Doubts about the effectiveness of medical care in improving patient he
alth have been raised by epidemiological studies and by studies of geo
graphical variation and inappropriate use of health care. To investiga
te this problem, the life expectancy gain (LEG) from consecutive admis
sions to a department of internal medicine during a six-week period wa
s assessed by two expert panels, each consisting of an internist, a su
rgeon and a general practitioner. The mean LEG for all admissions was
2.25 years (n = 422). Sixty-one percent had a LEG of 0.10 years or les
s, while 5% had a LEG of more than 9.98 years. In a probabilistic sens
itivity analysis, the mean LEG remained greater than zero under assump
tions oi overestimated positive LEG and underestimated negative LEG. W
e conclude that the life expectancy of the majority of the patients wa
s not influenced by the admission, but that a minority had substantial
gains, resulting in a high overall mean LEG. (C) 1997 Elsevier Scienc
e Inc.