Principles and Practice of Pharmaceutical Medicine

larger effect than treatment B in some cen-
ters and a smaller effect in other centers. If
the researcher cannot find the cause of this
interaction and correct for it, the study will
be inconclusive. This type of interaction
would occur if, for example, the data center
mislabeled the treatments for some centers.
This would be easy to rectify. However,
often there is no reasonable and acceptable

explanation for this, and the entire study

has to be invalidated.

The ICH guidelines address this issue as follows:

If heterogeneity of treatment effects is found, this

should be interpreted with care, and vigorous

attempts should be made to find an explanation in

terms of other features of trial management or subject

0

2

4

6

8

10

12

14

16

A B C D E F

Center

Response

TRT A

TRT B

Figure 25.2 Treatment by center interaction: quantitative interaction

0

2

4

6

8

10

12

14

A B C D E F

Center

Response

TRT A

TRT B

Figure 25.3 Center by treatment interaction: qualitative interaction

25.11 ISSUES IN STATISTICAL TRIAL DESIGN 335