Objective: To elucidate the nature of the etiologic relationship betwe
en personality and major depression in women. Design: A longitudinal t
win design in which twins completed a time 1 questionnaire and, 15 mon
ths later, were personally inter-viewed for the occurrence of major de
pression during the last year and completed a time 2 questionnaire. Bo
th questionnaires contained short forms assessing neuroticism and extr
aversion. Participants: 1733 twins from female-female pairs ascertaine
d from the population-based Virginia Twin Registry. Results: Extravers
ion was unrelated to lifetime or 1-year prevalence of major depression
. Neuroticism was strongly related to lifetime prevalence of major dep
ression and robustly predicted the prospective 1-year prevalence of ma
j or depression in those who, at time 1, denied previous depressive ep
isodes. However, controlling for levels of neuroticism at time 1, leve
ls of neuroticism at time 2 were moderately elevated in those who had
had an episode of major depression between times 1 and 2 (''scar'' eff
ect) and substantially elevated in those experiencing an episode of ma
jor depression at time 2 (''state'' effect). In those who developed ma
jor depression, levels of neuroticism did not predict time to onset. I
n the best-fit longitudinal twin model, the proportion of the observed
correlation between neuroticism and the liability to major depression
that is due to shared genetic risk factors was estimated at around 70
%, that due to shared environmental risk factors at around 20%, and th
at due to a direct causal effect of major depression on neuroticism (v
ia both ''scar'' and ''state'' effects) at around 10%. Approximately 5
5% of the genetic liability of major depression appeared to be shared
with neuroticism, while 45% was unique to major depression. Conclusion
: In women, the relationship between neuroticism and the liability to
major depression is substantial and largely the result of genetic fact
ors that predispose to both neuroticism and major depression.