Review data may be outdated
Zsolt Nyulaszi/Dreamstime.com
Care is needed when searching databases for clinical reviews |
Systematic reviews, such as those held in the Cochrane database, may go out of date sooner than initially thought, a Canadian study suggests.
Research was carried out to determine the “lifespan” of a
high-quality, systematic review, and whether any factors could predict
its duration of validity.
The researchers selected 100 reviews from the American College of Physicians
Journal Club database. For each review, a search was conducted for new
clinical trials that would have fitted the inclusion criteria of the
original review, and which reached conclusions that would result in a
clinician or policy maker altering their opinion of the original review.
The researchers found that on average, an update was necessary 5.5 years
after publication of a review, because new data had emerged. However,
for 15 per cent of reviews, significant new data was available within
one year, and for 23 per cent this occurred within two years. For seven
per cent, new data was available before the review had been published.
Reviews on cardiovascular subjects tended to have a shorter lifespan,
but there were no clear factors predicting a shorter duration of validity.
Nick Royle, chief executive of the Cochrane Collaboration, told Hospital
Pharmacist that this is an area they are looking into. He explained
that the collaboration’s initial practice of assessing reviews
every two years was not sustainable due to the number of reviews on its
database.
It has now changed its working practice to highlight when a
review has
been checked,whether or not a major update has occurred. A working
group has been set up whose remit will include attempting to classify
reviews
by the likelihood of them being changed by future research (Annals
of Internal Medicine 2007;147:224-33). |
Hospital
Pharmacist