Department of Biostatistics Seminar/Workshop Series

Analysis of double blind placebo controlled (DBPC) randomized trials (RCTs) with compliance information

James Robins, MD

Mitchell L. and Robin LaFoley Dong Professor of Epidemiology

Departments of Epidemiology and Biostatistics, Harvard School of Public Health

Wednesday, February 23, 1:30-2:30pm, MRBIII Conference Room 1220

Many DBPC RCTs of a failure time outcome now collect time-varying risk factor and compliance data (eg pill counting, self report, blood levels of active treatment and sometines placebo treatment).

In this talk I will describe how this additional compliance information can be used to in the analysis in an attempt to increase power and to estimate the true drug effiicacy, while protecting the alpha level.

I will specialize the discussion to the recently published PREP trial of pre-exposure prophylaxis to prevent HIV infection in high risk individuals.