ABSTRACT
In clinical trials, it is not uncommon to modify the trial and/or statistical procedures of on-going trials due to scientific/statistical justifications, medical considerations, regulatory concerns, and/or business interest/decisions. When modifications of trial and/or statistical procedures are made, a protocol amendment is necessarily filed to individual institutional review boards (IRBs) for review/approval before implementation. As discussed in the previous chapter, major (or significant) adaptation of trial and/or statistical procedures of a on-going clinical trial could alter the target patient population of the trial and consequently lead to a totally different clinical trial that is unable to answer the scientific/medical questions the trial is intended to address. In this chapter, we will examine the impact of protocol amendments on the target patient population through the assessment of a shift parameter, a scale parameter, and a sensitivity index. The impact of protocol amendments on power for detecting a clinically significant difference and the corresponding statistical inference are also studied.
