Under fire from critics, researchers from the Johns Hopkins University and elsewhere have released selected bits of data from a much-discussed study of how many Iraqis have died since the 2003 American invasion.

In an article published in October 2006 in The Lancet, the British medical journal, a team led by Gilbert Burnham of Hopkins estimated that there were 650,000 more Iraqi deaths after the invasion than would have been expected based on conditions before the war. The study was conducted by sending interviewers to neighborhoods to canvas 1,849 households (The Chronicle, October 20, 2006), and it followed up on a 2004 study.

At the time of the publication, the Johns Hopkins group did not disclose the identity of the interviewers or the neighborhoods surveyed because their Iraqi collaborators had asked them to withhold the data to protect the interviewers, Dr. Burnham said on his Web site.

Now that six months have passed, the team announced that it would release some data, but it is still not identifying the neighborhoods involved in the study. And it will send data only “to recognized academic institutions or scientific groups with biostatistical and epidemiological analytic capacity.” Another requirement is that the information go only to groups “without publicly stated views that would cause doubt about their objectivity in analyzing the data.”

Science magazine reports that the Johns Hopkins team has refused to send the data to at least one researcher and that several scientists are upset about the restrictions, which they say make it impossible to verify the study’s conclusions. —Richard Monastersky