To determine whether particular dysautonomia measures predict later Parkinson's disease (PD) phenoconversion and functional dependence in populations without PD but with other PD risk markers, and whether specific risk markers modify associations between dysautonomia and risk of the outcomes.

Identifying populations that will develop PD is important for recruitment into studies of putative disease-modifying therapies. Dysautonomia is a prodromal feature of PD that may help risk-stratify individuals in the presence of other risk markers.

Data were obtained from Parkinson’s Progression Markers Initiative; cohorts without PD but with PD risk markers (hyposmia and/or RBD, LRRK2 gene mutation, GBA gene mutation) were included. Dysautonomia measures included Scales for Outcomes in Parkinson’s Disease Autonomic (SCOPA-AUT), 7 SCOPA-AUT subscales, +/- neurogenic orthostasis, +/- supine hypertension, +/- low pulse pressure. Outcome measures were phenoconversion and Schwab and England Activities of Daily Living Scale (SE-ADL) ≤70. Cox proportional-hazards regression was used to evaluate phenoconversion/SE-ADL≤70 hazard for each dysautonomia measure, adjusting for relevant demographic/clinical variables in multivariable models. If a significant association was identified, a likelihood-ratio test was employed to evaluate for an interaction between the measure and cohort. If an interaction was identified, regression analysis was repeated stratified by cohort.