IFHS – “Violent deaths”
January 27, 2008
The main discrepancy between findings in the IFHS paper and those of Burnham et al. is not in the total excess deaths, but in the specific category “violent deaths”. It is therefore of interest to examine whether the classification methods used in those papers to assign deaths to the “violent” category are identical, and whether any differences in the classifications could account for some of the different findings. I notice two points on which the papers’ methodologies of classification differ: One is that Burnham et al. examined death certificates, while IFHS did not. A second difference is that they use different categories for injuries. Burnham et al. use two “accident” categories. One of those is included in the “non-violent” section, the other in the “violent” section. IFHS has no “violent accident” category, and has two categories, “road accidents” and “unintentional injuries”, counting injuries within the “non-violent” classification.
The difference on the matter of relying on death certificates could explain the large difference in the amount of deaths of “unknown” causes counted. While Burnham et al. have no deaths categorized as “unknown” for the pre-invasion period and 1% “unknown” for the post-invasion period, IFHS has about 6% of the deaths counted as “unknown” for both periods. IFHS classify “unknown” as “non-violent”. Therefore, if deaths certificate would indicate that some of those should be classified as “violent”, this could increase significantly the number of “violent” deaths – with a potential increase of up to 50%. It is relevant to note that the 6 deaths of “unknown” causes counted by Burnham et al. were all judged by the surveyors as “violent”.
The second methodological difference – i.e., the use of different injury categories – could also be a source of significant differences in the count as well. In light of the fact that both “road accidents” and “unintentional injuries” show substantial increases in mortality rates for the post-invasion period compared to pre-invasion period (4x for the former and 2x for the latter), it seems a reasonable guess that both categories, but in particular the latter category, could have a substantial overlap with some of the “violent” categories of Burnham et al. (which recorded no significant increase in the “non-violent” type of “accidents”.) “Road accidents” may be associated with violence at check points while the term “unintentional injuries” is vague enough to include many war related injuries (e.g., accidental shooting, or death in ordnance explosion). In fact, the term “Injuries/armed conflict” used by the IFHS to describe war related deaths that fall under the “violent” classification, may be interpreted (depending on how the Arabic term being used is understood and on the context) to mean only deaths of combatants, in which case any warfare related death by non-combatants would be counted as “unintentional injury”.
Moving the deaths counted under “unintentional injuries” into the violent deaths classification would increase that count by 35-40%. Adding in some of the “road accidents” could increase it noticeably further.
In conclusion, the deaths, or some of the deaths in the “unknown cause”, “unintentional injury” and “road accident” categories in the IFHS could represent mortality that was counted by Burnham et al. as “violent”. Including much of those deaths in the IFHS estimate for violent deaths could increase that estimate by 80-90% putting the point estimate at about 270,000-280,000 and the upper end of the confidence interval at more than 400,000. This would put it less at odds with the estimate given by Burnham et al. (600,000, 95% CI 425,000-800,000).