Biostatistics
Eva Petkova, Ph.D., Director
Mark Davies, M.S., Research Scientist, IV
Steve Ellis, Ph.D., Research Scientist, V
Hanga Galfalvy, Ph.D., Research Scientist, II
Xinhua Liu, Ph.D., Senior Biostatistician
Todd Ogden, Ph.D., Senior Biostatistician
Thaddeus Tarpey, Ph.D., Senior Biostatistician
Haying Zhang, Ph.D., Senior Biostatistician
The Biostatistics Department provides investigators of the New York Psychiatric
Institute access to and training in state-of-the-art statistical techniques as
needed for optimal use of their research data. This includes developing and
applying new methodology for the design and analysis of psychiatric studies,
participating in major funded research projects, teaching statistics to
researchers and fellows of training grants, and providing consultations on
data-analytic and design issues for grant submission and manuscript preparation.
In addition, the department provides continuous education for statisticians and
data analysts working at the Institute.
Department Staff
E. Petkova, M. Davies, X. Liu and T. Ogden hold academic appointments in the
Division of Biostatistics of the Columbia University School of Public Health,
and S. Ellis has a faculty appointment at the Columbia University. In September
2001, Thaddeus Tarpey, Ph.D., was hired to work full-time as a Biostatistician
for the Core Grant to Enhance Neuroscience Transfer (CoGent). Dr. Tarpey’s work
focuses on the application of multivariate statistical techniques to issues
related to the analysis of brain imaging data and its interpretation. H.
Galfalvy has been hired to work on statistical problems in the Neuroscience
Department. Guaguang (Julie) Ma, a doctoral student in the Ph.D. program of the
Division of Biostatistics at the Columbia University School of Public Health,
assisted the Division Staff as a Research Associate. Her duties included complex
statistical modeling and non-standard statistical analysis, consulting under
Biostatistics department staff’s supervision, manuscript preparation, training
of data analysts and network management. On graduating in November 2001, she
left the Department to obtain a full time position in another Institution.
Jianfeng Chen, a Ph.D. candidate of the Division of Biostatistics at the
Columbia University School of Public Health, who has been working part time was
hired full time after Dr. Ma left. The administrative assistant for the
department is Rosalind Russell.
Research
Brain Imaging
Dr. Ellis is developing methods for controlling for multiple comparisons in
looking for suicide effects in many brain regions in autoradiographic human
brain images. Very intensive computing is used to adjust p-values for multiple
comparisons and methods must be developed to deal with such issues as missing
data. Dr. Ellis is also fitting a complex statistical model to understand the
differences in dendrite branching patterns in pyramidal cells among controls and
subjects with mood disorder or schizophrenia.
Dr. Ogden continues his work on developing statistical methods for analysis of
PET data in collaboration with researchers from the Department of Neuroscience.
PET imaging is a useful tool to measure various components of neurotransmission
in the study of depression and other psychiatric disorders. Any of several
different models may then be used to describe the resulting image data. The
estimated model parameters are used in subsequent comparisons among subjects.
Recent work has focused on evaluating models and developing new methodology for
more efficient estimation of model parameters, as well as estimation of the
accuracy of the estimates themselves. By incorporating these new methods, the
resulting analysis can be considerably sharpened.
As part of her work in the Department of Neuroscience, Dr. Galfalvy has begun
research on the analysis of gene microarray data, analyzing autoradiography data
about serotonin and norepinephrin levels in the brainstem for suicide victims
and controls, comparing cortisol levels for depressed subjects with and without
PTSD.
Identification of placebo responders among subjects treated with active drugs
Drs. Tarpey, Petkova and Ogden have developed a method for identifying subjects
treated with drugs who respond to non-specific effects of the treatments
(placebo effect). The method is based on semi-parametric clustering of
functional coefficients using principal points. This work is in collaboration
with colleagues from Depression Evaluation Service Department.
Collaborative and Consultative Activities
Department staff has been very active in collaborative and consultative efforts,
especially with the departments of Therapeutics, Substance Abuse, Research
Assessment and Training, Depression Evaluation Services, Neuroscience, Child
Psychiatry, Lieber Center for Schizophrenia Research, Biological Psychiatry,
Clinical Psychopharmacology and Geriatrics. Department staff provided input on
design and data-analytic strategy for over 20 grant submissions.
Education and Training
Statistics Workshop
As part of its training activities, the department offers the NYSPI
Biostatistics Workshop, a series of seminars covering statistical methodology,
applications, and the use of advanced statistical software. Invited speakers
illustrate statistical and data management issues based on examples from
on-going research at the Institute, explore emerging methodologies, or present
relevant papers recently published in the statistical literature. These weekly
seminars are open to all Institute investigators and their staff members, as
well as students from the Columbia School of Public Health.
Columbia-Penn Forum on Statistics in Psychiatry
The Biostatistics Department is a co-founder of the annual Columbia-Penn Forum
on Statistics in Psychiatry. The Forum is unique in its focus on methodological
and statistical issues pertaining to research in psychiatry. The practice of
psychiatric research has features that make the statistical problems associated
with it distinct from the statistical issues in other areas of medicine, such as
cancer research, cardiovascular research, AIDS and others. The challenges of
biostatistics in psychiatry usually have a low profile at biostatistical
conferences and the Forum is the only meeting where these challenges are
exclusively discussed. The one-day round-table of biostatisticians from the
Institute and faculties from the Department of Epidemiology and Biostatistics at
University of Pennsylvania has become an important tool for collaboration,
development and exchange of ideas related to statistics in psychiatric research.
On April 17, 2001, the Department of Biostatistics hosted the Third Annual
Columbia-Penn Forum on Statistics in Psychiatry. Dr. Ogden presented a paper,
entitled, “Some Current Statistical Issues in Brain Mapping.”
BaDMaN Workshops
As part of the Division of Biostatistics work with the Biostatistics and Data
Management and Networking (BaDMaN) Core, Dr. Petkova organized the BaDMaN
Workshops to further the exchange of information, foster collaboration and
facilitate the development of new ideas between NYSPI researchers,
biostatisticians and data managers. This bi-monthly seminar runs during the
academic year and is open to all researchers at the Institute. Todd Ogden, Thad
Tarpey and Eva Petkova have been presenting at these series. Guest speakers from
the Technology and Information Systems Department have also given invited
lectures. Topics this year included: “Design of a study: Hypothesis (null and
alternative), decisions, power, sample size”, “Multiple comparisons”, “Analysis
of clinical trials with multiple end points: what are the null and alternative
hypothesis and how to choose the most appropriate test”, “Logistic regression
and interpretation of the regression coefficients -- odds ratios, risk ratios
and likelihood of event”, “Survival analysis”, “Factor analysis”, “Principal
components analysis”, “Varying coefficients models”, “Web-enabling complex,
multi-lingual, structured interviews using TRICEPS”, “Bayesian statistics”,
“Non-parametric statistics”, “Maximum likelihood estimation and related tests”,
“Bootstrap and other resampling methods”, “Towards understanding placebo
response among drug treated patients” and others.