Schizophrenia is a devastating, highlyheritable form of psychosis. It has been suspected that genetic risksare exacerbated by adverse life events, but until now there has beenlittle conclusive evidence of such an effect on the individual level.This month�s issue of the Archives of General Psychiatry contains acompelling study by Dr. Ali Khashan of the link between a mother's experience of tragic events and her child's schizophrenia.

Thetwo most difficult parameters for Dr. Khashan to define were trauma andvulnerable time period. As the definition of trauma is nebulous, Dr.Khashan narrowed the criteria down to four tragic events afflictingfirst degree relatives: death, heart attack, stroke, and diagnosis ofcancer. As he was unsure what time period would reveal vulnerability,he delineated four periods: six months before pregnancy, firsttrimester, second trimester, and third trimester. Pregnancy may havebeen chosen as the focus over early childhood because infections during pregnancy have already been correlated with schizophrenia.

Thescale of this study is truly impressive; 1.38 million Danish birthrecords were connected to hospital and genealogical records. Huttunenand Niskanen reported on a trauma-psychosis link as early as 1978, buttheir cohort size was much smaller and they looked only at one factor,the death of a spouse. Other studies have been as large, but lookedonly at the general impact of something like a war on the entirepopulation. This study discovered the individual risk: a mother exposedto a traumatic life event was 44% more likely to have a schizophrenicchild. This extra risk rose to 67% when only the death of a relativewas counted. This was after adjusting for sex, place of birth, familyhistory of illness, and maternal age. It is possible that there areadditional confounding factors that were not accounted for, such asgenes that increase the risk of both physical and mental illness.

Exposurerisk was not equal for every investigated interval. In fact, only thefirst trimester showed significantly elevated risk. The secondtrimester showed marginally elevated risk, the third none, andastoundingly, the 6-month period before pregnancy showed a slightdecrease in risk. This considerably narrows the search for possiblemechanisms by which trauma elevates risk. For example, the fact thatpre-pregnancy tragedy showed no increase in risk allows us to rule outthings like economic difficulties. It is likely that the fetus isdamaged by biochemical changes in the mother during the earliest andmost crucial months of its development. The authors of the studyexplain that before 3 months the fetus does not have an adequateprotection in place against over-activity of thehypothalamic-pituitary-adrenal axis and stress hormones.

Thisstudy adds another piece of evidence that neural development isvulnerable in the first trimester. While mothers obviously cannot avoidthe death of a loved one in order to spare their children, theseresults do have practical value. We have reported recently on advancesin schizophrenia detection andearly treatment, and mothers now know of a factor that may warrantextra monitoring of their children. In addition, it may be possible formothers to work on reducing stress, or receive supplements that shieldtheir fetuses from the worse effects of that stress.