The study by Jones et al (Reference Jones, Vermaas and McCartney2003) adds an interesting perspective on the concept of PTSD. However, there are methodological matters that concern me.

First, why are no subjects included from the Falklands Conflict of 1982? Jones et al cite O’Brien & Hughes (Reference O'Brien and Hughes1991), whose work suggests that a much higher incidence of flashbacks might be found among that population.

Second, how many raters were used to confirm the existence of PTSD symptoms in the case records? What were the interrater reliabilities? How was any disagreement resolved?

Third, during my brief sojourn as Medical Member (Psychiatrist) of the War Pensions Appeal Tribunals, I studied in detail some 80 War Pension Agency case records, many for non-psychological cases. My overriding concern was the lack of symptom recording. Frequently, the relevant questions on War Pension Agency medical assessments concerning mental state received one-word answers, or were deleted entirely. How did these researchers deal with such cases?