capture an integrated view of how pain experience, depressive symptoms, and central sensitization coevolve in the progression of vulvodynia. Project Number: 1R01HD117068-01A1 | Fiscal Year: 2025 | NIH Institute/Center: Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) | Principal Investigator: Caroline Mitchell (+1 co-PI) | Institution: MASSACHUSETTS GENERAL HOSPITAL, BOSTON, MA | Award Amount: $4,867,750 | Activity Code: R01 | Study Section: Special Emphasis Panel[ZRG1 CCHI-W (57)] View on NIH RePORTER: https://reporter.nih.gov/project-details/1R01HD11706801A1