10class.wk4 - PubH 7420 Clinical Trials Readings for Week 4...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
PubH 7420 Clinical Trials: Readings for Week 4 1. Friedman, Furberg, and DeMets. Fundamentals of Clinical Trials, Chapter 5 and Chapter 16, pages 297-304. Supplemental Reading/References 1. Grizzle JE. A note on stratifying versus complete random assignment in clinical trials. Cont. Clin. Trials , 3:365-368, 1982. 2. Meier P. Stratification in the design of a clinical trial. Cont. Clin. Trials , 1:355-361, 1981. 3. Scott NW, McPherson GC, Ramsay CR, Campbell MK. The method of minimization for allocation in clinical trials: a review. Cont. Clin. Trials , 23:662- 674, 2002. 4. Hallstrom A, Davis K. Imbalance in treatment assignments in stratified blocked randomization. Cont. Clin, Trials , 9:375-382,1988. 5. Pocock SJ: Clinical Trials. A practical approach. John Wiley and Sons, Ltd. Chapter 5 and Chapter 13, pages 216-220. 6 Clinical Trials: Design, Conduct and Analysis, Chapter 10. 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Stratification (def.) - A procedure whereby factors which are known to be associated with the response of interest (prognostic variables) are taken into account in the randomization scheme, i.e., in the design of the study. Stratification aims to help ensure that prognostic variables have the same distribution in all treatment groups. Stratification is used to refer to restrictions on the randomization other than time (blocking). In other words, blocking is a restriction placed on the randomization to ensure the desired allocation ratio while stratification is a restriction to ensure comparability of the treatment groups with respect to the stratifying variables. As noted previously, in multi-clinic trials stratification on clinic is usually carried out since the types of patients can vary widely from clinic to clinic as can use of concomitant treatments and compliance to study treatment, i.e., it is not surprising to see a marked clinic effect on the outcomes of interest. In the analysis, sites may have to be grouped, e.g., by region, by size, or by type (HMO, university), otherwise, the sparse strata could result in a loss of power. Stratum (def.) - a large group of experimental units more homogenous than a randomly assembled group of experimental units by virtue of classification on some variable or set of variables at baseline. Advantages : - May prevent bias (an unfair treatment comparison) arising as a result of a chance imbalance between treatment groups on an important baseline prognostic factor. - Will increase the precision (reduce the variance) of the treatment comparisons made. - Will facilitate within stratum (subgroup) analysis since the treatments will be balanced. - If important prognostic factors are balanced then the study will be subject to less criticism. Disadvantages : - Results in a randomization scheme which is more difficult to implement and therefore more prone to error. For example, in a multicenter trial if one is stratifying on three baseline variables and each variable has two possible outcomes, eight schedules would have to be prepared in advance for each clinic. It is important to differentiate stratified randomization (also referred to as
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 30

10class.wk4 - PubH 7420 Clinical Trials Readings for Week 4...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online