A potential limitation of this review, and most reviews in
general, is the literature search. It is possible that the
search did not identify all trials published, which could
lead to a selection bias, but this was minimized by
checking the references of previously published reviews
and the articles retrieved in the search. Furthermore,
the organization of primary care differs worldwide, and
a clear and worldwide-accepted definition of primary
care is not available. A different definition or interpretation
of primary care could have led to the inclusion of
a different set of studies. The criterion to exclude
unpublished trials might cause publication bias, which
is known to cause bias to positive findings.48