9-606-015
R E V : O C T O B E R 2 0 , 2 0 0 6
________________________________________________________________________________________________________________
Professor Andrew F. McAfee prepared this case. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve
as endorsements, sources of primary data, or illustrations of effective or ineffective management.
Copyright © 2005 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685,
write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be
reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical,
photocopying, recording, or otherwise—without the permission of Harvard Business School.
A N D R E W F . M C A F E E
Pharmacy Service Improvement at CVS (A)
On a Thursday afternoon in July of 2002 Jon Roberts, Josh Flum, Tom Grossi, and Mitch Betses
walked into a cluttered conference room at CVS headquarters in Woonsocket, Rhode Island. For
several months, the room had served as the data repository, meeting space, and nerve center for the
company’s Pharmacy Service Initiative (PSI). Most horizontal surfaces were stacked high with
folders, binders, and books, and most vertical ones were covered with whiteboards, sticky notes,
sheets of paper, and hand-drawn flow charts. The four men cleared off enough space to sit down
around a table.
Their eyes were drawn to two recently added pieces of paper on the nearest wall. One was a list
of the problems the PSI team had uncovered during a recent series of observations at CVS
pharmacies around the country (Exhibit 1); the other was a description of the problems encountered
over the course of a single shift by the person staffing the prescription pickup counter in one
pharmacy (Exhibit 2).
Flum looked at Betses. “You told us it was bad, but this bad?”
“I told you there were service issues in our pharmacies. But I have to admit, even I didn’t know
the whole story.”
“So what do we do about it?”
“Well, we can’t have 67 solutions for the 67 problems we identified,” Roberts said.
“Definitely not,” Grossi agreed. “But do you have an idea what we should do? If you erased that
whiteboard and grabbed a pen, could you draw the ‘right’ flow chart for pharmacy operations?”
“Actually, I think I could come pretty close. And I think my flow chart would look a lot like both
of yours. I’m just not sure which parts of it would be easy to implement and which would be tricky.
Mitch, you know these places better than anyone—what kinds of changes would make them really
unhappy?”
“Anything affecting safety. Everyone—not just the pharmacists—is a fanatic about making sure
we fill prescriptions accurately and watch out for the health of our customers. So for example if we
said, ‘In the interests of efficiency we want to have the system spit out fewer alerts about drug-drug
interactions,’ we would get killed. The pharmacists would march us right out the front door of their
stores and tell us never to come back. And I wouldn’t blame them.”
“Got it. What else?”
606-015 Pharmacy Service Improvement at CVS (A)
2
“Anything that increased customer waiting times. People in the pharmacy feel like customers
already wait way too long when they come to pick up prescriptions, especially at peak times. They’re
not in a good mood when they get to the front of the line, and it can get really ugly if after they’ve
waited all that time they’re told their medicine isn’t ready.”
Roberts nodded. “OK. Hand me that whiteboard eraser and pen. Here’s the new process. It
doesn’t degrade safety at all, it decreases waiting time, and it improves customer satisfaction. Of
course, convincing the pharmacies that’s true might not be easy.”
Pharmacy Operations at CVS
The first “Consumer Value Store” opened in Lowell, Massachusetts in 1963. The company grew
quickly after that, both organically and by acquisition, and by 2002 CVS was one of America’s largest
retail drugstores, with over 4,000 stores and revenue of $24.2 billion, over two-thirds of which was
generated by the pharmacies (see Exhibit 3 for selected corporate financial information).1
The Pharmacy Service Initiative
As the company grew, managers started to worry that pharmacy operations were not performing
well. Anecdotes from both customers and employees indicated that many locations had serious
problems with customer service. The company’s pharmacy business, however, grew as quickly as
the industry average. Some interpreted this to mean that CVS did not in fact have serious service
problems.
To understand the true state of pharmacy customer service and to make any required fixes, CVS
launched the PSI and staffed it with operations executives and managers, including Roberts, the
senior vice president of store operations; Flum, the director of store technology; and Betses, the
director of pharmacy operations. Also on the team were a top pharmacy supervisor, a top
pharmacist, and consultants from the Boston Consulting Group, including Grossi.
Interviews and Analysis
The PSI team began gathering information by analyzing historical data and interviewing current
and former customers, as well as customers of other pharmacies. This work quickly confirmed that
problems existed at CVS. As Flum explained:
It was true that we were growing at market rates, but that was only because customers
believed that no one provided great service. If they came to us or stuck with us, it was because
they didn’t think anyone else would take better care of them, not because we were so fantastic.
One of our interviewees said, “I’ve had problems at CVS, but why would I leave? All
pharmacies probably have some problems.”
Luckily for us, they also thought that it was really difficult to switch from one pharmacy to
another. Another interviewee said, “I don’t even know what’s involved in transferring a
prescription. Do I have to call my doctor to get a new prescription? It just seems like it would
be such a hassle.”
1 Pharmacies were responsible for a roughly equivalent share of CVS profits.
Pharmacy Service Improvement at CVS (A) 606-015
3
Actually, it’s not a hassle for the customer at all. We’re required by law to immediately
transfer customer records to another pharmacy whenever asked. It’s a good thing for us that
we weren’t asked more often.
Even though customers believed that switching was difficult, deeper analyses showed that many
of them took their business elsewhere each year. CVS had 29.5 million pharmacy members at the
start of 2000, a year in which total revenue for the corporation year was $20 billion. PSI team analyses
indicated that approximately 7.2 million regular pharmacy customers left CVS during the year.2 The
total volume of filled prescriptions grew during 2000 because the company also attracted 8.5 million
new regular members over the course of the year, but the PSI team’s work clearly highlighted that
customer defections were hampering growth. The regular customers who left in 2000 took with them
an estimated 55 million annual prescriptions that, had they been filled by CVS, would have
contributed $2.5 billion to revenue.
Early interviews and analysis also revealed that different kinds of customers left for different
reasons. The PSI team divided regular CVS pharmacy members into two categories. Light users,
who filled an average of five scripts per year, were most likely to defect because of the pharmacy’s
location (see Figure A). Heavy users filled an average of 40 scripts a year and were most likely to
leave because of poor service. According to Grossi, “We thought that a better fulfillment process in
the pharmacies could prevent 60%–90% of the customer defections that were due to service. The PSI
team had a pretty big opportunity.”
Figure A Reasons Given by Former CVS Pharmacy Customers for Switching to Another Pharmacy
Light Users
0
13
18
69
0 10 20 30 40 50 60 70
Insurance
Service
Price/Mail
Location
% of Customers
Reason for Switching
Heavy Users
6
44
33
17
0 10 20 30 40 50 60 70
Insurance
Service
Price/Mail
Location
% of Customers
Reason for Switching
Source: CVS.
2 In addition to these regular customers, an estimated 10.9 million infrequent customers left in 2000. Because infrequent
customers contributed so little to total volume of prescriptions filled by CVS, the PSI team did not focus on them or include
them in analyses.
606-015 Pharmacy Service Improvement at CVS (A)
4
Field Work
PSI team members spent time in many CVS pharmacies, systematically observing how
prescriptions were filled or not filled. In addition to the comprehensive list of problems (Exhibit 1),
they gathered other evidence that things were not working well. Approximately one in four scripts
experienced a problem at some point in the fulfillment process, and 16% of all scripts had problems
that were still unresolved at customer pickup. This not only slowed down pickup for other
customers but also made working at the pickup station a stressful and unpleasant job. During a
single eight-hour shift observed by a PSI team member, 40% of customers voiced a complaint. The
tech was asked 10 questions that he was not qualified to answer and was verbally abused four times.
When asked, he said that he felt he was responsible for none of the problems encountered by
customers and could have done nothing to prevent them (Exhibit 2). As Betses explained:
The people working at pickup are our lowest paid, least trained people, but we were asking
them to do something that’s both no fun and super difficult—dealing with angry customers all
day. No wonder lots of them left after less than a year on the job! All of us on the PSI gained a
real appreciation for how hard it was to work effectively in our pharmacies. We saw that in
the few that were working well, people had either developed elaborate workarounds or were
making heroic efforts or both.
The Pharmacy Fulfillment Process
The PSI team found that virtually all CVS pharmacies followed the same multistep process to fill
prescriptions and experienced the same exceptions to it. The process consisted of five basic steps,
diagrammed below in Figure B.
Drop-off
When a customer dropped off a script, a tech asked when they would return to pick it up. The
tech wrote the requested pickup time on the script itself, then put it in a box that was divided into a
number of slots. Each slot was assigned to a specific time period—2 p.m., 3 p.m., 4 p.m., and so on.
The tech put the script into the slot corresponding to the hour before the desired pickup time. If the
customer wanted the prescription filled immediately, the tech put the script in the slot corresponding
to the current time.
Although customers dropped off their prescriptions throughout the day, the busiest times at the
drop-off window were before work, lunchtime, and after work. Regardless of when they dropped
them off, more customers wanted to pick up their filled prescriptions after work than at any other
time.
Data Entry
Each hour, a tech took that hour’s scripts from the box and entered all required data about them
into the pharmacy information system, an application used by all locations and connected to CVS’s
central databases of drug, prescription, customer, payment, and insurance information. Required for
each prescription were patient and doctor contact information, data about any third-party payors
such as insurance companies or employers, and the specifics of the prescription itself: medication,
dosage, number of doses, and so on.
Pharmacy Service Improvement at CVS (A) 606-015
5
Figure B Basic Flow for CVS Prescription Fulfillment Process
Quality
Assurance
I
N
V
E
N
T
O
R
Y
INVENTORY
Drop-off Data
Entry Production Quality
Assurance Pick-up
Standard
script path
Data entry
RPh
RPh
Shelves
Dr. call or
Production
Consultarea
Drop-off
Pick-up
Quality
Assurance
I
N
V
E
N
T
O
R
Y
INVENTORY
Drop-off Data
Entry Production Quality
Assurance Pick-up
Standard
script path
Data entry
RPh
RPh
Shelves
Dr. call or
Production
Consultarea
Drop-off
Pick-up
Source: CVS.
Drug utilization review As soon as data entry was complete, the system performed an
automated “drug utilization review” (DUR). The DUR checked the script against all other
prescriptions in the database for that patient (in other words, all prescription drugs that had ever
been dispensed by CVS to the patient) to see if there existed any possibility for harmful drug-drug
interactions. The DUR also checked to make sure the drug was appropriate for the patient, given the
patient’s age, gender, and other demographic data stored in the system.3
If the DUR revealed any potential problems, the systems came to a “hard stop” and fulfillment
could not proceed until the DUR was reviewed by a pharmacist. In the great majority of cases the
pharmacist did not need to involve the customer when reviewing the DUR. In fact, many within the
industry considered it better for the customer not to be involved, reasoning that if the DUR gave the
impression that a prescribed drug could be harmful, the customer might be less likely to take it.
Everyone at CVS felt that the DUR was an essential part of good pharmacy operations and
customer service and that the automated review should be a very careful and conservative one.
Insurance check After the DUR was complete and any hard stops were reviewed, the system
performed an insurance check. Most CVS pharmacy customers had their prescriptions paid for by a
third party such as an employer, an insurance company, or a government agency. These customers
3 CVS maintained a separate application that allowed customers to request refills via telephone. This system stored refill
requests until 1.5 hours before the requested pickup time, then transferred them to the pharmacy system for fulfillment,
beginning with the DUR.
606-015 Pharmacy Service Improvement at CVS (A)
6
paid only a small amount of their own money, called a “copayment,” when they picked up their
medicine.4 Payors had complicated rules about the drugs they would cover and the conditions under
which they would pay for them. The insurance check verified that a script followed all of these rules.
As Flum explained:
One of the biggest changes in our industry is the fact that in recent years more and more
pharmacy customers have third parties that help pay for prescriptions—over 90% of our
customers now. Payors have been putting in place more and more complicated formularies5 to
try to control their costs. This complicates our work a lot.
Say a doctor prescribes a drug that’s not on a patient’s formulary, which happens all the
time because doctors and patients don’t usually have formularies at their fingertips. Our
insurance check is the first time anyone learns that there’s a problem. We would then need to
work with the doctor, the patient, and the payor to switch the prescription. Payors have also
tightened rules about when they’ll allow a prescription to be refilled, so patients basically have
to wait longer before coming in for a refill. If they don’t wait long enough the payor will refuse
to cover the fill. This type of insurance rejection is called “refill too soon,” and we’ve been
seeing more and more of them.
In most cases the fulfillment process would continue even if one of these rules was violated; CVS
pharmacy employees would attempt to identify and correct the problem while the process continued
or when the customer came to pick up their prescription.
Production
The drugs to fill the script were counted and verified by certified pharmacy technicians in the
production area, which was near the shelves where medicine was stored.
Quality Assurance
After production, a pharmacist reviewed each script to make sure that it contained exactly the
right drugs in the right quantities and that all other details were correct. Quality assurance (QA) was
one of a pharmacist’s most important tasks and was never delegated to a technician or other
employee in the pharmacy.
The steps from data entry to QA could be completed in approximately five minutes if there were
no problems.
Pickup
After QA, each completed script was sealed in a bag. Bags were stored in the pickup area in
alphabetical order. When customers arrived to pick up their prescriptions, the technician staffing the
pickup window searched for the right prescription among the bags, verified customers’ identities,
and took any required payments from them.
4 Copayments were typically between $5 and $20, which was a small fraction of the cost of most nongeneric pharmaceuticals.
5 A formulary is a set of rules governing the medicines a third party will pay for and the circumstances under which they will
pay. A formulary might state, for example, that a third party will only pay for a generic version of a certain antibiotic and will
only pay for 30 doses a month. Formularies were so complicated that many payors worked with separate companies called
pharmacy benefit managers (PBMs) to define, update, and enforce them.
Pharmacy Service Improvement at CVS (A) 606-015
7
Problems during the Process
Pickup window technicians also dealt with customers who did not get what they were expecting.
Based on their analyses and observations, the PSI team estimated that 16% of customers fell into this
category. The team was even more disturbed to find that 27% of scripts encountered a substantial
problem at some point in the fulfillment process.
Drop-off
The only substantial problem that arose at this step, the PSI team found, was an unmanned drop-
off window. As Grossi explained, issues were not common at this stage because “nothing happened
at drop-off. The customer just handed over a script and walked away while the tech filed it in the
box according to pickup time.”
Data Entry
When the tech took scripts from the box and entered their details into the system, a number of
problems could occur.
No refill allowed Many scripts allowed the customer to refill the prescription at least once.
Customers could lose track of how many refills were allowed, however, and drop off an ineligible
script. When this occurred the system printed a label for the ineligible script, which was put in a “Dr.
call bin.” A tech would periodically take the contents of this bin and make phone calls or send faxes
to doctors’ offices asking for their approval to refill the prescription. If the tech reached the doctor
immediately and the doctor approved the refill, the script proceeded to the next step in the process.
If the doctor rejected the refill, the label was put in a “Dr. denied” box near the pickup area;
customers learned about refill denials when they returned to pick up their prescriptions.
If the tech could not reach the doctor immediately, the label was put in a “Dr. call-back box.”
Problems stemming from “no refill allowed” scripts required from 20 minutes to three days to
resolve, with an average resolution time of one day. “No refill allowed” scripts were 6% of total
scripts.
DUR hard stop The DUR generated a hard stop for 20% of all scripts. Over 90% of hard stops
were resolved by pharmacists without involving the prescribing doctor. As Betses explained:
The system checks each script against all others for that patient dispensed over the last 12
months. So the DUR for script A could generate a hard stop because of the possibility of a
drug-drug interaction with script B, which was a 10-day course of antibiotics prescribed eight
months ago. Pharmacists would clear that kind of hard stop after a careful review. They
would clear others after calling up the patient to determine, say, that their weight was
appropriate for the dosage prescribed. In both of these cases the system is working as planned;
we want hard stops every time there’s even a small chance of harm, and we want the
pharmacist to take action on them quickly. In a few cases, though, there is a serious potential
problem with the script as written. The DUR generates a hard stop, and the pharmacist needs
to call the doctor to resolve the potential problem.
Insurance check Seventeen percent of all scripts encountered a problem during the automated
insurance check. The majority of these problems were easy to resolve; they were due to date-of-birth
errors on the script or to a customer’s having changed jobs or insurers. Some errors of this type could
be resolved by the data-entry technician alone; others required a phone call to the customer. Other
606-015 Pharmacy Service Improvement at CVS (A)
8
insurance problems were harder to resolve and required a phone call to the insurer and/or the
prescribing doctor. Scripts were filled even if insurance problems were not resolved. When this was
the case, the customer was asked to pay the full amount of the prescription at pickup.
Production and Quality Assurance
The only problem identified at the production step was insufficient inventory to completely fill
the script. Seven percent of scripts encountered partial or complete stock shortages of the required
medicine.
The PSI team did not find any issues with quality assurance as practiced at CVS. Pharmacists
diligently and completely reviewed each filled script and made sure that the drugs dispensed were
actually the ones prescribed.
Pickup
Team members documented a variety of issues at the pickup window. The most common were
unpleasant customer surprises: unauthorized refills, scripts that had not been paid for by insurance,
or scripts that were simply not ready yet. Some of these issues prevented fulfillment, causing
customers to walk away from the pickup window without medicine and with a bad impression of
CVS customer service. Even when problems could be fixed, the resolution process took a long time
and increased wait time for other customers in line. The situation at the pickup window was worst
between 5 p.m. and 7 p.m., when customers came after work to pick up the prescriptions they had
dropped off or called in earlier. Most CVS locations found it difficult to staff this time period simply
because pharmacy employees did not want to work then. As one tech said to the PSI team, “I hate
the late afternoon shift. You spend all your time dealing with angry people, and you can’t do
anything to make things better for them.”
Flum commented: “Pickup is where customers wait in line, get bad news, get mad, and yell at the
poor tech, but that doesn’t mean that we need to fix pickup. It means that we need to fix whatever’s
causing pickup not to have the completed script with the right copayment amount ready when the
customer walks up to the counter.”
Conclusion
The PSI team felt that they had a great deal of freedom to change pharmacy fulfillment operations.
Their work was sponsored and supported by senior management, and CEO Tom Ryan had stated
that pharmacy service improvement was the most important corporate initiative for the coming year.
Team members therefore knew that their recommended changes to tasks, responsibilities, and
processes would carry much weight. They also knew that they could get information systems
changed, if necessary; pharmacy IT at CVS was part of the operations function, which had sponsored
the PSI.
Team members also realized, however, that any changes they made could not compromise
customer safety. Even changes that appeared to do so would be difficult to sell to the organization.
As Roberts started to sketch a new fulfillment process on the whiteboad, Flum, Betses, and Grossi
wondered exactly what it would look like and how it would be accepted by CVS and its pharmacies.
60
6-
01
5
-9
–
Ex
hi
bi
t 1
C
V
S
Ph
ar
m
ac
y
Fu
lfi
llm
en
t P
ro
bl
em
s
N
ot
ic
ed
b
y
PS
I T
ea
m
•
G
ro
up
o
f s
cr
ip
ts
a
rr
iv
es
se
pa
ra
te
ly
•
T
ec
h
ca
nn
ot
d
et
er
m
in
e
w
he
th
er
IV
R
s
cr
ip
ts
a
re
pa
rt
o
f g
ro
up
•
F
ai
lu
re
to
p
ro
pe
rly
pr
io
rit
iz
e
la
be
l p
rin
t
•
O
O
S
is
su
e
fir
st
de
te
rm
in
ed
•
C
us
to
m
er
n
ot
c
al
le
d
ab
ou
t O
O
S
/p
ar
tia
l
•
O
O
S
n
ot
o
rd
er
ed
fr
om
O
V
•
B
as
ke
ts
d
el
iv
er
ed
to
Q
A
o
ut
o
f p
rio
rit
y
•
R
x
fil
le
d
in
co
rr
ec
tly
•
P
ill
im
ag
e
no
t i
n
sy
st
em
•
H
ar
d
co
py
im
ag
e
no
t i
n
sy
st
em
/il
le
gi
bl
e
•
R
P
h
bu
sy
a
t o
th
er
st
at
io
ns
, f
al
ls
b
eh
in
d
at
Q
A
•
R
P
h
fa
ils
to
a
nn
ou
nc
e
P
4
an
d
P
3
sc
rip
ts
•
C
om
pl
et
ed
o
rd
er
m
is
fil
ed
•
S
cr
ip
t g
ro
up
s
no
t
m
ai
nt
ai
ne
d
•
G
re
en
z
on
es
n
ot
cl
ea
ne
d
re
gu
la
rly
•
N
ob
od
y
w
at
ch
in
g
in
-s
to
re
dr
op
-o
ff
w
in
do
w
•
E
xt
en
de
d
w
ai
t b
ef
or
e
in
iti
al
gr
ee
tin
g
in
s
to
re
•
N
ob
od
y
re
sp
on
ds
to
d
riv
e-
th
ru
•
T
ec
h
m
us
t c
ho
os
e
be
tw
ee
n
se
rv
in
g
cu
st
om
er
in
s
to
re
o
r
se
rv
in
g
dr
iv
e-
th
ru
•
D
riv
e-
th
ru
c
lo
se
d
•
Lo
ng
w
ai
t b
eh
in
d
so
m
eo
ne
at
d
riv
e-
th
ru
•
C
us
to
m
er
c
al
lin
g
in
r
ef
ill
le
ft
on
h
ol
d
•
P
ro
vi
de
r
le
ft
on
h
ol
d
•
P
ro
vi
de
r
vo
ic
em
ai
l n
ot
re
gu
la
rly
c
he
ck
ed
•
F
ax
m
ac
hi
ne
n
ot
r
eg
ul
ar
ly
ch
ec
ke
d
•
C
us
to
m
er
c
an
no
t l
oc
at
e
dr
op
-o
ff
•
In
iti
al
in
te
rv
ie
w
fa
ils
to
o
bt
ai
n
cr
iti
ca
l d
at
a
•
F
ai
lu
re
to
p
ro
pe
rly
p
rio
rit
iz
e
sc
rip
ts
•
W
ai
t t
im
es
m
ad
e
up
/o
ve
r-
pr
om
is
ed
•
O
O
S
n
ot
c
he
ck
ed
•
C
us
to
m
er
c
an
no
t v
er
ify
de
m
og
ra
ph
ic
s
•
S
cr
ip
t m
is
fil
ed
in
ti
m
e
sl
ot
or
ga
ni
ze
r
•
C
us
to
m
er
w
ith
n
o
re
fil
ls
n
ot
in
fo
rm
ed
p
ro
ce
ss
m
ay
ta
ke
up
to
th
re
e
da
ys
•
In
co
m
pl
et
e
in
fo
rm
at
io
n
ta
ke
n
fr
om
p
ro
vi
de
r
–
C
us
to
m
er
la
te
r
ca
nn
ot
be
fo
un
d
in
s
ys
te
m
•
P
ro
vi
de
r
ca
lls
/fa
xe
s
no
t
pr
io
rit
iz
ed
a
s
P
2
•
D
U
R
s
ar
is
e
fre
qu
en
tly
a
nd
R
P
h
m
us
t i
nt
er
ve
ne
•
N
o
cl
ea
r
pr
ot
oc
ol
fo
r
re
so
lv
in
g
th
ird
-p
ar
ty
p
ro
bl
em
s
•
R
es
ol
va
bl
e
th
ird
-p
ar
ty
is
su
es
fi
lle
d
fo
r
ca
sh
•
S
te
ps
ta
ke
n
to
r
es
ol
ve
th
ird
-p
ar
ty
is
su
es
n
ot
r
ec
or
de
d
w
ith
s
cr
ip
t
•
90
-m
in
ut
e
le
ad
ti
m
e
on
IV
R
d
oe
s
no
t p
ro
vi
de
a
de
qu
at
e
tim
e
to
re
so
lv
e
th
ird
-p
ar
ty
is
su
es
•
C
us
to
m
er
n
ot
n
ot
ifi
ed
o
f t
hi
rd
-p
ar
ty
is
su
es
•
In
-s
to
re
p
ro
ce
ss
fo
r
ha
nd
lin
g
re
fil
l
au
th
or
iz
at
io
ns
n
ot
o
pt
im
iz
ed
•
N
o
st
an
da
rd
fo
rm
fo
r
fa
xi
ng
pr
ov
id
er
s
•
F
ax
es
/c
al
ls
to
p
ro
vi
de
rs
n
ot
d
on
e
in
tim
el
y
m
an
ne
r
–
Le
ft
in
d
oc
to
r
ca
ll
bo
x
•
“M
D
w
ill
c
al
l b
ac
k”
b
ox
n
ot
re
vi
ew
ed
r
eg
ul
ar
ly
•
P
at
ie
nt
n
ot
n
ot
ifi
ed
o
f n
ee
d
fo
r
re
fil
l
au
th
or
iz
at
io
n
•
IV
R
h
ol
ds
s
cr
ip
ts
th
at
n
ee
d
re
fil
l
au
th
or
iz
at
io
n
in
“
D
”
qu
eu
e
•
IV
R
“
D
”
qu
eu
e
no
t r
ou
tin
el
y
ch
ec
ke
d
•
IV
R
in
fo
rm
s
pa
tie
nt
th
at
s
cr
ip
t
w
ith
ou
t r
ef
ill
a
ut
ho
riz
at
io
n
ca
n
be
re
ad
y
af
te
r
fiv
e
ho
ur
s
•
P
er
so
n
at
D
E
c
an
no
t r
ea
d
pr
ov
id
er
ha
nd
w
rit
in
g
•
P
er
so
n
at
D
E
c
an
no
t r
ea
d
te
ch
ha
nd
w
rit
in
g
•
P
re
sc
rib
ed
m
ed
ic
at
io
n
no
lo
ng
er
m
an
uf
ac
tu
re
d/
in
co
rr
ec
t d
os
ag
e
O
O
S
:
O
ut
o
f s
to
ck
D
U
R
:
D
ru
g
ut
ili
za
tio
n
re
vi
ew
R
P
h:
P
ha
rm
ac
is
t
IV
R
:
In
te
gr
at
ed
v
oi
ce
r
es
po
ns
e
(a
ut
om
at
ed
s
ys
te
m
a
llo
w
in
g
cu
st
om
er
s
to
p
ho
ne
in
r
ef
ill
s)
M
D
:
P
hy
si
ci
an
D
E
:
D
at
a
en
tr
y
R
x:
P
re
sc
rip
tio
n
P
S
A
:
P
ha
rm
ac
y
se
rv
ic
e
as
so
ci
at
e
•
T
ec
h/
P
S
A
c
an
no
t f
in
d
sc
rip
t
•
T
ec
h/
P
S
A
fi
nd
s
pa
rt
o
f
gr
ou
p
of
s
cr
ip
ts
•
T
ec
h/
P
S
A
d
oe
s
no
t
fo
llo
w
a
ny
p
ro
ce
ss
to
fin
d
sc
rip
t
•
T
ec
h
ca
nn
ot
ad
eq
ua
te
ly
e
xp
la
in
in
su
ra
nc
e
is
su
e
•
T
ec
h
no
t s
ur
e
if
do
ct
or
ca
lle
d
ab
ou
t r
ef
ill
•
R
ud
e
st
af
f
•
R
P
h
m
ak
es
n
o
ef
fo
rt
to
he
lp
r
es
ol
ve
p
ro
bl
em
s
•
Lo
ng
li
ne
s
at
p
ic
ku
p
•
N
o
ad
di
tio
na
l r
eg
is
te
rs
op
en
ed
to
e
as
e
tr
af
fic
•
N
ob
od
y
re
sp
on
ds
to
dr
iv
e-
th
ru
•
N
o
ca
sh
r
eg
is
te
r
at
dr
iv
e-
th
ru
•
T
ec
h/
P
S
A
fo
rc
ed
to
ch
oo
se
b
et
w
ee
n
lin
e
at
pi
ck
up
a
nd
d
riv
e-
th
ru
•
D
riv
e-
th
ru
c
lo
se
d
•
Lo
ng
w
ai
t b
eh
in
d
so
m
eo
ne
a
t d
riv
e-
th
ru
So
ur
ce
:
C
V
S.
D
ro
p-
of
f
D
at
a
En
tr
y
Pr
od
uc
tio
n
Q
A
W
ai
tin
g
B
in
s
Pi
ck
up
60
6-
01
5
-1
0-
Ex
hi
bi
t 2
Is
su
es
F
ac
ed
b
y
Te
ch
ni
ci
an
S
ta
ff
in
g
C
V
S
Pr
es
cr
ip
tio
n
Pi
ck
up
W
in
do
w
, a
s
N
ot
ic
ed
b
y
PS
I T
ea
m
M
em
be
rs
S
om
e
th
in
gs
te
ch
s
he
ar
in
a
ty
pi
ca
l d
ay
In
o
ne
e
ig
ht
-h
ou
r
sh
ift
a
t t
he
r
eg
is
te
r
“W
hy
t
he
h
el
l
is
t
hi
s
fo
ur
h
un
dr
ed
d
ol
la
rs
?
I
a
lw
ay
s
pa
y
fif
te
en
fo
r
an
y
dr
ug
s!
”
“I
’m
n
ot
m
ov
in
g
fr
om
th
is
s
po
t u
nt
il
w
e
ge
t t
hi
s
cl
ea
re
d
up
.
I
am
N
O
T
p
ay
in
g
th
at
m
uc
h!
T
hi
s
m
us
t b
e
a
jo
ke
!”
“Y
ou
lo
st
m
y
pr
es
cr
ip
tio
n!
I
’m
ti
re
d
of
th
is
a
ll
th
e
tim
e!
W
hy
ca
n’
t y
ou
e
ve
r
do
a
ny
th
in
g
rig
ht
?”
“W
he
re
a
re
m
y
ot
he
r
sc
rip
ts
?”
“I
’m
n
ot
s
ur
e
ho
w
m
an
y
sc
rip
ts
I
ha
ve
.”
“W
ha
t
do
y
ou
m
ea
n
yo
u
ca
n’
t
fin
d
m
y
or
de
r,
I
c
al
le
d
it
in
ye
st
er
da
y.
”
“I
t
al
ke
d
to
s
om
eo
ne
f
ou
r
da
ys
a
go
t
o
m
ak
e
su
re
m
y
or
de
r
w
ou
ld
b
e
re
ad
y,
a
nd
y
ou
’re
te
lli
ng
m
e
it’
s
no
t d
on
e?
!”
80
—
T
ot
al
n
um
be
r
of
c
us
to
m
er
s
de
al
t w
ith
32
—
T
ot
al
c
us
to
m
er
s
w
ith
p
ro
bl
em
s
•
21
—
C
us
to
m
er
s
w
ho
w
ai
te
d
lo
ng
er
th
an
th
ey
th
ou
gh
t
th
ey
s
ho
ul
d
ha
ve
•
16
—
C
us
to
m
er
s
w
ith
o
rd
er
s
no
t r
ea
dy
/c
om
pl
et
ea
•
9
—
C
us
to
m
er
s
pa
yi
ng
m
or
e
th
an
e
xp
ec
te
db
•
4
—
C
us
to
m
er
s
co
m
pl
ai
ni
ng
a
bo
ut
fr
on
t s
to
re
-r
el
at
ed
is
su
es
(
i.e
.,
co
up
on
s,
s
al
es
, s
to
ck
)
10
—
N
um
be
r
of
ti
m
es
te
ch
w
as
a
sk
ed
a
q
ue
st
io
n
th
ey
w
er
e
no
t q
ua
lif
ie
d
to
a
ns
w
er
4
—
N
um
be
r
of
ti
m
es
te
ch
w
as
v
er
ba
lly
a
bu
se
d
by
th
e
cu
st
om
er
0
—
N
um
be
r
of
p
ro
bl
em
s
te
ch
fe
el
s
th
ey
w
er
e
re
sp
on
si
bl
e
fo
r
So
ur
ce
:
C
V
S.
a In
cl
ud
es
o
ut
-o
f-
st
oc
k,
p
ar
tia
l f
ill
s
w
he
re
c
us
to
m
er
is
n
ot
p
re
vi
ou
sl
y
co
nt
ac
te
d,
r
ef
ill
a
ut
ho
ri
za
tio
n
re
qu
ir
ed
, t
hi
rd
p
ar
ty
.
b In
cl
ud
es
th
ir
d-
pa
rt
y
re
je
ct
s
fil
le
d
fo
r
ca
sh
a
nd
c
us
to
m
er
m
is
un
de
rs
ta
nd
in
gs
o
ve
r
po
lic
y
co
ve
ra
ge
.
Pharmacy Service Improvement at CVS (A) 606-015
11
Exhibit 3 CVS Historical Financials (millions of dollars)
Year 2002 2001 2000 1999 1998
Net Operating Revenues 24181.5 22241.4 20087.5 18098.3 15273.6
Cost of Goods Sold 18112.7 16544.7 14725.8 13236.9 11134.4
Gross Margin 6068.8 5696.7 5361.7 4861.4 4139.2
Operating Expenses & D, D. & A 4862.6 4577.1 4058.2 3725.9 3198.7
Operating Profits 1206.2 1119.6 1303.5 1135.5 940.5
Non-Operating Expenses 50.4 -288.0 98.5 59.1 -127.7
Pre-Tax Income 1155.8 1407.6 1205.0 1076.4 1068.2
Income Tax 439.2 296.4 497.4 441.3 306.5
Minority Interest 0.0 0.0 0.0 0.0 0.0
Net Income 716.6 413.2 746.0 635.1 384.5
Net Margin 2.96% 1.86% 3.71% 3.51% 2.52%
Inventories 4013.9 3918.6 3557.6 3445.5 3190.2
Accounts Receivable 1019.3 966.2 824.5 699.3 650.3
Cash and Cash Equivalents 700.4 236.3 337.3 230 180.8
Other Current Assets 248.5 333 217.2 233.2 327.9
Total Current Assets 5982.1 5454.1 4936.6 4608.0 4349.2
Property, Plan and Equipment 2215.8 1847.3 1742.1 1601 1351.2
Other Non Current Assets 1447.4 1326.8 1270.8 1066.4 985.8
Total Assets 9645.3 8628.2 7949.5 7275.4 6686.2
Asset Turnover 2.51 2.58 2.53 2.49 2.28
ROA 7.43% 4.79% 9.38% 8.73% 5.75%
Accounts Payable 1707.9 1535.8 1351.5 1454.2 1286.3
Other Current Liabilities 1398.0 1530.1 1612.6 1435.7 1847.0
Total Current Liabilities 3105.9 3065.9 2964.1 2889.9 3133.3
Non Current Liabilities 1342.4 995.4 680.8 705.8 442.3
Total Liabilities 4448.3 4061.3 3644.9 3595.7 3575.6
Equity 5197.0 4566.9 4304.6 3679.7 3110.6
Total Liabilities & Equities 9645.3 8628.2 7949.5 7275.4 6686.2
Levarage (Equity/Total Assets) 0.54 0.53 0.54 0.51 0.47
ROE 13.79% 9.05% 17.33% 17.26% 12.36%
Source: Standard & Poor’s Research Insight.
Note: Fiscal Year ends December 31.
CVS: Process
Design Principles
TC1 TEAM
Summary
Combine several tasks into one
Enable workers to make decisions
Arrange the process steps in a natural order
Create multiple versions of processes
Perform the work where it makes sense
Checks and controls are reduced
Reconciliation steps are minimized
Chris – Combine several tasks into one
The CVS pharmacy fulfillment process does a good job in combining several tasks
into one, but at least one place could benefit from some adjustment:
● Combining DUR and Insurance checks to run at the same time would save
time. These are currently ordered, with “stop” processes in place. However
with the rate of resolution being high in both cases and knowing that the
fulfillment process continues, even if a rule above was violated CVS should
eliminate the wait. The suggestion is to combine and make this one step.
Angela – Enable workers to make decisions
The CVS pharmacy fulfillment process does a good job of enabling workiner to
make decisions:
● Due to the compliance and safety nature of pharmacies, CVS does a good
job utilizing the pharmacy techs when they can so that the pharmacist can
attend to the tasks that they alone need to address such as DUR & insurance
check exceptions.
Chris – Arrange the process steps in a natural order
The CVS pharmacy fulfillment process is rated good follows a natural order but
could be improved in a few areas.
● Technicians should not wait until the hour to start entering information. This
could be a bottleneck in the process if the orders became too many or there
were unexpected issues that arose. The orders should be done in order of
pickup, but waiting, rather than just doing, seems an unnecessary potential
problem.
● Currently the DUR and the Insurance check occur in order and have a “stop”
point in the process. However the rate of resolution is high in both cases and
the fulfillment process continues even if a rule above was violated CVS
should eliminate the wait. The suggestion to combine and make this one
step.
Ibrahim – Create multiple versions of processes
The CVS pharmacy fulfillment process does a fair job of creating multiple versions
of processes:
● There are two versions pickup options; immediate or delayed. In order to
provide better customer service they have divided these processes.
Immediate pickup is processed right away and delayed pickups are placed in
a file folder system by pick up time of day.
Angela – Perform the work where it makes sense
The CVS pharmacy fulfillment process does a fair job of performing the work
where it makes sense:
● CVS’ pharmacies are typically limited space but have specific areas for
certain activities like customer pickups and dropoffs.
● Insurance issues that come back requiring a doctor intervention should go to
the doctor before coming back to the pharmacy, the pharmacy should be
notified so they can tell the patient there might be a delay.
● Inform patients of issues with the prescription before they come to pick it up.
Gloria – Checks and controls are reduced
The CVS pharmacy fulfillment process does a fair job of reducing checks and
controls:
● The Quality Assurance Check at CVS is one of the most important ones
where the pharmacist reviews each script to ensure it contains the right drugs
and the right quantities. However, this is part of the process and it is not a
meaningful check that is rare and significant. The pharmacist is performing
the checks at one stage of the process rather than rare meaningful checks
throughout the process.
Gloria – Reconciliation steps are minimized
The CVS pharmacy fulfillment process does a poor job of minimizing reconciliation
steps:
● For the case of No refill allowed, there is not a systematic accurate method
to track refills which creates the need for manual reconciliation to call the
doctor’s office to get their approval to refill. This creates the issue of duplicate
information across different touchpoints. There is not a notification system for
the patient to know if their prescription was approved for refill or not.
Now what?
We believe that at this point in time the best route is an incremental approach to
the items identified above:
● Combing the DUR and Insurance checks to run synchronously
● Improve customers notifications on issues and complications
● Add a drop box for prescriptions to be dropped off unattended in addition to
in-person
