We thank Ben-Ezra & Aluf (Reference Ben-Ezra and Aluf2005) for their letter, in which they broadly support our findings (Reference Mol, Arntz and MetsemakersMol et al, 2005) that life events may cause as many symptoms of post-traumatic stress disorder (PTSD) as traumatic events classified according to the A1 criterion of the DSM–IV. However, they also have some criticisms. Ben-Ezra & Aluf argue that ‘serious illness (self)’ – classified as a life event in our study – can be considered a traumatic event. We decided against this classification as many respondents had experienced an illness that was chronic but not life-threatening in the short term. However, when we re-analysed the data with ‘serious illness (self)’ as a traumatic event the PTSD scores of the traumatic and life events groups still did not differ (total log PTSD score 0.68 in both groups).

As suggested by Ben-Ezra & Aluf we have also excluded accidents and sudden deaths from the trauma events group, since this might be a heterogeneous group regarding the magnitude of the event. This resulted in a mean total log PTSD score of 0.76 (v. 0.71), which is not an essential change compared with the original difference.

Ben-Ezra & Aluf argue that the magnitude (severity) of an event is related to the likelihood of developing PTSD, and that we should have allotted events to either of our two groups on the basis of their severity. We agree that symptoms are related to severity but we found a striking overlap in PTSD symptomatology after life events and traumatic events (Tables 2 and 4) and similar mean symptom levels (Table 3).

The severity of an event can be assessed objectively and subjectively. Ben-Ezra & Aluf allude to the objective assessment but the subjective appraisal of an experience also plays an important role (Reference McNally, Bryant and EhlersMcNally et al, 2003). It is likely that objective and subjective severity are associated with PTSD symptoms after both traumatic and life events.