That post-traumatic stress disorder (PTSD) is not associated with disability and that its clinical importance ‘may be questionable’ are huge claims and somewhat counterintuitive. For Neal et al (Reference Neal, Green and Turner2004) to come to such counterintuitive conclusions would require a very strong piece of research. Unfortunately, since their paper does not seem to offer a sound methodology or results, the force and the validity of their conclusions is debatable.

Six years (on average) after a traumatic event the armed services personnel they study report high levels of disability. From the paper it appeared that the authors thought that the association was not mediated via diagnoses of PTSD, depression or alcohol dependence on statistical grounds (although these disorders were all frequently present in the 70 armed services personnel referred to the PTSD unit). The presence of high scores on the Beck Depression Inventory (BDI) accounted for a ‘high proportion of the variability’. In other words, diagnoses did not account for variability but ‘depression consequent upon trauma’ did. It is difficult to understand how the findings of relatively high disability 6 years (on average) after a trauma associated with depressive scores on the BDI cannot be linked to the diagnoses found at interview, as the conclusion of the abstract section clearly implies.

There are several problems with the study design. There is no control group. All the interviews were conducted by a single individual, a nurse, who was presumably not masked to the origin of the patients. The Structured Clinical Interview for DSM–IV (SCID) is not primarily intended as an instrument to detect PTSD. An easily administered alternative to the SCID might have been the Short Post-traumatic Stress Disorder Rating Interview (SPRINT) (Reference Connor and DavidsonConnor & Davidson, 2001) which has solid psychometric properties. There is perhaps an over-reliance otherwise on self-report questionnaires.

The study also fails to refer to relevant literature. A study published in 2001 by Tucker et al recruited 307 patients with PTSD and was both double-blind and placebo-controlled. This found a significant reduction in disability, as measured by the same Sheehan Disability Scale, after 12 weeks of treatment with paroxetine (Reference Tucker, Zaninelli and YehudaTucker et al, 2001).

At this point I must state my own interest in that I have written court reports on PTSD, but on joint solicitors’ instructions or single solicitor's instructions for claimants or defendants. However, with regard to the authors – although one would like to believe in their independence – surely it is not credible for there to be no declaration of interest stated when all three authors are employed either directly or indirectly by the Ministry of Defence?

Declaration of interest

B.H.G. has written numerous personal. injury and clinical negligence medico-legal reports and is Editor of. Psychiatry On-Line.