Statistical Methods in Psychiatry Program

Overview

The fundamental goal of the program is to foster novel statistical methods and analytical plans that identify and statistically validate biomarkers and novel treatment targets corresponding to psychiatric disorders. The program encourages grant applications that explore parsimonious statistical models, power analyses and analytical strategies and overcome the unique challenges of neuroimaging and psychiatric data. The program is interested in statistically sound novel inferential approaches, efficient methods of designing clinical and/or nonclinical studies suitable to psychiatry with the goal of reducing the burden of mental disorders.

New exploration methods and applications that seek to advance statistical techniques for designing and analyzing studies are important for translational psychiatric research. Also, developing innovative statistical challenges in assessing and fostering the reproducibility of translational mental health studies are essential to the program.

Areas of Emphasis

• Developing advanced statistical applications in designing randomized clinical trials in psychiatric populations; statistical power, sample size determination, methods to improve efficiencies of study design of translational mental health research; methods to handle data anomalies, such as multiple testing, correlation, clustering, heterogeneity and missing data in psychiatric studies
• Innovative application of existing methods and/or development of new statistical methodologies with efficient analytic strategies in translational psychiatry
• Analytical methods of high dimensional and spatiotemporally correlated neuroimaging data; including data-driven clustering methods and/or biomarker identification in the brain
• Statistical research promoting quantitative methods to aid in the design, analysis, and interpretation of clinical, neuroimaging and population-based research studies
• Development and adaption of methods to estimate and improve data generalizability, precision, reliability, and validity
• Methods to estimate and adjust for bias, measurement error, confounding, sampling and non-sampling error in translational psychometric methods
• Novel methods of adaptive study design; sequential hypothesis testing
• Software development for novel analytic methods and secondary data analyses utilizing existing mental health database resources

Contact

Abera Wouhib, Ph.D.
Program Chief
6001 Executive Boulevard, Room 7109, MSC 9637
301-594-9195, abera.wouhib@nih.gov