Abstract:
|
There has been considerable progress in understanding the biology of Parkinson's disease (PD). Reliable biomarkers are still lacking, however, for early stage detection of PD and for characterizing disease progression. Substantial neurodegeneration occurs prior to the onset of hallmark motor symptoms of PD that lead to diagnosis. There is a pressing need for earlier detection of characteristics reflecting high risk of PD, creating an optimal window for therapeutic intervention. We consider electronic medical records from an integrated healthcare system to determine clinical precursors to PD. We analyze longitudinal data from PD patients and matched control subjects to identify commonalities in medication history, previous diagnoses, and laboratory results. Both hypothesis based features with known relevance to premotor PD, as well as other features with less established connection to the PD phenotype are analyzed in our models. We identify a clinical risk profile that suggests an elevated risk of PD, which through external validation could have a tremendous impact on patient management and the design of future clinical trials.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.