A meta-analysis examined the relationship between psychosocial factors and
the development of breast cancer. Average effect sizes (Hedges's g) were ca
lculated from 46 studies for 8 major construct categories: anxiety/depressi
on, childhood family environment, conflict-avoidant personality, denial/rep
ression coping, anger expression, extraversion-introversion, stressful life
events, and separation/loss. Significant effect sizes were found for denia
l/repression coping ig =.38), separation/loss experiences (g =.29), and str
essful life events (g =.25). Although conflict-avoidant personality style w
as also significant (g =.19), the effect size was less robust, and a modera
te number of future studies with null results would reduce the significance
. Results overall support only a modest association between specific psycho
social factors and breast cancer and are contrary to the conventional wisdo
m that personality and stress influence the development of breast cancer.