**Statistics 565**

**Methods in Biostatistics**

**Basic Information:**

**Meeting Time: **Tuesday and Thursday,
8-9:20 am in Molecular Biology 1428

**Final Exam: **Monday,
December 12, 7:30-9:30 am

**Instructor: **Kenneth
J. Koehler

Office: 102F Snedecor Hall

Telephone : 515-294-4181

Fax : 515-294-4040

E-mail: kkoehler@iastate.edu
**
**Office hours: To be announced

**Textbook**

Collett, David,Modelling Survival Data in Medical Studies Data, 2nd edition,

Chapman&Hall/CRC, 2003.

ISBN 1-58488-325-1Diggle, Peter J., Heagerty, Patrick, Liang, Kung-Yee, Zeger, Scott L.

Analysis

of Longitudinal Data, 2nd edition, Oxford University Press, 2002.

ISBN 0-19-852484-6Course notes will be posted as the course progresses.

This course will review basic features of observational studies, including case-referent and cohort studies, and randomized clinical trials. It will provide an introduction to the analysis of time-to-event data and data obtained by taking repeated measurements on subject across time. After taking this course students will be familiar with survival and hazard functions. They will know how to estimate and compare survivor functions using non-parametric Kaplan-Meier estimation. Students will also learn how to model the effects of covariates using proportional hazards models. Some diagnostic procedures will be reviewed. Accelerated failure models and parametric estimation using exponential and Weibull probability models will be introduced, but this will not be a major feature of the courses. The second part of the course will examine statistical methods for analyzing data from longitudinal studies. Students will learn how to work with linear and generalized linear models containing both fixed and random effects. Some strategies for dealing with missing data will also be considered.Along with introducing students to statistical methods for analyzing longitudinal data, some insight into the underlying models and theory will be presented to enable students to identify situations in which it is appropriate to apply particular procedures. Another objective of this course is to help students become familiar with software for applying these methods. Examples will be presented in the lectures using both the SAS and R packages, and assignments will require the application of statistical software. Students may use any software they choose in this course, but assistance will only be given for the SAS and R packages. R code presented in this course should run in S-PLUS with little modification. Links to files containing SAS and R code will be made available on this web page as they are presented in the lectures.

Windows versions of SAS and R can be accessed through the machines in the computing labs in Snedecor 321 and 322, and at other locations on campus. R can be downloaded from the web for free (www.r-project.org). Students can obtain a free copy of S-PLUS for windows that they can load on their own computers from Jeanette in 102 Snedecor. Under our site license, this copy of S-PLUS must be loaded on a computer that you own. It cannot be loaded on any computer belonging to any company or university, including Iowa State University.

Copies of S-PLUS and SAS manuals are available in Snedecor 115, the Computation Center Library, and the Parks Library. On-line help is also available with in each package.

You should have a calculator that you can bring to exams.

**Course grades**

About eight written assignments, one mid-term exam, possibly one project, and a final exam.

**Material to be Covered**

TopicReading Assignments1. Observational studies and

randomized clinical trials.Class notes 2. Survival analysis: Survival

functions and hazard functionsCollett, Chapter 1 3. Non-parametric methods:

Kaplan-Meier estimation,

comparing survival functionsCollett, Chapters 2 and 12 4. Parametric methods: Weibull

and exponential modelsCollett, parts of Chapter 5 5. Cox regression model,

proportional hazards, diagnosticsCollett, Chapters 3 and 4 6. Time dependent covariates

Collett, Chapter 8

7. Longitudinal studiesDiggle, et al, Chapters 1-3

8. Mixed linear modelsDiggle, et al, Chapters 4 and 5

9. Methods for count dataDiggle, et al, parts of Chapters 7-9

10. Meta analysisClass Notes