We accept that our findings require confirmation from further studies. However, human intuition has often been shown to be incorrect in the face of scientific research and perhaps ought not to be taken too seriously. A good example of this is the recent history of psychological debriefing to prevent PTSD (Reference Rose, Bisson and WesselyRose et al, 2003).

The finding that the categorical measures of depression (according to DSM–IV) did not concur with the continuous measure of depression (according to the BDI), in terms of predicting disability, may be evidence for the unreliability of the way we categorise psychiatric disorder, in terms of individual functioning. This is a possible area for further investigation.

The study was a cross-sectional survey examining the within-subject variability and relationships between variables. It has not been explained how a control group would add anything to the findings. The origin of the subjects was not a variable in the study design and it has not been explained why the assessor should have been masked to this information. The SCID is the most widely used and the most thoroughly researched clinical interview format for PTSD (Reference Wilson and KeaneWilson & Keane, 1997). The SPRINT is one of numerous other measures of PTSD. A search on the National Center for PTSD database showed 127 hits for the SCID and 3 hits for the SPRINT. The use of self-report questionnaires as continuous variables was integral to study design and was not an ‘over-reliance’.

The study by Tucker et al (Reference Tucker, Zaninelli and Yehuda2001) does not tell us anything about the relative contribution of PTSD, depression or alcohol dependence to disability, which was central to our hypothesis. Paroxetine is effective in the treatment of depression as well as PTSD.

Employment by the Ministry of Defence does not introduce an obvious partisan interest in this study. On the one hand, the Ministry might benefit from showing that PTSD does not cause disability, but on the other hand, if PTSD has little relevance, then the need to employ military psychiatrists may be questionable. Either way, the employing organisation can hardly be said to have been concealed by the authors from Dr Green.

Declaration of interest

At the time of data collection, L.A.N. was employed by the UK Ministry of Defence. At the time of submission of the publication, he had no links with the Ministry.