From Pacific Standard [wiki] by Michael Spagat (2013): The Iraq Sanctions Myth]. See also: The Nation (2001) &c &c]
Sanctions allegedly killed hundreds of thousands of children in Iraq and provided a rationale for invasion, a line still heard today. But those deaths almost certainly never happened.
The claim that sanctions killed hundreds of thousands of Iraqi children originated in a 1995 letter to The Lancet which, in turn, was based on a Baghdad survey done by Sarah Zaidi and colleagues. After other researchers identified anomalies in the survey data, Zaidi, to her great credit, re-investigated the work from the ground up. Having sub-contracted the original interviews to the Iraqi government, she traveled to Baghdad and re-interviewed many of the original households. When Zaidi failed to confirm quite a few of the reported deaths in these follow-up interviews, she retracted her results.
From note #6, page 2, PDF: Estimating the Human Costs of War: The Sample Survey Approach (Michael Spagat, in Garfinkel, Michelle R. and Skaperdas, Stergios, (eds), The Oxford Handbook of the Economics of Peace and Conflict, 2012):
Indeed, survey approaches to measuring deaths due to economic sanctions in Iraq have encountered similar problems. Zaidi (1997) withdrew her survey-based estimate of 567,000 child deaths due to sanctions (Zaidi and Smith, 1995) after revisiting households from the original survey and failing to replicate many of the deaths. According to a subsequent UNICEF survey, child mortality nearly doubled in the early 1990’s in Iraq (Ali and Shah, 2000 and Ali, Blacker and Jones, 2003), resulting in an estimated 400,000 to 500,000 “excess deaths” of children. However, Dyson (2009) shows that these estimates are inconsistent with a range of credible evidence and argues that it is likely that the UNICEF “survey data were deliberately manipulated by the then government of Iraq.”
Basing decisions, policy or statements on (as Albright unfortunately discovered) incorrect or invalid data is never a good idea. If you’re considering voting for Hillary in the primaries, I just wanted to provide more information on this issue. This incorrect data, as noted by Pacific Standard, was still being used by Tony Blair in 2010 to justify the invasion of Iraq, and the Washington Post for George Bush.
Michael Spagat is a professor of economics at Royal Holloway, University of London. He provides a detailed response to the Zaidi & UNICEF estimates in this 2014 interview.