Body Scans and Bottlenecks:: Optimizing Hospital CT Process Flows
Body Scans and Bottlenecks:: Optimizing Hospital CT Process Flows
KEL592
It was November 2005, and Dr. Steve Foster, head of radiology at a large, multispecialty
hospital in the Midwestern United States, faced a challenge. His hospital had just purchased three
new computed tomography (CT) scanner units and was preparing to install them. This was good
news—each new scanner could provide high-resolution images in less than seven minutes,
whereas the hospital’s six older CT scanners required more than half an hour to scan each
patient—but the new units’ arrival also meant that Foster and his team had to decide how to
maximize utilization of the hospital’s increased scanning capacity. Foster sat down and began to
review the existing process flow and plan the changes he would need to make.
Processing even the small number of images acquired by the first CT scanner took two and a
half hours. Improvements in scanner technology occurred slowly until 2000, when multidetector
CT scanners were developed; thereafter, the technology advanced quickly as manufacturers
shifted from single-detector designs to configurations using 64 to 320 detectors, which resulted in
shorter scans and better image quality. Improvements in computer processing dramatically
reduced the time required to generate the reconstructed images, which numbered in the
thousands. Similar to processing power in the computer industry, CT technology continued to
advance in accordance with Moore’s Law, representing a doubling of capability about every 18
months.1
Intravenously administered contrast agents were used to enhance image quality and better
highlight the blood vessels in the body. CT scans were useful for diagnosis of medical diseases
1
Moore’s Law originally applied to the number of transistors that could be placed cost effectively on an integrated circuit. See Gordon
E. Moore, “Cramming More Components Onto Integrated Circuits,” Electronics 38, no. 8 (1965): 114.
©2011 by the Kellogg School of Management at Northwestern University. This case was prepared by Sachin Waikar under the
supervision of Professor Sunil Chopra and Dr. Scott Flamm. 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. To order copies or
request permission to reproduce materials, call 800-545-7685 (or 617-783-7600 outside the United States or Canada) or e-mail
custserv@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
the Kellogg School of Management.
This document is authorized for use only by Anthony Huaman in IN2006 Fundamentos de Operaciones 2024-1 - Prof. Óscar Gamonal taught by Óscar Gamonal, Universidad de Ingenieria y
Tecnologia UTEC from Apr 2024 to Oct 2024.
For the exclusive use of A. Huaman, 2024.
BODY SCANS AND BOTTLENECKS KEL592
throughout the body; the high speed of the scanners enabled them to image even moving organs
such as the heart.
The components of the process included patient preparation, the actual scan, and image
reconstruction. One nurse and one CT technologist were required to carry out the tasks for each
scan. The room that housed the CT scanner was a 45-second walk from the patient waiting area
(Exhibit 2).
1. The nurse first went and located the patient in the waiting area (2 minutes).
2. The nurse then moved the patient from the waiting area into the CT scanning room (2
minutes).
3. The nurse placed the IV line for contrast injection (2 minutes). At the same time, the CT
technologist selected the CT scan protocol and prepared the machine (3 minutes).
4. The technologist performed the CT scan (17 minutes).
5. While the technologist was monitoring the CT image reconstruction (10 minutes), the
nurse assisted the patient off the table (1 minute), returned the patient to the changing
area (2 minutes), changed the linens and cleaned the CT scan room (2 minutes), refilled
the contrast injector (3 minutes), and got the next patient from the waiting area (2
minutes).
6. The technologist coded and distributed the CT image (2 minutes).
From that point on, the process repeated for each new patient.
As Foster and his team awaited the arrival of the new CT scanners, they realized that
installing them in the same configuration as the older scanners and conducting business as usual
would not maximize their value.
The new machines not only required less time for scans (2.5 minutes versus 17 minutes for
the old machines), but the duration of the technologist’s tasks was also reduced: selecting and
preparing the CT scan protocol took 1 minute (versus 3 minutes), CT image reconstruction took 2
minutes (versus 10 minutes), and CT image coding and distribution required only 1 minute
(versus 2 minutes).
Given the dramatically shortened scanning time, the CT scanner had shifted from being the
bottleneck in the process to a much less time-consuming component. That meant that the team
had to consider reconfiguring the scanning process to maximize utilization of the new units.
But where to start? The existing CT process—patient preparation, scanning protocol, image
reconstruction, nurse and technologist tasks, and even the configuration of the scanning area—
had been designed around the long scanning time of the original units. The much shorter scanning
time of the new CT units meant Foster’s team had to answer questions such as: Where could time
be saved in non-scanning parts of the process? What changes in the physical layout would be
necessary? What would be the most efficient use of the nurses and technologists? (See Exhibit 3
for hourly personnel costs.) Would it be possible to make previously sequential tasks parallel, or
to centralize certain tasks?
Foster also wondered about the implications for CT scanning for the healthcare system at
large. If future technological advances continued to yield scanners with greater speed and image
quality, would hospitals have more scanning capacity than needed? (See Exhibit 4 for the growth
in number of scans performed in the United States between 1998 and 2007.) What would happen
when scanning capacity (or supply) eventually outstripped demand? Would it still make sense for
every hospital of a certain size to have its own CT scanners?
Foster had authorization from hospital management to make any changes he felt necessary.
He knew he and his team needed to think logically and creatively in order to make the most of
this opportunity.
CT Image, 2008
Expenses
Technologists: $35/hour
Nurses: $50/hour
Supplies: $50/scan
80
70
60
50
Millions
40
30
20
10
0
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Year