We read with interest the article by Jones et al (Reference Jones, Vermaas and McCartney2003), who reported an absence of flashbacks in the symptom reports of exservicemen from the Second World War awarded pensions for post-combat disorders. As acknowledged by the authors, a limitation of the study is the retrospective review of historical descriptions of post-combat disorder symptoms in their sample population.

We administered the Clinician Administered Posttraumatic Stress Disorder Scale (CAPS) (see Weathers, 2001) to 82 American combat veterans of the Second World War who had also been held as prisoners of war (POWs). These veterans were seen as part of a compensation and pension examination conducted by the US Veterans Administration to examine the overall health status and presence of service-connected disabilities in these highly stressed veterans. CAPS interview question B-3 specifically asks the frequency (range: 0=none to 4=daily) and intensity (range: 0=none to 4=extreme) of flashback phenomena. Six of 41 ex-POWs (14.6%) who met criteria for post-traumatic stress disorder (PTSD) reported flashbacks in the month prior to the CAPS interview at a combined frequency and intensity score of 4 or greater. None of 41 ex-POWs who did not meet criteria for PTSD reported flashbacks at this level of frequency and intensity. In contrast, 75 of 124 Vietnamera veterans (60%) who had been diagnosed with combat-related PTSD and were administered the CAPS while participating in a Veterans Administration PTSD treatment programme reported flashbacks at this level of severity.

Consistent with the report of Jones et al, we found a striking difference in the prevalence of flashback symptom severity across generational cohorts. However, with specific questioning about this symptom using the standard diagnostic instrument for PTSD, the Second World War cohort who suffered extreme stress and currently meet criteria for PTSD did report flashbacks. The six former POW subjects who reported a current clinically significant level of flashbacks have informed us that this phenomenon was present in the 1940s. Of course, we cannot know whether these subjects would have reported flashbacks in the 1940s without having been exposed to the interim cultural changes.