BIOS 6321: Clinical trials and experimental design (Spring 2018)

For students only

• Restricted access

Instructor

• Tatsuki Koyama, PhD, Associate professor of biostatistics
• tatsuki.koyama@vanderbilt.edu; PRB 582; 615.936.1232
• Office hours: Thursday 3-4 in 11128A WEA

Teaching Assistant

• Jonathan Chipman, PhD Candidate
• jonathan.chipman@vanderbilt.edu
• Office hours: Friday 1:30-2:30 where?

Schedule

• Lectures: Tuesday and Thursday 1:30-3:00, 2525 Conference Room #11105
• Midterm exam: 2/27/18 (due)
• Final exam: 4/24/18 (due)

Course description

This course covers the statistical aspects of study designs, monitoring and analysis. Emphasis is on studies of human subjects, i.e. clinical trials. Topics include: principles of measurement, selection of endpoints, bias, blinding, randomization and balance, blocking, study designs, sample size, interim monitoring of accumulating results, flexible and adaptive designs, sequential analysis and adjustment techniques, history of clinical trials, and ethics of animal and human subject experimentation.

Prerequisite

If you are not a graduate student in the Department of Biostatistics, please email me to confirm you have sufficient math background to take this course.

• At least one semester of graduate level statistics
  • Bios 6311 or equivalent
• Calculus (3+ semesters)
• Familiarity with R
  • Bios 6301

Recommended materials

The lectures will loosely follow "Statistical Design and Analysis of Clinical Trials: Principles and Methods" by Weichung Joe Shih and Joseph Aisner.
Other good books: "Introduction to Statistical Methods for Clinical Trials" by Thomas D. Cook and David L. DeMets; ''Fundamentals of clinical trials (4th ed)'' by Friedman LM, Furberg CD, DeMets DL

Assignments

There will be homework assignments throughout the semester. Collaboration with classmates is encouraged, but you need to submit homework individually.

Exams

The midterm and final exams will be take-home. You will have a weekend (Thursday - Tuesday) to work on them. You are to work by yourself; no discussion is allowed.

Grading

• Homework and project (and class participation): 30%
• Midterm Exam (Take home): 35%
• Final Exam (Take home): 35%

The final exam will focus on the post-midterm materials, however, key concepts covered before the midterm may (almost surely) also appear.

Topics

1. Introduction and topics in statistics
   • Experiments and observational studies (1 lecture)
   • Clinical trial phases (1 lecture)
   • Topics in hypothesis testing (2 lectures)
   • Bayesian approaches (1 lecture)
2. Some fundamental topics in experimental design
   • Randomization (2 lectures)
   • Sample size and power (2 lectures)
3. Early phase clinical trials
   • Phase I clinical trials (1 lecture)
     • Class discussion (1 lecture)
   • Phase II clinical trials (2 lectures)
4. Phase III clinical trials
   • Introduction and group sequential designs (2 lectures)
   • Adaptive and flexible designs (2 lectures)
     • Guest lecturer (1 lecture)
   • Factorial designs (2 lectures)
   • Crossover designs (2 lectures)
5. Other topics
   • Fraud and misconduct
   • Quality of life
   • Noninferiority designs