Objective/Rationale:
Currently, the diagnosis of Parkinson disease (PD) relies on recognition of impaired movement, which occurs decades after neurodegeneration begins. By looking at changes in sleep, mood, bowel and bladder habits, blood pressure or substances in blood in people prior to their PD diagnosis, we may be able to identify non-motor features or biomarkers which allow earlier diagnosis or treatment. We will look for earlier markers of PD in the Cardiovascular Health Study (CHS).
Project Description:
CHS is a project which has followed 5,888 people aged 65 years and older since 1989. Over 150 participants have gone on to develop PD. Although CHS was originally designed to study risk factors for heart disease and stroke, the data gathered is useful in studying PD (e.g., medical histories, physical examinations, blood samples). We propose classification tree analysis as a novel approach to analyzing this wealth of data. This process generates clinical decision rules physicians and researchers can use for categorizing people according to their risk of developing PD. This approach would result in a decision tree that would identify important pieces of information prior to someone being diagnosed with PD that could influence their chances of developing motor signs of PD.
Relevance to Diagnosis/Treatment of Parkinson’s Disease:
Currently, there is nothing we can do for someone before they are diagnosed with PD. Our study is three fold: (1) it will provide a framework clinical decision making prior to PD diagnosis that can be further tested and refined in other patient populations; (2) lead to clinical research, including interventions, in the period prior to movement problem developing (e.g. trials for neuroprotective agents, neuroimaging, biomarker development); and (3) generate hypotheses for potentially novel and modifiable risk factors of PD.
Anticipated Outcome:
The decision tree generated would be a tool to identify those at risk to develop motor signs of PD. This would: (1) establish a powerful strategy to study PD by selecting cohorts of individuals that are at higher risk to develop PD; (2) enable counseling of patients prior to motor signs and offer surveillance for PD; and (3) allow selection of participants for PD trials which could potentially prevent motor signs altogether.
Final Outcome
We identified 209 of 5,888 people in the Cardiovascular Health Study (CHS) as having PD, of which 44 had PD at the beginning of the study, and 165 were diagnosed later. We looked at Medicare claims linked with CHS within five years and one year prior to diagnosis. Within five years of diagnosis, compared to non-PD participants, we found that those who would go on to be diagnosed with PD had more claims for gastrointestinal, sleep, genitourinary, and sweat dysfunction. Within one year prior to diagnosis, the PD group had more claims for cognitive, psychiatric and dizziness issues. This is consistent with current models of PD that suggest the gastrointestinal tract and sleep centers of the brain are affected early in PD, while areas affecting cognition and psychiatric issues are affected later. We further explored our finding using data collected in CHS. We found that cognitive function, body fat (affected by gastrointestinal dysfunction), and cardiovascular issues that may underlie dizziness (carotid stenosis and ECG abnormalities) were all significant risk factors for PD. Whether these factors cause PD or are very early features of PD prior to motor problems requires further investigation. We submitted an NIH R01 grant application to further fund this work.