Turton et al (Reference Turton, Hughes and Evans2001) describe substantial morbidity in pregnant women whose previous pregnancy (or pregnancies) had ended in spontaneous loss after 18 weeks' gestation. She found that 21% of women reporting stillbirth as a criterion A stressor had post-traumatic stress disorder (PTSD) symptoms at caseness level in the third trimester of the next pregnancy. Turton et al conclude that women are vulnerable to PTSD in the pregnancy following stillbirth.

I take issue with these findings. There are several methodological problems with the study. First, stillbirth is not defined as pregnancy loss after 18 weeks' gestation. An infant born after the 28th week of gestation who does not breathe after birth or show any other sign of life is termed a stillbirth (Reference Beischer and MackayBeischer & Mackay, 1988). Hence, by definition, Turton et al have included 41 women (out of their total number of 66 subjects) who have had miscarriages. It would have been better to report foetal loss figures on babies with a birth-weight of <500 g, which is current widespread practice. Second, the authors state that 14 out of 66 women did not see their stillborn infants. No reason is given for this. Was this because of the gestational age of the infant (<28 weeks' gestation)? Third, the use of the term PTSD must be questioned. The authors describe stillbirth as a criterion A stressor. One would therefore expect the onset of PTSD within 6 months of the stillbirth. The authors appear to have ignored this time criterion in making a diagnosis of PTSD (World Health Organization, 1993). Similarly, it is difficult to see how the persistent avoidance criterion (criterion C) was met. None of the subjects avoided pregnancy but became pregnant following stillbirth. What the authors describe are symptoms precipitated by the subsequent pregnancy, with the previous ‘stillbirth’ as a vulnerability or predisposing factor. Perhaps the diagnosis of adjustment disorder would be more appropriate