The latest data on combat deaths in Afghanistan show that the gap in fatality rates between the UK and the US has widened still further.
 
Between May 2009 and June 2010, UK death rates were 2.6 times greater than those of the US, for whatever reason. In the earlier phase of the war, May 2006 to May 2009, the ratio was 1.7.
 
US deaths are greater in number, but that is because the US commitment is greater. When fatality rates are calculated per 1,000 personnel-years, the UK can be seen to have suffered disproportionately.  
 
Since 1 May 2006, Clive Fairweather and I have reported every 20 weeks on military fatalities in Afghanistan, switching to 10-weekly reporting since 3 May 2010. The primary reason for the changed reporting frequency is that there are now at least as many coalition military fatalities in 10 weeks as hitherto there were in 20 weeks. Not only are more coalition troops now deployed to Afghanistan but their fatality rate has gone up.
 
The UK had deployed 4,500 troops to Afghanistan by 1 May 2006 but this had risen to 10,000 by 3 May 2010. The US had 23,300 troops in Afghanistan by 1 May 2006, rising to 90,000 by 3 May 2010. By then, 20,000 of the US deployment was to Helmand province, the area where  UK forces mainly operate.
 
In the 160 weeks from 1 May 2006 to 17 May 2009, the UK commitment amounted to 20,500 personnel-years (pys), with the loss of 152 lives – that is, a fatality rate of 7.4 per 1,000 pys (95% CI: 6 to 9). In the same period, the US commitment was 90,000 personnel-years, with the loss of 400 lives - a much lower fatality rate of 4.4 per 1,000 pys (95% CI: 4 to 5).
 
So in this earlier period, the US committed more than four times as many personnel-years, but suffered only 2.6 times as many deaths.  The UK’s military fatality rate was 1.7 times (7.4/4.4) that of the US.
 
The gap has since widened. In the subsequent 60 weeks from 18 May 2009 to 11 July 2010, the UK commitment has been been 11,000 personnel-years (just over half that of the preceding 160 weeks). But the cost in blood has been as high, at 155 lives lost, which means that the fatality rate has increased dramatically, to 14 per 1,000 pys (95% CI: 12 to 16).
 
In the same 60-week period, the US committed 91,000 personnel-years (about as high as in the preceding 160 weeks) with 484 lives lost, a fatality rate of 5.3 per 1,000 pys (95% CI: 4.8 to 5.8). While the UK fatality rate has almost doubled, that of the US has increased by only a fifth.  
 
In these 60 weeks, the US committed 8.3 times as many personnel years, but suffered only 3.1 times as many deaths. UK death rates were 2.6 times greater than those of the US. There may be several reasons for the difference, the most obvious being that different areas of operation within Afghanistan vary in the intensity of combat and hence in fatality rates.
 
Absent from this analysis is any count of fatalities among Afghan  forces. In an ideal world we should count, on an equal basis, the price being paid by the Afghan troops whom the coalition forces partner. 
 
Last week, Radio 4’s Today programme commented on the high number of US military fatalities in Afghanistan during June this year. The number of fatalities may increase because deployment has increased, or fatality rate increased, or both.  It is important therefore  to remember also to report fatality rates, because they reflect the tempo of operations – in particular, the extent of major combat.
 
This evening, when BBC2 considers the blood and treasure that UK has expended in Afghanistan, serious non-fatal casualties, on which Ministry of Defence regularly reports, will also feature.
 
Professor Sheila M. Bird is from the MRC Biostatistics Unit in Cambridge