Furthermore, rather than writing the protocol so
that dropout by a participant results in a lack of
measured study outcomes, which can limit the
inferences regarding the effects, the NAS report
on missing data recommends that:
‘‘
Sponsors should continue to collect
information on key outcomes on
participants who discontinue
their protocol-specified intervention
in the course of the study, except
in those cases for which a
compelling cost-benefit analysis
argues otherwise, and this
information should be recorded
and used in the analysis.3
In addition, to avoid data-driven method selection,
all of the relevant aspects of the analysis, including
those for missing data and interim analyses,
should be included in the protocol and SAP.5
To minimize the potential of introducing
operational biases into the trial by making changes
to the trial conduct based on the accumulating
data, the protocol should also specifically address
who will have access to the data and conclusions.
Ideally, the personnel who see unblinded
results should not have involvement in trial
conduct or analysis decisions. For example, the
IDMC could see the unblinded results, but the
recommendations to the trial sponsor would
not include these results.
12
The FDA and other regulatory authorities have
published several documents that highlight
potential concerns and issues that can occur
d \