|
ISPOR Short Courses- Sunday, May 19th, 2002 |
 |
| MORNING SESSION
- (8:00-12:00PM) |
Meta-Analysis and Systematic Literature Review - Introduction
Joseph C. Cappelleri PhD, MPH, Associate
Director, Pfizer Inc., CT, USA and
Joseph Lau MD, New England Medical Center and
Tufts University Medical School, MA, USA
Course Description: Meta-analysis may be defined as the statistical
analysis of data from multiple studies for the purpose of synthesizing and
summarizing results, as well as for quantitatively evaluating sources of
heterogeneity and bias. A systematic literature review often includes
meta-analysis and involves an explicit, detailed description of how a review
was conducted. This short course highlights and expounds upon four key areas:
1) impetus for meta-analysis and systematic reviews, 2) basic steps to perform
a quantitative systematic review, 3) statistical methods of combining data,
and 4) appraisal and use of meta-analytic reports. The material is motivated
via applications in pharmacoeconomics, outcomes research, and clinical studies
from the published literature and hypothetical examples. Interactive exercises
are part of the course. After attending the workshop, participants should be
able to 1) express the need for meta-analysis and systematic literature
review, 2) describe the basic steps to perform a quantitative systematic
review, 3) explain statistical methods of combining data, and 4) appraise
meta-analytic reports and decide whether to use their results. This course is
designed for those with little experience with meta-analysis.
Introduction to Patient Reported Outcomes (PRO)
Assessment:
Designing a PRO Strategy
Linda Abetz MA, Project Director, Mapi Values, UK
Course Description: Conceptual, methodological, and practical methods
for measuring patient reported outcomes. An emphasis will be placed on
health-related quality of life; with additional discussion provided on symptom
and treatment satisfaction assessment strategies. A strategy to aid in
selecting appropriate instruments will be presented. Reliability, validity,
responsiveness, methods of administration and scoring, and issues of analysis
and interpretation will be discussed using practical examples and exercises.
This course is designed for those with either little or intermediate
experience in health-related quality-of-life assessment.
Statistics for Non-Statisticians
Thomas Einarson PhD, Associate Professor, Faculty
of Pharmacy, University of Toronto, Toronto, Canada
Course Description: This course is designed to present an overview of
the foundations upon which major statistical tests are based and tests which
may be used to address pharmacoeconomic problems. The major emphasis will be
on application and interpretation of statistical results, not statistical
theory. Surveyed will be the most commonly utilized statistical tests (t-test,
ANOVA, correlation, regression, chi square) and the types of variables
associated with each test. The course will assist ISPOR attendees who are
interested in increasing their understanding of the general principles
associated with elementary statistics and how to apply these tests to problems
presented in their work environment. This course is suggested as a
prerequisite to "Statistical Considerations in Pharmacoecomonic Evaluations –
Advanced" for those who lack a background in basic
inferential statistics or need a brief review.
Decision Analysis – Introduction
Mark Roberts MD, MPP, FACP, Associate Professor
of Medicine, Chief, Section of Decision Sciences and Clinical Systems
Modeling, University of Pittsburgh School of Medicine, PA, USA
Course Description: Decision analysis is a tool that provides an
explicit structure for solving complicated healthcare problems, and is
commonly used as the tool for conducting cost-effectiveness analyses as well.
This course will describe the basic structure and analysis of decision trees,
the calculation of expected utilities, sensitivity analysis, the assessment of
patient values, and inclusion of quality-of-life parameters. Simple examples
will be used to illustrate these concepts, and more complex examples from the
literature will be presented. This course is designed for those with little or
no experience in decision analysis.
Introduction to Bayesian Approaches to Health
Economics and Outcomes Research
Bryan Luce PhD MBA, Senior Research Leader & CEO,
MEDTAP International, MD, USA, Tina Shih PhD,
Research Scientist, MEDTAP International, MD, USA, Chris
M. Barker PhD, Director of Statistical Research, MEDTAP International,
MD, USA, Frank E. Harrell, Jr. PhD, Professor of
Biostatistics and Statistics, University of Virginia School of Medicine, VA,
USA
Course Description: This course is designed to provide an overview of
the Bayesian approach and its applications to health economics and outcomes
research. The course will cover basic elements of the Bayesian statistics
(prior, likelihood, and posterior distribution, conjugate families, EVPI,
etc.), discuss differences between Bayesian and classical (Frequentist)
approach, and demonstrate how to apply the Bayesian approaches to clinical
trials and cost-effectiveness analyses. Studies conducted in Excel or WinBUGS,
statistical software designed for Bayesian computations, will be presented as
teaching examples. Participants will learn how to interpret and apply the
Bayesian approaches to their work. This course is designed for the beginner or
intermediate POR professional.
Introduction to Pharmacoepidemiology
David Lilienfeld MD, MS, MPH, Associate Director,
Pharmacoepidemiology, Bristol Myers-Squibb, NJ, USA
Course Description: Pharmacoepidemiology is the application of
epidemiological knowledge and methods to study the effects (both positive and
negative) of drugs in human populations. Its purpose is to describe and
predict drug treatment in a defined time, space, and population. This course
will provide an overview of the contribution of epidemiology to the study of
drug uses and effects. Pharmacoepidemiologic study design strategies
(observational, analytic, and interventional studies) including their
strengths and weaknesses will be presented. This course is for those with no
or little experience with pharmacoepidemiology.
|
| AFTERNOON SESSION (1:00PM-5:00PM) |
Decision Analysis -
Advanced Applications
Marc F. Botteman MSc, MA, Executive Director,
International Health Economics, HERQuLES, Abt Associates Clinical Trials, MD,
USA
Course Description: This course will teach participants how to use and
implement intermediate-to-advanced decision analysis techniques (such as
Markov models and first- and second-order Monte Carlo simulations) for
cost-effectiveness analyses. After a brief review of these methods and their
appropriate use, the course will focus on the steps involved in developing
Markov models and conducting Monte Carlo simulations. To illustrate how these
methods are applied, examples will be demonstrated in Microsoft Excel and
Monte Carlo simulation software, as appropriate. Issues related to the
selection of model input parameters and their distributions for use in
probabilistic sensitivity analyses will be reviewed. Participants will learn
how to analyze, interpret, and present results (e.g. using acceptability
curves or analyses of the cost-effectiveness plane). If time permits, the
course will also briefly cover more advanced topics (e.g. discrete event
microsimulations). Publications from healthcare journals presenting additional
examples and theoretical considerations will be provided as a course
supplement. This intermediate course requires basic understanding of decision
analysis. Also, because computers are not available for
participants' use, please note that the course will be characterized more
as seminar-style, rather than as a hands-on modeling workshop.
Elements of Pharmaceutical Pricing
Jack M. Mycka, President & GM, Pricing on
Purpose, a division of Roger Green & Associates, Inc., PA, USA
Course Description: The elements of pharmaceutical pricing decisions
and the role of pharmacoeconomics in shaping those decisions will be
discussed. This course is designed for those with some experience in either
pharmacoeconomics or pharmaceutical pricing.
Prospective Economic Trials
Gerry Oster PhD, Vice President, Policy Analysis
Inc. (PAI), MA, USA
Course Description: Randomized
controlled trials are the accepted standard for demonstrating the efficacy and
safety of new pharmaceuticals. They also can be a useful venue for collecting
economic data. This course will provide participants with an overview of
methodologic and practical issues surrounding economic data collection that is
"piggybacked" onto otherwise-traditional randomized controlled trials,
including protocol development, data collection, and data analysis. The course
also will consider circumstances under which such studies do not provide an
appropriate setting for collecting "real-world" economic data, and will
discuss key aspects of clinical trials that are conducted specifically to
inform economic decision making ("pragmatic clinical trials") as well as
issues in the design and conduct of these types of investigations. Course
participants will be asked to solve specific problems in the design of
trial-based economic investigations. This course is designed for persons with
a basic knowledge of and familiarity with clinical trial methodology as well
as techniques of economic investigation.
Quality-of-Life – Advanced
Karen F. Gold PhD, Director of Biostatistics and
Outcomes Research, Abt Associates Clinical Trials, MD, USA
Course Description: This section will discuss the role of advanced
psychometric analysis in the construction and evaluation of quality-of-life
instruments. The course will use the latent variable model as an organizing
theme around the following approaches to scale evaluation: 1) internal
consistency, 2) exploratory factor analysis, 3) confirmatory factor analysis,
4) structural equation models with latent variables and 5) item response
modeling. Two detailed worked examples will be presented examining handling
loss to follow-up due to death in a QoL study (structural equations model) and
adaptive Quality of Life assessment tools (item response theory). This course
is for those with experience with quality-of-life instruments and psychometric
models.
Advanced Retrospective Database Analyses
Thomas E. Delea MBA, Senior Consultant, Policy
Analysis Inc. (PAI), MA, USA
Course Description: Large-scale retrospective databases provide a
unique opportunity to examine the effects of drug use on clinical and economic
outcomes in "real world" settings. The learning objective for this course will
be to gain in-depth understanding and hands-on experience with methods for
estimating the relationship between drug use and clinical and economic
outcomes using retrospective databases. An overview of the challenges
associated with confounding (i.e., selection bias) and censoring in
retrospective database analyses will be described, along with general
approaches for addressing these challenges including sample selection and
analytical approaches. Selected analytical methods will be described including
general linear regression, logistic regression, Poisson regression, and Cox
proportional hazards regression. Topics to be covered will include methods for
assessing the degree of confounding, selection of covariates, assessing the
appropriateness of model assumptions, incorporating time-dependent covariates,
analyzing multiple events per patient, and estimation of adjusted event rates
and survival functions. The uses of propensity scores to control for
confounding through
matching, stratification, or as covariates in multivariate analyses will be
examined as well. Specific examples of these various techniques will be
provided. Course participants will be asked to participate in a case study in
which the methods described in the course will be applied. This course is for
persons with a basic understanding of statistics and principles of clinical
epidemiology.
Statistical Considerations in Pharmacoeconomic
Evaluations – Advanced
Kevin D. Frick PhD, Johns Hopkins Bloomberg
School of Public Health, Department of Health Policy and Management, MD, USA.
Course Description: This course will
provide an overview of the terminology and key methodological features of
pharmacoeconomic evaluations. The primary focus will be on the statistical
considerations of planning and analyzing studies and interpreting the results,
including comparisons of types of analyses used and discussing important study
design features for clinical economic trials. Methods of statistical analysis
for cost data and for estimating cost-effectiveness ratios will be thoroughly
reviewed, making extensive use of real-life examples and published studies.
This course is designed for those with experience in conducting
pharmacoeconomic studies. |
| |
|
