The May 6, 2006, issue of Science News has a short article in it under the above title. Some quotes from the article follow.
Researchers have found evidence of bias when scientists review data and the researcher's name and affiliation are available to the reviewers...
... [in] some 67,000 research abstracts submitted to the American Heart Association... Experts... annually review the abstracts and deem about 30 percent... acceptable for presentation... at the annual meeting.
...the AHA changed its review process so that the authors' names and affiliations were stripped from the abstracts before they were sent out for peer review... The change triggered major shifts in which categories of authors were most likely to have... abstracts accepted.
...during 2000 and 2001, abstracts from U.S. authors were 80 percent more likely to be accepted than those from non-U.S. authors... After [the removal of names and affiliations] the U.S. based papers were only 41 percent more likely to be accepted... Similarly, the share of abstracts from from faculty at highly regarded U.S. research universities dropped by about 20 percent... For authors in government agencies, the acceptance rate fell by 30 percent.
Although the study focused on abstract[s] from one organization..., there's no reason to assume [that] the same thing doesn't happen at other meetings or in other disciplines.