You’ve
spent a lot of time and effort conducting a clinical study and now you find
that it’s flawed either when it comes time to try and get it published,
or because a systematic reviewer finds the problems.
It doesn’t have to be this way.
Most clinicians aren’t trained in clinical design of studies, and it’s one thing to understand clinical study design in general and another to organize the protocol for a study. We can help. We can either offer a brief critique or we can help you design the study from scratch and generate a protocol—your choice.
Some examples of design flaws:
· Data in a cohort study that is massively skewed with regard to
patient parameters because data was not
collected in proportion
to each patient enrolled.
· An RCT that has so many exclusion criteria that it doesn’t
have much relevance to real clinical practice.
· A case-control study that hasn’t properly matched control
subjects to the “experimental” subjects.
· Loss to follow-up in a longitudinal study that is excessive
· RCTS that don’t adhere to the CONSORT protocol
· A non-inferiority RCT design in which the study is powered to detect
only a 25% difference in the primary
outcome parameters.
· A comparative design in which there are major differences between
groups with regard to study parameters
at baseline.
A dollar well spent at the design stage is worth hundreds spent at the execution stage. Be clear in what your objectives are, what you would like to demonstrate. Rigorously examine your protocols. Ask tough questions … what if? If you don’t ask the tough questions, it’s almost certain a reviewer will after the study is complete and then it’s too usually too late to fix problems.
Most journals are now getting a lot tougher about
all the other items that go with designing a trial, too. If your study is
testing an intervention or treatment, you should get it registered before
you start (for example, with clinicaltrials.gov). Are you sure you don’t
need an IRB approval for a study? Think again! Many journals are now insisting
on this, even for trivial or “routine care” studies, whether
we think that’s a good idea or not. It’s not hard to set up
an institutional IRB, so this should not pose a barrier to your studies,
and we can advise you on this aspect if you need assistance.
