I was not disputing the fact that ghostwriting is an unethical practice, or
that narrative reviews may be more easily biased (by ghostwriting or
otherwise) than systematic ones. Neither I was hinting that looking for
better ways to prevent ghostwriting or other sources of bias (or at least
to make them more identifiable), in narrative reviews as in other types of
articles, should be discouraged.
I was making two points, in response to a specific suggestion raised in the
original post:
1) Narrative reviews should never be used as a base for clinical decisions,
regardless if they contain sections that seem to suggest it. And this not
(only) because of possible ghostwriting, but because simply they don't
guarantee to contain the right information for that, that is original data
or systematic analysis of original data.
2) When there is a need for broader information to put a problem in
perspective (which is common), systematic reviews are mostly useless and
cannot be used in place of narrative reviews, commentaries, editorials etc.
regards,
Piersante Sestini
At 13.14 07/08/2009 -0400, roy poses wrote:
>But narrative reviews often have embedded within them discussions of the
>sorts of questions better addressed by systematic reviews, e.g., specific
>approaches to particular diseases or problems.
>
>However, it is true that narrative reviews are often broader in scope than
>systematic reviews. That does not make the original issue, the
>ghost-writing of such reviews sponsored by corporations with products to
>sell, any less important. Narrative reviews do influence how physicians
>approach particular clinical problems. If the reviews have been
>engineered by vested interests to focus on that which will improve the
>marketing of their products, the result may be that physicians have a
>distorted view of the clinical problem, and hence a distorted approach to it.
|