SECTION NEWS
Biometrics
Meetings Dot Section Landscape
Jerry Heatley, Section Continuing Education Chair, and Page Moore, Biometrics Section Publications Officer
The Biometrics Section is proud to cosponsor the following four short
courses during JSM 2009 in Washington, DC:
Missing Data in Longitudinal Studies: Strategies
for Bayesian Modeling and Sensitivity Analysis
Instru ctors: Michael Daniels of the University of Florida and
Joseph Hogan of Brown University
This course will provide a survey of modern model-based
approaches to handling dropout in longitudinal studies and
illustrate the use of newly developed methods for sensitivity
analysis and incorporation of prior information. Emphasis
will be on Bayesian approaches, but the models and meth-
ods discussed can be implemented in non-Bayesian settings.
The course will be divided into three parts: (1) includes a
brief review of models for longitudinal data and the basics
of Bayesian inference; (2) focuses on formal classification of
dropout and missing data mechanisms, describes classes of
models that can be used to adjust for biases caused by drop-
out, and examines the logistics of model fitting; (3) deals with
specification and fitting of models to handle nonignorable
(informative) dropout, with emphasis on the role of sensitiv-
ity analysis and informative prior distributions for encoding
key assumptions.
Integrated into the course will be three case studies that
illustrate many of the concepts introduced during the course.
A view of the Walter E. Washington Convention Center from 7th Street NW
We will build on each case study to illustrate progressively
and Mount Vernon Place/New York Avenue
more complex analyses (e.g., progressing from analysis under
Photo courtesy of Destination DC
MAR to analysis under MNAR to use of informative priors
and sensitivity analyses). The case studies will include real-
time demonstrations of model fitting using WinBUGS soft-
log-likelihood, so it naturally provides for tests about model
ware and the R-to-WinBUGS interface.
parameters and goodness of fit; and the QIF provides for diag-
nostic tests about the mechanisms underlying any missing data.
Longitudinal Data Analysis: Semiparametric and
Nonparametric Approaches
Methodology for Competing Outcomes: The
Instructors: Annie Qu of the University of Illinois at Urbana-
Analysis of Multiple Mutually Exclusive Endpoints
Champaign and Peter Song of the University of Michigan
in a Clinical Trial
This short course will introduce semiparametric and non-
Instructor: David C. Naftel of the University of Alabama at
parametric approaches that do not require specification of the
Birmingham
likelihood function. In particular, the application of the qua- Many clinical trials today have a primary endpoint that is
dratic inference function (QIF) will be illustrated through the actually a composite of several separate endpoints that can be
development of statistical theory and methods in addition to constructed so they are mutually exclusive. Medical devices
real-world data examples and the use of the newly developed that provide mechanically assisted circulatory support often
SAS MACRO QIF. The QIF can easily take into account cor- are evaluated on the basis of multiple endpoints, including
relation within subjects and deal directly with both continuous death, removal of the device due to cardiac transplantation,
and discrete longitudinal data under the framework of general- and removal of the device due to recovery of the heart. A fourth
ized linear models. This course will demonstrate that the QIF ‘event’ is when the patient is alive with the device in place at
yields a more efficient estimator than the currently popular a specified point in time. Estimating the distribution of time
GEE approach; the QIF is analogous to twice the negative until the occurrence of the composite endpoint is important,
MAY 2009 AMSTAT NEWS 49
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84