BIOS 6649

Clinical Trials: Statistical Design and Monitoring

Syllabus and Lecture Schedule

Spring 2015

Course Instructors:Dr. John Kittelson

Department of Biostatistics and Informatics

Colorado School of Public Health

University of Colorado Anschutz Medical Campus

Building 500, Room W3163

, (303) 724-4364

Web page (course materials):

Prerequisites: Bios6611 and concurrent enrollment Bios6612.

Lecture times/place:TuTh10:30-11:50am; Ed2-South Room 2304

Office hours: Tu12:30-1:30pm; Bldg 500 room W3163 or by arrangement

Course Description: BIOS 6649 – Scientific and statistical design of clinical trials including the scientific parameterization of outcome space, frequentist and Bayesian standards for scientific evidence and statistical inference, fixed-sample trial design, and group sequential designs for trial monitoring. The course will use the R software library RCTdesign for the design and analysis of sequential clinical trials. The course is intended for biostatistics graduate students.

Objectives: Upon completion of this course the student will:

  1. Understand the scientific and statistical setting for clinical trials including:
  2. The scientific method as applied in clinical trials.
  3. The phased approach to developing a new therapy, and its efficiency properties.
  4. The statistical parameterization of the common types of questions encountered in clinical trials (superiority, non-inferiority, inferiority), and how that parameterization relates to the statistical standards for decision-making.
  5. Know the relationship between the scientific principles of human intervention trials and their relationship to the statistical design and conduct of clinical trials.
  6. Parameterization of the outcome space including choice of: (1) the probability model, (2) the functional on the distribution to measure treatment effect, and (3) the statistical model (contrast) that describes differences between groups.
  7. The specification of statistical hypotheses based on the scientific design.
  8. How frequentist measures of uncertainty relate to decision-making in clinical trials.
  9. How Bayesian measures of uncertainty relate to decision-making in clinical trials.
  10. Using the above scientific and statistical setting, construct an appropriate statistical design for a fixed-sample clinical trial, including:
  • Principles that determine the amount of information that is required to inform clinical practice, and how these principles determine the sample size requirement for a clinical trial.
  • The frequentist evaluation of the sample size including the discriminating ability of the trial and statistical power.
  • The Bayesian evaluation of the sample size including posterior discriminating ability and characterization of the prior opinion that is ``convinced” by trial results.
  1. Develop and evaluate a group sequential design for monitoring a trial using the RCTdesign library in the R statistics package. Specific elements include:
  2. Understand the motivation for interim analyses in a trial, the scientific implications, and the implications for statistical inference and control of statistical operating characteristics.
  3. Understand the classes of group sequential stopping rules and the major design families.
  4. Be able to evaluate the frequentist and Bayesian properties of a group sequential design.

Competencies:

  1. Scientific setting: Be able to frame the major trial questions/hypotheses within the context of the larger scientific setting. This includes appropriately categorizing the trial phase, and selecting appropriate trial endpoints and appropriate study population. Be able to correctly specify the form of the scientific hypotheses (e.g., superiority, non-inferiority, inferiority; one- versus two-sided hypotheses).
  2. Scientific design: Be able to formalize the scientific setting using an appropriate choice of the probability model, functional of interest, statistical model, and the statistical hypotheses.
  3. Statistical setting: To understand the relationship between statistical inference (point estimates, interval estimates, measures of uncertainty; frequentist and Bayesian measures) and the scientific standards for decision-making.
  4. Fixed-sample trial design: To appropriately discern the information requirements for a trial and use those to determine the sample size for a trial. To evaluate the implications of a sample size from both a frequentist and Bayesian perspective.
  5. Group sequential design: To develop and evaluate an appropriate statistical design for interim analyses of the primary endpoint. To evaluate the properties of this design from both a frequentist and Bayesian perspective.
  6. Software: To use modern software for the design and analysis of sequential clinical trials. Software is use to structure interim decision criteria, to evaluate their statistical properties, to adapt decision criteria in the midst of a trial, and for bias-adjusted inference at trial completion.

Textbook:

Required: Original sources including (software, affiliated teaching resources)

Recommended texts include:

Friedman L.M., Furberg C.D., DeMets D.L. (1998) Fundamentals of Clinical Trials, Third edition. Springer, N.Y.

Jennison C, Turnbull BW (2000) Group Sequential Methods with Applications to Clinical Trials. CRC press, Florida.

Assessment:

Exam 1 (30%)

Exam 2 (30%)

Project (30%)

Assignments (approximately weekly) (10%)

Course Policies:

Make-up Exam Policy: Make-up exams must be approved prior to the exam time except in the case of unexpected emergencies and these are permitted at the discretion of the professor.

Attendance Policy: Attendance is recommended but not required.

Academic Conduct Policy: All students are expected to abide the honor codes of the Colorado School of Public Health. Unless otherwise instructed, all of your work in this course should represent completely independent work. Students are expected to familiarize themselves with the Student Honor Code that can be found at:

Any student found to have committed acts of misconduct (including, but not limited to cheating, plagiarism, misconduct of research, and breach of confidentiality) will be subject to the procedures outlined in the Honor Code.

Disability Policy: For students requesting accommodations, contact the Office of Disability Resources and Services. Their staff will assist in determining reasonable accommodations as well as coordinating the approved accommodations. Phone number: (303) 724-5640. Location: Building 500, Room W1103. The physical address is 13001 E. 17th Place.

Lecture Topics: The schedule of lecture topics is attached as a separate document.

BIOS 6649 Course Syllabus, Spring 2015, Dr. John Kittelson 1