| |
As
was the case with estimating probabilities accurately, assessing
the desirability or utility of various outcomes is also essential
for good decision making. Choosing wisely among options is impossible
if one’s evaluation of an option is inaccurate. To demonstrate
the evaluation of outcomes consider Brickman,
Coates, and Janoff-Buhlman (1978).
Seeing
the results, most people are surprised to learn that the newly
rich lottery winners rated their happiness at this stage of their
life as only 4.0, which does not differ significantly from the
rating of the control subjects. Also surprising to many people
is the fact that the paraplegics and quadriplegics rated their
lives at 3.0, which is above the midpoint of the scale (2.5).
To
the extent that people cannot predict the utility of unknown states
of health, to that extent they are unable to make decisions which
are “in their own best interest.” The ratings given by the control
subjects, the accident victims, and the lottery winners pertain
to their “experienced utility,” which is their opinion of the
state they are actually experiencing at the current time (Kahneman
and Snell, 1992). However,
decisions are often made before all possible outcome states are
experienced. Therefore one can only predict what the experienced
utility will be like. If the prediction is in error, then a decision
contrary to one’s best interests may occur. For example, if people
knew that winning a huge prize in a state lottery would cause
only an insignificant change in their happiness, then many people
would cease buying lottery tickets.
As
individuals face important medical decisions, the outcome states
might be quite unfamiliar. The choice between radiation or surgery
for prostate cancer might hinge on one’s appreciation of the outcome
states following the selection of each option. Can laypersons
with absolutely no experience with the outcome states be expected
to make an adequate appraisal of each of them? Much research suggests
that they cannot. For example, Boyd
et al., (1990) solicited the utility of life with a colostomy
from several different groups: patients who had rectal cancer
and who had been treated by radiation, patients who had rectal
cancer and who had been treated by a colostomy, physicians who
had experience treating patients with gastrointestinal malignancies,
and two groups of healthy individuals. The patients with a colostomy
and the physicians rated life with a colostomy significantly higher
than did the other three groups. Similarly, Sackett
and Torrance (1978) found that those on dialysis rated their
state of health higher than did members of the general public
who predicted how they would rate their state of health if they
were on dialysis. However other researchers have found that people
can make relatively accurate predictions concerning the utility
of impending health states (e.g. Llewellyn-Thomas,
Sutherland, and Thiel, 1993; Rachman
and Eyrl, 1989). To the extent that persons cannot
predict what being in a future health state would be like, it
will be extremely difficult for them to make decisions in their
own best interest. Interactive videos and other educational programs
have been developed to try to inform patients what living with
particular outcomes would be like (e.g. Barry
et al., 1995; Liao
et al., 1996). Whether such programs will enable patients
to rate such health states the same as those already in the health
state is unknown. It is also unknown whether patients should
rate such health states the same as those already in the state.
Individual differences exist, and we should not expect all patients
to rate the same health state in the same way. What is needed
for good decision making, however, is that the patient be able
to predict what the experienced utility will be. This is often
an unrealized expectation.
Note
that the inability to appreciate the utility of future health
states also has important implications for “informed consent.”
If a patient facing a medical decision does not assign the same
utilities to various possible outcome states that a typical person
in each health state assigns, to what extent is the patient sufficiently
well informed to make the decision? Can that person render decisions
in his or her own best interest? Can we blame patients who do
not?
|