Users' Guides to the Medical Literature
Guyatt G, Rennie D, Meade MO, Cook DJ
Part G Moving from Evidence to Action
Chapter 22.2. Decision Making and the Patient
Victor M. Montori, P. J. Devereaux, Sharon Straus, Brian Haynes, Gordon Guyatt
Severe Stroke
Topics Discussed:
alteplase, atrial fibrillation, cerebrovascular accident, clinical decision making, decision analysis, decision making, decision support techniques, gastrointestinal bleeding, incorporating patient values and preferences, measuring patient experience, myocardial infarction, patient-important outcome, streptokinase, stroke prevention, visual analogue pain scale
Excerpt:
"Table 22.2-1 summarizes decision-making
approaches theoretically available to the clinician and patient
facing an important decision.When clinicians offer patients minimal information about the
options and make the decision without patient input, a style commonly
referred as parental or paternalistic, they are not considering
patient values and preferences. This does not mean that
patients do not have an opportunity to express their wishes, but
they may do so in a delayed fashion and through actions. For instance,
if the choice was not consistent with their values and preferences,
then patients may not act on the decision or may quickly abandon
the plan shortly after the visit with the clinician. To the extent
that EBM requires the incorporation of patient values and preferences
in decision making, a parental approach to clinical decisions is
inconsistent with the practice of EBM.Heyland et al5 asked 120 at-risk patients
to consider whether they would prefer streptokinase or tissue plasminogen
activator (TPA) if they were to have a myocardial infarction. To
obtain their informed preference, they used a decision tool that
described the outcomes (myocardial infarctions, death,
and thrombolytic-associated stroke) and the likelihood of death
and stroke when using TPA and streptokinase (ie, TPA use was associated
with 9 fewer deaths and 4 more strokes per 1000 patients treated
compared with streptokinase) derived from the Global Use of Strategies
to Open Occluded Coronary Arteries (GUSTO) trial.6 The
tool worked insofar as all patients chose TPA when the difference
in stroke risk was reduced to 0, and all patients chose streptokinase
when the difference in mortality risk was reduced to 0. Decision
analyses under the expected utility theory assumptions found
TPA to be the dominant option. In the study, only half of the patients,
however, chose TPA. The other half might have considered the additional
1% mortality benefit TPA afforded to not be worth the 0.33% additional stroke
risk and therefore opted for streptokinase.Devereaux et al9 used a technique called probability tradeoff
(including clear descriptions of the outcomes of interest and iterative
testing of preferences with changing likelihood of competing outcomes)
to determine the strength of preference for anticoagulation to prevent stroke
in 61 at-risk patients and 63 physicians who treated patients
with atrial fibrillation. The figure in this box shows the maximum
number of excess upper gastrointestinal bleeding episodes per 100
patients treated to prevent 8 additional strokes (4 major and 4
minor) that patients and physicians found acceptable. The figure
shows the following: (1) there is variability in stroke aversion
among patients and among physicians; (2) patients seem more stroke
averse than physicians; (3) physicians seem more averse to adverse
outcomes that they "cause" with their prescription
(eg, bleeding) than to adverse outcomes that result from clinical
course (eg, strokes). If one believes that patient preferences should
guide treatment, these data suggest the following: if clinicians
fail to incorporate patient values and preferences in the decision-making
process, they will recommend against anticoagulation more often
than is appropriate and, depending on which physician patients see,
they will or will not get the treatment they would prefer.To effectively communicate the nature of the options, researchers
have devised and tested tools called decision
aids. These tools are an alternative to the use of intuitive
approaches to communicating concepts of risk and risk reduction
that clinicians may have developed through clinical experience. Decision
aids present, in a patient-friendly manner, descriptive and
probabilistic information about the disease, treatment options,
and potential outcomes.16-18 A well-constructed decision
aid is valuable in that someone has performed a systematic
review of the literature and produced a rigorous summary of
the outcomes and their probabilities. Clinicians who doubt
that the summary of probabilities is rigorous can review the primary studies on which those probabilities
are based and, using the principles of this book, determine their
accuracy. Furthermore, a well-constructed decision aid offers
a pretested and effective way of communicating information to patients
who may have little background in quantitative decision making.
Most commonly, decision aids use visual props to present
the proportion of people who experience the outcomes of importance
with and without the intervention (Figure
22.2-2A and B).Having a mild stroke causes you to slur your words. After a mild stroke,
you are able to fully understand what is being said to you. Your
thoughts remain clear and you can carry out a conversation without
much trouble, but sometimes you cannot find the right word to use.
Your thinking ability is otherwise normal. There is some weakness
and numbness in your right arm and your face has a slight droop.
You are able to feed, dress, and bathe yourself. However, you cannot
grip objects as tightly as you could before the stroke, objects sometimes
fall from your hands, and you have difficulty writing. Your condition
will not get better.Should clinicians interested in practicing EBM and expecting
to make clinical decisions that incorporate the values and preferences of
the informed patient use one or more of the above approaches for all
decisions? The ultimate constraint of clinical practice is time. Many
clinicians have more to do in each encounter than they did in the
past.24-26 Attention to the patient's
agenda competes with other activities that clinicians ought to do
(eg, documentation, routine preventive care27)
during visits that have not increased in duration to accommodate
these additional activities and demands. Thus, it is not surprising
that clinicians frequently cite time as a key barrier to patient
education about options and to enhanced patient participation in
decision making.28-32
Table 22.2-2 provides
some suggestions for what to do when time is limited...."
Log in to read the full chapter:
Get full access to JAMAevidence two ways:
Subscribe to JAMAevidence
JAMAevidence is a subscription-
based website dedicated to the learning, teaching, and practicing of evidence-based medicine.
Pay Per View
Timed access to all of JAMAevidence
24 hours for $29.95
48 hours for $49.95