3
the
can
individualize the
the
Asubjects
required
format,
to
physician
guidelines
during
testing phase.
Group
a
mean of three
to reach their
conclusions,
any patient.
pages
1.5 minutes
To evaluate this DSS for
we
read
to
and
scenario
preventive
averaged
they
therapy,
per
recommendation.
scenarios of
in-
and
reach
tuberculosis
the cases
The
a
developed eight patient
fection Four local tuberculosis
intui-
found the tool
A
(appendix).
experts
investigational
group
validated the DSS
that
the recommen-
DSS was the
recommenda-
to
and effective.
The
use,
Apartici-
tive,
by assuring
the
easy
group
dation
rated the
tool
as
web-based
provided by
expert
pants
user-friendly,
very
tion an
would
the
with a score of
4.56
0.17
sub-
All
(mean
or5 ona
ATS/CDC
DSSrec-
SEM).
provide
using
All
four
with the
rated the tool
1
either
4
scale,
jects
with
When
experts agreed
guidelines.
five-point
the
ommendations for
scenarios 100% of the
useless and
5
eight
being
rating
user-friendly.
being very
time.
the effectiveness of the
tool
web-based
tested the DSS’s effectiveness in
for
Wethen
when to use
treat-
providing
internal
recommending
preventive
recommendations
rated the tool 4.36
0.16
(mean
ment,
using general
participants
guideline
medicine residents. This evaluation was done dur-
on
a
with 1
scale,
SEM)
useless,
five-point
being
a
scheduled
5
and
effective.
and
ar-
conference,
ing
regularly
teaching
being very
the residents were not
rived for the
selected. As
uniquely
conference,
they
the residents were ran-
to
a
computer-based
domly assigned
group (group
Discussion
n =
and a written-resource-based
group
A,
12),
n
The two
were
The decision to treat
has
with tuberculosis
17).
B,
groups
presented
patients
implications
(group
sub-
forbothindividual
case
and
were
they
with the same
scenarios,
eight
they
prophylaxis
infected
would treat or not treat each
and the
asked whether
jects
public. Appropriately treating
in
with INH.
A
WWW
accessed the DSS tool on
will
individual disease and assist
prevent
Group
patients
patient
a
health The
efforts.
are
a
Tl line over the
4.0
ATS/CDC
using
did not receive
Netscape
public
based on skin-test reactions to
decision to treat is based on
guidelines
PPD. The
5 TU of
This
browser.
any specific
accessing
treatment
group
the DSS
in
to
it.
a
balance of risk of
using
prior
They
training
asked to decide on
disease
risk of
and disease
and
were
(test
preventive
using
prevalence)
specificity
To assist
B
in group
the DSS. Theresidents
wereallowed
the
ATS/
clinicians,
only
drug toxicity.
to use a
card we have
orto
at our
CDC established &dquo;cut
size,
These
based on skin-test re-
guideline
(figure
developed
other
points&dquo;
underlying
institution
resource
use
risk of
and risk of
2)
would
disease,
thus
any
paper-based
sponse
consult. After
clini-
complet-
drug
provide
they
usually
toxicity.4
guidelines
for the
the resi-
cians with
in
their choices
scenarios,
patient
ing
important guidance
for their with
choosing drug
in
A
dents
were asked to evaluate the com-
tuberculosis infec-
patients
group
therapy
a
short
tion.
puter program by answering
questionnaire.
The results for the individual case scenarios were
The
initial
for
guidelines
treating skin-test-posi-
Students’s
andsignificant dif-
tive
were
with
t-test,
However,
compared
patients
using
straightforward.
<
ferences were defined as
0.05.
of the
of
p
increasing understanding
epidemiology
skin-test
with increased use of immuno-
reactivity,
andwith the
hu-
of
appearance
suppressive agents,
man
in
the
virus
(HIV)
have become more
immunodeficiency
popula-
Results
the
tion,
guidelines
complex by
each
for the
scenarios
We are
The results for
in
unaware of
studies
eight
computer-DSS
group
Group
necessity.
any
investigating howaccurately
physiciansapply
directly
1.
A
are shown
table
thelat-
(the
with the
est
demonstrated better
to individual
How-
with
con-
ATS/CDC
compliance
patients.
compliance
guidelines
group)
ATS/CDC
guidelines
for each scenario than did
a
recent
of
ever,
guidelines
written-resource
study
physician
B
the WHO tuberculosis-treatment
cluded that
(the
Overall,
group
recommenda-
group).
ATS/CDC
group
A
reached the
waslow.s
appropriate
possible
compliance amongphysicians
X
12 sub-
tions in
cases
scenarios
These data
that our results are not
(8
92/96
in the
unique.
suggest
or 95.8%. The residents
As
a
we have
a
of
in group
profession,
long history
and clinical
skills
jects
B,
group),
trying
as the
to
maintain our decision
reached the same conclusions
however,
through
in
cases
medical
education
traditional
or
(8
(CME)
77/136
ATS/CDC
scenarios
difference
One
possible
or56.6%.
continuing
education
guidelines
only
in the
X
This
units
not
17
(CEUs).6
subjects
group),
continuing
most
Unfortunately,
<
0.001.
CMEattempts
was
at
do
their
of
p
accomplish
statistically significant
goal
criticism
DSSs
or to
attitudes and
behaviors.
of
Those
altering physicians’
frequent
computer-assisted
that do
learn
that
time to
succeed share somecharacteristics. Theyfo-
is
they require significant
the server
cus on
to
we had
and
measure
are
use. In order
available as
this,
activities,
legitimate
they
time
logall
by
they
selections
and
reminders for
when
see
computer
by
computer
physicians
patients.’
Zweifel
