Statistic Course at CVCT 2018 "design, data monitoring and reporting of clinical trials" Investigators Meetings at CVCT.jpg

Design, Data Monitoring and Reporting of Clinical Trials

Timetable: Thursday Nov 29, 2018

9:30am to 11:00am - Design and Statistical Fundamentals (Stuart Pocock)

11am - Coffee Break

11:30am to 1pm - Data Monitoring (Janet Wittes)

1pm to 2pm - Lunch

2pm to 3:30pm - Reporting (Stuart Pocock)

 

Learning Objectives

To provide a comprehensive review of important statistical and scientific aspects of major clinical trials in cardiovascular diseases. The content will be aimed at cardiologists, regulators, academic scientists, and statisticians: indeed all health professionals wanting to expand their knowledge of how to design, conduct, monitor, analyse, report and interpret a major randomized trial. Topics will be illustrated by real examples of cardiovascular trials. The two lecturers will actively encourage audience participation in a lively discussion.

 

Course Description

The course will comprise three 90 minute sessions, one each on trial design, data monitoring, and reporting. Topics covered will include the following:

 

1) Designing the Trial and Statistical Fundamentals

Types of trial; choices of patients, treatments and outcomes; randomization, how and why; blinding, how and why; size of trials and power calculations; non-inferiority trials; factorial trials.

 

Hypothesis testing, including the value and the misuse of P-values; estimation of treatment effects; Kaplan Meier plots; expressing uncertainty using confidence intervals; need for both absolute and relative effect measures.

 

2) Monitoring the Ongoing Data

The role and function of Data Monitoring Committee: practicalities, ethics and statistics of monitoring accumulating data; stopping for efficacy, for futility or for safety; statistical stopping guidelines; adaptive designs; the reality of decision-making when data suggest it’s time to stop a trial.

 

3) Reporting the Results

Multiplicity of data, hierarchical testing; handling of composite endpoints, subgroup analyses and covariate-adjusted analyses; intention-to-treat and per-protocol analyses; balancing efficacy and safety; Bayesian methods; constructive critical appraisal.

 

All three sessions will be full of topical examples to illustrate each point. Throughout we will point to common sources of bias that arise in designing, monitoring, and reporting about trials. Our goal is to structure the whole experience to be of direct practical value. We will provide appropriate references to enhance your knowledge further.