FEW MEDICAL JOURNALS have the storied reputation of The Lancet, a British publication founded in 1823. In the course of its long history, The Lancet has published work of exceptional influence, such as Joseph Lister's principles of antiseptics in 1867 and Howard Florey's Nobel Prize-winning discoveries on penicillin in 1940. Today it is one of the most frequently cited medical journals in the world.
So naturally there was great interest when the Lancet published a study in October 2006, three weeks before the midterm US elections, reporting that 655,000 people had died in Iraq as a result of the US-led war.
Hundreds of news outlets, to say nothing of antiwar activists and lawmakers, publicized the astonishing figure, which was more than 10 times the death toll estimated by other sources. The Iraqi health ministry, for example, put the mortality level through June 2006 at 50,000. Iraq Body Count, a nonpartisan anti-war group that maintains a public database of the war's victims, tallied some 45,000 Iraqi dead. If The Lancet's number was accurate, more Iraqis had died in the two years since the US invasion than during the eight-year war with Iran.
President Bush, asked about the study, dismissed it out of hand: "I don't consider it a credible report." Tony Blair's spokesman also brushed it off as "not . . . anywhere near accurate."
But the media played it up. "One in 40 Iraqis killed since invasion," blared a front-page headline in the Guardian, a leading British paper. CNN's story began: "War has wiped out about 655,000 Iraqis, or more than 500 people a day, since the US-led invasion, a new study reports." Few journalists questioned the integrity of the study or its authors, Gilbert Burnham and Les Roberts of Johns Hopkins University's Bloomberg School of Public Health, and Iraqi scientist Riyadh Lafta. NPR's Richard Harris reported asking Burnham, "Right before the election you're making this announcement. Is this politically motivated? And he said, no, it's not politically motivated." Burnham told Newsweek the same thing: "There's no political motivation in this. I feel very confident in the numbers."
But the truth, it turns out, is that the report was drenched with politics, and its jaw-dropping conclusions should have inspired anything but confidence.
In an extensively researched cover story last week, National Journal took a close look under the hood of the Lancet/Johns Hopkins study. Reporters Neil Munro and Carl M. Cannon found that it was marred by grave flaws, such as unsupervised Iraqi survey teams, and survey samples that were too small to be statistically valid.
The study's authors refused to release most of their underlying data so other researchers could double-check it. The single disk they finally, grudgingly, supplied contained suspicious evidence of "data-heaping" - that is, fabricated numbers. Researchers failed to gather basic demographic data from those they interviewed, a key safeguard against fraud.
"They failed to do any of the [routine] things to prevent fabrication," Fritz Scheuren, vice president for statistics at the National Opinion Research Center, told the reporters.
Bad as the study's methodological defects were, its political taint was worse:
- Much of the funding for the study came from the Open Society Institute of leftist billionaire George Soros, a strident critic of the Iraq war who, as Munro and Cannon point out, "spent $30 million trying to defeat Bush in 2004."
- Coauthors Burnham and Roberts were avowed opponents of the Iraq war, and submitted their report to The Lancet on the condition that it be published before the election. Roberts, a self-described "advocate" committed to "ending the war," even sought the Democratic nomination for New York's 24th Congressional District. "It was a combination of Iraq and Katrina that just put me over the top," he told National Journal.
- Lancet editor Richard Horton "also makes no secret of his leftist politics," Munro and Cannon write. At a September 2006 rally, he publicly denounced "this axis of Anglo-American imperialism" for causing "millions of people . . . to die in poverty and disease." Under Horton, The Lancet has increasingly been accused of shoddiness and sensationalism. In 2005, 30 leading British scientists blasted Horton's "desperate headline-seeking" and charged him with running "badly conducted and poorly refereed scare stories."
The claim that the US-led invasion of Iraq had triggered a slaughter of almost Rwandan proportions was a gross and outlandish exaggeration; it should have been greeted with extreme skepticism.
But because it served the interests of those eager to discredit the war as a moral catastrophe, common-sense standards were ignored. "In our view, the Hopkins study stands until someone knocks it down," editorialized the Baltimore Sun.
Now someone has, devastatingly. But will the debunking be trumpeted as loudly and clearly as the original report? Don't hold your breath.
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
February 25, 2009 update: The Nation Journal report prompted an internal investigation by Johns Hopkins, which found improprieties in the way the Burnham/Roberts research was conducted. As a result, Johns Hopkins has now announced, "the School has suspended Dr. Burnham's privileges to serve as a principal investigator on projects involving human subjects research."
(Jeff Jacoby is a columnist for The Boston Globe).
-- ## --
Follow Jeff Jacoby on Twitter.
Discuss his columns on Facebook.
Want to read more? Sign up for "Arguable," Jeff Jacoby's free weekly email newsletter.