1

Evaluating the Performance of Hospitals

 Using DEA to measure the performance of each hospital relative to the

performance of all four hospitals
 Input Measures
1. The number of full-time equivalent (FTE) nonphysician personnel
2. The amount spent on supplies
3. The number of bed-days available
 Output Measures
1. Patient-days of service under Medicare
2. Patient-days of service not under Medicare
3. Number of nurses trained
4. Number of interns trained

2
 In this application of DEA, a linear programming model
is developed for each hospital whose efficiency is to be
evaluated.

3
 DEA Linear Programming Model
 To determine the weight that each hospital will have in computing the
outputs and inputs for the composite hospital, we use the following
decision variables:
wg = weight applied to inputs and outputs for General Hospital
wu = weight applied to inputs and outputs for University Hospital
wc = weight applied to inputs and outputs for County Hospital
ws = weight applied to inputs and outputs for State Hospital

 The DEA approach requires that the sum of these weights equal Thus,
the first constraint is
wg + wu + wc + ws = 1

4
 ANNUAL SERVICES PROVIDED (OUTPUTS) BY THE FOUR HOSPITALS

Every DEA linear programming model will include a constraint that requires the weights for the operating
units to sum to 1.

5
 for each output measure, the output for the composite hospital is determined by
computing a weighted average of the corresponding outputs for all four
hospitals. For instance, for output measure 1, the number of patient-days of
service under Medicare, the output for the composite hospital is:

 Substituting the number of Medicare patient-days for each hospital as shown in

Table 2, we obtain the following expression:
Medicare patient-days for Composite Hospital = 48.14wg + 34.62wu + 36.72wc +
33.16ws
 the general form of the output constraints is:

Output for the Composite Hospital >= Output for County Hospital

6
RELATIONSHIP BETWEEN THE OUTPUT MEASURES FOR THE FOUR
HOSPITALS AND THE OUTPUT MEASURES FOR THE COMPOSITE HOSPITAL

7
Because the number of Medicare patient-days for County Hospital is
36.72, the output constraint corresponding to the number of Medicare
patient-days is

Formulated a constraint for each of the other three output measures, with
the results as shown:

8
 Input for the Composite Hospital <= Resources available to the
Composite Hospital

 For each input measure, the input for the composite hospital is a
weighted average of the corresponding input for each of the four
hospitals. Thus, for input measure 1, the number of full-time equivalent
nonphysicians, the input for the composite hospital is:

9
 obtain the following expression for the number of full-time equivalent
nonphysicians for the composite hospital:
285.20wg + 162.30wu + 275.70wc + 210.40ws

 To complete the formulation of the input constraints, we must write

expressions for the right-hand-side values for each constraint. First,
note that the right-hand-side values are the resources available to the
composite hospital. In the DEA approach, these righthand- side values
are a percentage of the input values for County Hospital. Thus, we must
introduce the following decision variable:
E = the fraction of County Hospital’s input available to the Composite
Hospital

10
RELATIONSHIP BETWEEN THE INPUT MEASURES FOR THE FOUR
HOSPITALS AND THE INPUT MEASURES FOR THE COMPOSITE HOSPITAL

11
 the input constraint corresponding to the number of FTE nonphysicians
available to the composite hospital:
285.50wg + 162.30wu + 275.70wc + 210.40ws <= 275.70E
 the input constraints for the supplies and bed-days are written as follows:

123.80wg + 128.70wu + 348.50wc + 154.10ws <= 348.50E Supplies

106.72wg + 64.21wu + 104.10wc + 104.04ws <= 104.10E Bed-days
• The DEA efficiency conclusion is based on the optimal objective function
value for E. The decision rule is as follows:
• If E = 1, the composite hospital requires as much input as County Hospital
does. There is no evidence that County Hospital is inefficient.
• If E < 1, the composite hospital requires less input to obtain the output
achieved by County Hospital. The composite hospital is more efficient;
thus, County Hospital can be judged relatively inefficient

12
13
 The Slack/Surplus column provides some additional information about the efficiency of
County Hospital compared to the composite hospital.
 Specifically, the composite hospital has at least as much of each output as County Hospital
has (constraints 2–5) and provides 1.6 more nurses trained (surplus for constraint 4) and 37
more interns trained (surplus for constraint 5).
 The slack of zero from constraint 8 shows that the composite hospital uses approximately
90.5% of the bed-days used by County Hospital.
 The slack values for constraints 6 and 7 show that less than 90.5% of the FTE nonphysician
and the supplies expense resources used at County Hospital are used by the composite
hospital.
 The composite hospital is more efficient than County Hospital, and we are justified in
concluding that County Hospital is relatively inefficient compared to the other hospitals in
the group.
 Given the results of the DEA analysis, hospital administrators should examine operations to
determine how County Hospital resources can be more effectively utilized.

14
THE SOLUTION FOR THE COUNTY HOSPITAL DATA
ENVELOPMENT ANALYSIS PROBLEM

15
THE SOLUTION FOR THE COUNTY HOSPITAL DATA
ENVELOPMENT ANALYSIS PROBLEM (cont..)

16
 Data Envelopment Analysis

 Data envelopment analysis (DEA) is an LP application

used to determine the relative operating efficiency of
units with the same goals and objectives.
 DEA creates a fictitious composite unit made up of an
optimal weighted average (W1, W2,…) of existing units.
 An individual unit, k, can be compared by determining
E, the fraction of unit k’s input resources required by
the optimal composite unit.
 If E < 1, unit k is less efficient than the composite unit
and be deemed relatively inefficient.
 If E = 1, there is no evidence that unit k is inefficient,
but one cannot conclude that k is absolutely efficient.

17
 Data Envelopment Analysis

 The DEA Model

Min E
s.t. Weighted outputs > Unit k’s output
(for each measured output)
Weighted inputs < E [Unit k’s input]
(for each measured input)
Sum of weights = 1
E, weights > 0

18
 Data Envelopment Analysis

The Langley County School District is trying to

determine the relative efficiency of its three high
schools. In particular, it wants to evaluate Roosevelt
High.
The district is evaluating performances on SAT
scores, the number of seniors finishing high school,
and the number of students who enter college as a
function of the number of teachers teaching senior
classes, the prorated budget for senior instruction,
and the number of students in the senior class.

19
 Data Envelopment Analysis

 Input

Roosevelt Lincoln Washington

Senior Faculty 37 25 23
Budget ($100,000's) 6.4 5.0
4.7
Senior Enrollments 850 700 600

20
 Data Envelopment Analysis

 Output

Roosevelt Lincoln Washington

Average SAT Score 800 830
900
High School Graduates 450 500
400
College Admissions 140 250
370

21
 Data Envelopment Analysis

 Define the Decision Variables

E = Fraction of Roosevelt's input resources required
by the composite high school
w1 = Weight applied to Roosevelt's input/output
resources by the composite high school
w2 = Weight applied to Lincoln’s input/output
resources by the composite high school
w3 = Weight applied to Washington's input/output
resources by the composite high school

22
 Data Envelopment Analysis

 Define the Objective Function

Minimize the fraction of Roosevelt High School's input
resources required by the composite high school:
Min E

23
 Data Envelopment Analysis

 Define the Constraints

Sum of the Weights is 1:
(1) w1 + w2 + w3 = 1

Output Constraints:
Since w1 = 1 is possible, each output of the composite
school must be at least as great as that of Roosevelt:
(2) 800w1 + 830w2 + 900w3 > 800 (SAT Scores)
(3) 450w1 + 500w2 + 400w3 > 450 (Graduates)
(4) 140w1 + 250w2 + 370w3 > 140 (College Admissions)

24
 Data Envelopment Analysis

 Define the Constraints (continued)

Input Constraints:
The input resources available to the composite school is
a fractional multiple, E, of the resources available to
Roosevelt. Since the composite high school cannot use
more input than that available to it, the input constraints
are:
(5) 37w1 + 25w2 + 23w3 < 37E (Faculty)
(6) 6.4w1 + 5.0w2 + 4.7w3 < 6.4E (Budget)
(7) 850w1 + 700w2 + 600w3 < 850E (Seniors)
Nonnegativity of variables:
E, w1, w2, w3 > 0
25
 Data Envelopment Analysis

 Computer Solution

Objective Function Value = 0.765

Variable Value Reduced Cost
E 0.765 0.000
W1 0.000 0.235
W2 0.500 0.000
W3 0.500 0.000

26
 Data Envelopment Analysis

 Conclusion
The output shows that the composite school is made
up of equal weights of Lincoln and Washington.
Roosevelt is 76.5% efficient compared to this composite
school when measured by college admissions (because
of the 0 slack on this constraint #4).
It is less than 76.5% efficient when using measures
of SAT scores and high school graduates (there is
positive slack in constraints 2 and 3.)

27
End of Chapter 5, Part A

28