Michael Donohue, PhD

Title(s)Associate Professor of Neurology
Phone+1 858 964 0790
vCardDownload vCard

    Collapse Overview 
    Collapse Overview
    Dr. Donohue’s methodological interests include semiparametric and generalized linear mixed-effects models, model selection, asymptotics, and clinical trials design. His applied work includes the design, monitoring, and analysis of clinical trials and observational studies in the fields of Alzheimer’s disease, anxiety disorders, fMRI pain response, smoking cessation, and childhood obesity.

    Dr. Donohue received his PhD in Mathematics from UCSD in 2005. His thesis work involved related problems in rank regression and synergy detection. He proposed a procedure for fitting a regression surface with linear level sets, which is useful for determining synergy between two or more agents. More recently he has been investigating model selection and asymptotics for generalized linear and proportional hazards mixed-effects models. He received a KL2 Post-Doctoral Scholar Award to investigate efficient analysis methods for longitudinal clinical trials, particularly in the face of missing data typical in Alzheimer’s disease trials.

    Dr. Donohue was a senior statistician on two observational studies: Alzheimer’s Disease Neuroimaging Initiative (ADNI) and Home Based Assessments (HBA). Both studies are designed to improve the statistical and economical efficiency of future Alzheimer’s disease clinical trials. He has been working on power and study design for prodromal or asymptomatic Alzheimer’s disease incorporating recent advances from ADNI in neuroimaging and fluid assay biomarkers.

    Collapse Research 
    Collapse Research Activities and Funding
    Estimating Long-Term Disease Trajectories from Short-Term Data
    NIH/NIA R01AG049750Feb 1, 2016 - Jan 31, 2020
    Role: Principal Investigator