The costs associated with depressive disorders are substantial, being
related not so much to drug acquisition (which contributes around 10%
of the total costs) as to hospitalization, physician attendance and nu
rsing care. In addition, there are indirect costs arising out of depre
ssion which is untreated, or ineffectively treated, because of inappro
priate or non-diagnosis; these include the economic losses incurred th
rough reduced work output and the demands made upon various social sup
port systems, as well as intangible personal costs to the patient, fam
ily, friends and others. In choosing a cost-effective medication for d
epressive illness, it is necessary to examine all available evidence f
or the efficacy of the range of therapies which are currently availabl
e. This has to be done by imposing strict criteria upon the methodolog
ical soundness of the studies which are to be taken into account. In a
ddition to efficacy data, attention has also to be given to the likeli
hood of treatment drop-outs, occasioned either by adverse effects asso
ciated with treatment, or by treatment inefficacy, because the seconda
ry medication which needs then to be instituted carries its own additi
onal costs. In recent studies in which efficacy and drop-out rates hav
e been determined for various classes of antidepressant drugs on the b
asis of carefully conducted meta-analyses, the insertion of the result
ing values into decision analysis models for treating inpatients and o
utpatients suggests that venlafaxine is cost-effective when compared w
ith tricyclic antidepressants and selective serotonin reuptake inhibit
ors and that its inclusion in formularies therefore may be warranted.