• Researcher Profile

    Paul J. Catalano, ScD

     
    Paul J. Catalano, ScD
     
    Senior Lecturer in Biostatistics, Harvard T.H. Chan School of Public Health

    Office phone: 617-632-2441
    Fax: 617-632-2444
    Email: pcata@jimmy.harvard.edu

    Preferred contact method: office phone
     
     

    Research Department

    Biostatistics and Computational Biology

    Area of Research

    Biostatistical Methods for Multiple Outcomes and Clustered Data


    Dana-Farber Cancer Institute
    450 Brookline Avenue
    CLSB 11007
    Boston, MA 02215

    Biography

    Dr. Catalano received his ScD in biostatistics from Harvard School of Public Health in 1991. He then completed a two-year postdoctoral fellowship there and at DFCI, joined the faculty in 1993, and became Associate Chair of the department in 2001 and Senior Lecturer in 2005. In 2007 he became Director of the Biostatistics Core Facility of the DF/HCC. His collaborative work is primarily in radiation oncology, radiology, outcomes research and cancer biology with investigators at the DF/HCC, cancer clinical trials and translational studies in gastrointestinal malignancies with the Eastern Cooperative Oncology Group and quality-of-care research in lung and colorectal cancer with the CanCORS Consortium. His methodological research is in the areas of multiple outcomes modeling and methods for hierarchical clustered data.

    Research

    Biostatistical Methods for Multiple Outcomes and Clustered Data

    The main line of my methodologic statistical research is the design and analysis of studies involving mixtures of discrete and continuous outcomes, with a focus on dose-response modeling and quantitative risk assessment. Toxicologic and other early evaluation studies in animals are a specific application area, but problems related to multiple outcomes arise in a variety of research contexts.

    One important application of such methods is the determination of overall risk from an exposure to an agent (pharmaceutical, environmental contaminant, etc.) when the outcomes are naturally multivariate. Although assessing probabilistic risk is usually straightforward with a single outcome, in the multivariate setting the situation is often more complex because outcomes may be observed simultaneously or in a hierarchical fashion. The statistical analysis must carefully account for these features of the response profile.

    My collaborative research in cancer involves statistical support for the design, monitoring, and analysis of clinical trials in gastrointestinal malignancies, design and analysis of prospective and retrospective studies in radiation oncology, radiology, outcomes research and cancer biology, collaboration on studies with quality-of-life and translational-science endpoints, and analysis of all phases of clinical trials and quality of care studies in lung cancer and colorectal cancer.

View Physician Directory

Find a Clinical Trial

Support Cancer
Research

Give Now