# Efficient statistical methods for assessing dementia risk in Parkinson's disease

> **NIH NIH R01** · UNIVERSITY OF PENNSYLVANIA · 2020 · $311,573

## Abstract

“Efficient statistical methods for assessing dementia risk in Parkinson's disease”
Summary/Abstract:
The proposed R01 grant is in response to PAR-16-260 “Methodology and Measurement in the Behavioral and Social
Sciences (R01)”. Disease-modifying therapies targeting Parkinson's disease (PD) dementia are likely to be most
efficacious before significant cognitive decline has occurred, as has been proposed for Alzheimer's disease (AD). Thus,
cognitive biomarker studies in PD are significant because biomarkers may signal an increased risk of future cognitive
decline prior to measurable impairment on standard neuropsychological testing. Longitudinal design is particularly
desirable because it allows ongoing monitoring of pathophysiological processes associated with cognition and
identification of those biomarkers most sensitive to ongoing or future cognitive decline. A major challenge in longitudinal
biomarker studies is the difficulty in obtaining all biomarker outcomes serially for every participant, due to limitations in
study resources and priorities. Current available statistical procedures such as mixed-effects models ignore missing data,
which results in low efficiency (power) of the analyses in the presence of missing data. Thus, our ability to detect
significant longitudinal changes in biomarkers is limited by the current available statistical methods due to this
inefficiency. This R01 aims to develop more efficient longitudinal methods than the current available methods in the
presence of missing biomarker outcome or covariate data. The new methods will require less biomarker data than current
methods to achieve the same analytic statistical power (efficiency). This will be a significant methodological advance, as
it will reduce future study costs and patient burden without sacrificing power. It has broad applications in PD dementia
and other neurodegenerative diseases such as AD, as well as general biomedical research. We also plan to study
progression of three potential cognitive biomarkers (cerebrospinal fluid [CSF], brain MRIs, and dopamine transporter
[DAT] SPECT imaging) and establish their temporal ordering in relationship to cognitive decline in PD participants in the
Parkinson's Progression Markers Initiative (PPMI) study by applying these new statistical methods. The results will
inform the design of future studies testing possible disease-modifying therapies in treating PD dementia.

## Key facts

- **NIH application ID:** 9925847
- **Project number:** 5R01NS102324-04
- **Recipient organization:** UNIVERSITY OF PENNSYLVANIA
- **Principal Investigator:** DANIEL WEINTRAUB
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $311,573
- **Award type:** 5
- **Project period:** 2017-07-01 → 2024-04-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/9925847

## Citation

> US National Institutes of Health, RePORTER application 9925847, Efficient statistical methods for assessing dementia risk in Parkinson's disease (5R01NS102324-04). Retrieved via AI Analytics 2026-05-24 from https://api.ai-analytics.org/grant/nih/9925847. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*