In the 30 years that therapeutic drug monitoring (TDM) has moved from an ab
stract consideration to a routine intervention, issues remain over justifyi
ng the benefits in the light of the ever-increasing competition for budgeta
ry resources. Resolving the issues is constrained by various methodological
concerns. These include considerations such as: (i) the changed environmen
t of knowledge and practice during the generation in which TDM has been use
d and evaluated; (ii) the predominance of studies using system-related rath
er than patient-centred outcomes; (iii) using a timeframe for analysis that
is too short; (iv) a lack of rigour in many of the pharmacoeconomic analys
es; and (v) excessive use of a site-specific rather than a societal perspec
tive. Current observation suggests that the greatest benefit of TDM accrues
from targeted or specialty populations: those with severely decompensated
renal function, those at the extremes of age, and those using immunosuppres
sive, some antineoplastic, some psychotherapeutic and some anticonvulsant d
rugs. In these situations, safe and humane practice considers TDM a necessi
ty without respect to cost. But for many routine situations with drugs for
which TDM has commonly been used in the past, present reliance on the inter
vention may have become excessive in the light of today's knowledge base of
practitioners.