Both kinds of studies supported the
ERI assumption, but only longitudinal studies could
strengthen the predictive value of the model. In addition,
and even more important, with the help of longitudinal
studies the direction of the relation between ERI and
adverse health could be further examined.