An intriguing aspect of the meta-analysis by Chiles, Lambert, and Hatch (th
is issue) is the finding that medical cost offsets associated with psycholo
gical interventions vary across patient populations. Our commentary describ
es four types of patients: the worried well, the worried unwell, the chroni
cally disabled, and the chronically ill. We suggest the kinds of behavioral
medicine interventions most likely to lead to restored health and decrease
d medical utilization for each patient type, and the changes in research an
d practice necessary to maximize clinical psychology's impact on physical h
ealth intervention.