We read with great interest the paper by Mol et al (Reference Mol, Arntz and Metsemakers2005). However, we would like to point out some weaknesses. First, ‘serious illness (self)’ was considered a life event rather than a traumatic event. There is a vast literature concerning post-traumatic stress disorder (PTSD) in people with AIDS and cancer. Serious illness definitely meets the DSM-IV criterion A1 for life-threatening situations (Reference Barak, Achiron and RotsteinBarak et al, 1998).

Second, there is a big problem with Mol et al's definition of ‘sudden death’ of loved ones, which ranges from watching a loved one die violently to hearing about the death of a loved one or a close relative. The same is also true for accidents and serious illness. The magnitude of a traumatic event is linked directly to PTSD symptomatology (Reference Sunger and KayaSunger & Kaya, 2001). If we were to exclude sudden death and accidents from the traumatic events group we would see a significant difference between the life events group and the traumatic events group, with more symptoms in the latter. This is a crucial point since most people in the traumatic events group reported sudden death or accident as their worst event; they also had a low level of PTSD symptomatology. If Mol et al had sub-divided the sudden death and accident groups according to the magnitude of the event, this would have helped to determine whether the event could be considered a life event or a traumatic event. This is important when dealing with the issue of traumatic grief (Reference Stroebe, Schut and FinkenauerStroebe et al, 2001), which is a combination of PTSD and bereavement. If the participants had undergone normal grieving the sudden death should be considered a life event rather than a traumatic event.

Third, the magnitude of the traumatic event was clearly associated with PTSD symptomatology, with extremely traumatic situations such as physical and sexual abuse being related to very high PTSD symptomatology scores. However, the number of participants with exposure to such traumatic events was very small (n=9 for physical abuse, n=4 for sexual abuse and n=13 for physical and sexual abuse as a child).

Notwithstanding, the basic message of the paper is important: the line between life events and traumatic events is at best thin, and sometimes nonexistent. The best support for this can be found in the case of the Dutch farmers (Reference Olff, Koeter and Van HaaftenOlff et al, 2005) whose cattle were exposed to foot and mouth disease leading to the killing of the herds. This was not a life-threatening event for the farmers, but was a major life event that can easily be considered traumatic.