Bayesian Network Based Triage Successfully Predicts ER Disposition

brewerobstructionΤεχνίτη Νοημοσύνη και Ρομποτική

7 Νοε 2013 (πριν από 3 χρόνια και 8 μήνες)

79 εμφανίσεις


Bayesian Network Based Triage Successfully Predict
s

ER
Disposition

Sarmad Sadeghi, MD, Curtis Kennedy, MD, Craig W. Johnson, Ph. D.

University of Texas Health Science Center at Houston


Abstract:

A recently developed Bayesian network
based triage system was evaluated against an
emergency room (ER) physician’s triage assessment
to determine its classification performance. Binary
logistic regres
sion analysis revealed that the Bayesian
network’s classification significantly predicted actual
emergency room disposition, whereas the ER
physician’s classification did not. These findings
support expansion of the pilot project to obtain more
definitive
analyses.


Background:
Bayesian networks used for clinical
prediction are becoming more common
1
. We are
assessing the utility of such a tool to predict ER
disposition in a population of patients who
presented
to

an emergency department with a chief complai
nt
of abdominal pain.


Methods:

Our data set is from a retrospective review
of the historical elements (excluding physical
examination findings)
of
95 consecutive ER patient
records where the chief complaint was abdominal
pain. 5 records were excluded from

the analysis
because of missing disposition.

We performed
binary logistic regression on the data to determine the
accuracy of the Bayesian network (BN) and the ER
physician (MD). We included “go to ER” (serious)
as our predictor for both BN and physicia
n:. We used
the actual ER disposition as the dependent measure:
admit versus discharge or referral. There were
insufficient cases in the physician “call 911” category
(most serious)
for use in any analysis.


Results:

42 of the 90 patients were admitted.
The
BN classification was significantly correlated with
admission (p = 0.014) while physician classification
was not (although it approached significance, at p=
0.053). Table 1 summarizes binary logistic regression
models, with their associated significanc
e and odds
ratios. Tables 2 and 3 are classification accuracy
tables for BN and physician, respectively, and
contain sensitivity, specificity, and positive (PPV)
and negative (NPV) predictive values.

Table 1: Binary logistic regression results:
significan
ce

(Sig.)
, odds
-
ratio

(OR) and 95%
confidence interval
, coefficient (B)

and
s
tandard
e
rror

(SE)
.



Table 2: Bayesian network classification table.
Sensitivity =
81%, Specificity = 44%, PPV = 56%,
NPV = 72%.


Table 3: Physician classification table. Sensitivity =
64%, Specificity = 56%, PPV = 56%, NPV = 64%.


Discussion:

In general, the Bayesian network is
more conservative than the ER physician (it assigns a
higher level of disposition). Given the potential
application of serving as an initial triage system for
people with medical concerns, we feel the negative
impact o
n specificity is acceptable since it is safer to
err on the side of caution.


The utility of this application is likely to be
triaging potential patients that are still at home. Since
the subjects are from a population that has already
presented to th
e ER, these results cannot yet be
generalized to the population at large.


Based on this preliminary analysis, we feel further
research is justified and necessary to better describe
the scope of potential benefits that the Bayesian
network triage syste
m may provide.


Reference:
1. P.J.F. Lucas, H. Boot and B.G. Taal.
Decision
-
theoretic network approach to treatment
management and prognosis.
Knowledge
-
based
Systems
11 (1998) 321

330.

Model

Sig.

OR

(95%

CI)

B

S.E.

BN:ER

0.014

3.3
(1.
27
-
8.6
2
)

1.196

0
.48
9

MD:ER

0.053

2.3
(0.9
9
-
5.4
2
)

0.839

0
.434

ER Admission

Prediction

False True

Total

False

21

27

48

True

8

34

42

Total

29

61

90

ER Admission

Prediction

False True

Total

False

27

21

48

True

15

27

42

Total

42

48

90