267
Chapter 32
Industrial Engineers
in Public Health
Michael L. Washington
Introduction
Researchers note that the public health system needs improvements. More specically, Mays et al.
(2004) state that the public health system needs “better information on how to organize, nance,
and deliver public health services to achieve improvements in the population health” (p 183). ey
also state that one key element in improving the public health system is to increase the pipeline
of public health systems researchers and that these researchers must be able to develop analytical
methods and tools to help answer questions of policy and practice.
e National Academy of Engineering Committee on Engineering and the Health Care
System (2005) issued a report titled Building a Better Delivery System: A New Engineering/Health
Care Partnership that identies industrial engineering as a discipline that improves health ser-
vices, in that a better relationship needs to be developed between healthcare and engineering to
improve healthcare systems. For decades, engineers have been using analytical tools to improve
many sectors all over the world. One way to improve the public healthcare system would be
to merge the research tools of industrial engineering with the knowledge of epidemiology to
improve policy decisions, logistics, and delivery capabilities. While generally unknown, this idea
Contents
Introduction .............................................................................................................................267
History of Industrial Engineering ............................................................................................ 268
Linear Programming ............................................................................................................... 269
Forecasting ...............................................................................................................................270
Simulation................................................................................................................................271
Conclusion .............................................................................................................................. 273
References ................................................................................................................................275
268 ◾ Michael L. Washington
is not new. Industrial engineers have been involved in improving the healthcare system since 1913
(Salvendy, 2001).
History of Industrial Engineering
In general, people have little knowledge of industrial engineers. erefore, in order to understand
the benets they can provide to the public health system, and the tools and techniques they use,
a description of an industrial engineer is needed. e operational denition of an industrial engi-
neer or industrial and systems engineer (ISE) is (Salvendy, 2001, 5):
An ISE is one who is concerned with the design, installation, and improvement of
integrated systems of people, material, information, equipment, and energy by draw-
ing up specialized knowledge and skills in the mathematical, physical, and social sci-
ences, together with the principles and methods of engineering analysis and design to
specify, predict, and evaluate the results to be obtained from such systems.
Traditionally, industrial engineers have concentrated on improving the work environment
in areas like manufacturing, transportation, and distribution. Ergonomics, the study of ways to
improve job satisfaction, employee health and safety, and job performance, is a specic industrial
engineering concentration that involves healthcare. Since the 1970s, more industrial engineers, or
individuals using industrial engineering–related tools, have entered the healthcare arena as con-
sultants and management engineers in hospitals and other healthcare settings (Benson and Harp
1994; Butler 1995; Davies 1994; Isken and Hancock 1998; Liyange and Gale 1995; Mahachek
1992; Saunders, Leblanc, and Makens, 1989; Tomar et al., 1998; Washington 1997; Whitson
1997).
Although most industrial engineers in healthcare commonly work in acute health-care set-
tings, their skills can and have been applied in public health systems.
In 1913, Frank Gilbreth used motion-study techniques to improve surgical procedures. Lillian
Gilbreth, Frank’s wife, published articles in the 1940s on the use of industrial engineering tools
in hospitals. e growing use of industrial engineering techniques in hospitals led to the develop-
ment of the Hospital Management Systems Society (now called the Healthcare Information and
Management Systems Society) in 1961, which started at the Georgia Institute of Technology in
Atlanta, Georgia. Also in Atlanta, the Institute of Industrial Engineering created a hospital section
in 1964 that eventually led to the birth of the Society for Health Systems (Salvendy 2001). e
Georgia Institute of Technology has the largest industrial engineering department in the world
and they developed the rst academic program in health systems in 1958 (Sainfort 2004).
e National Academy of Engineering and the Institute of Medicine state that barriers exist
in using some engineering tools (e.g., statistical process controls; queuing theory; quality function
deployment; failure-mode eects analysis; mathematical modeling; discrete event computer simu-
lation; linear, nonlinear, and mix-linear programming; neural networks; optimization techniques
(e.g., tabu and scatter search); market models; and agency theory) in healthcare. As stated in the
National Academy of Engineering and the Institute of Medicine report (2005, 3):
[R]elatively few health care professionals or administrators are equipped to think ana-
lytically about health care delivery as a system or to appreciate the relevance of sys-
tems-engineering tools. Even fewer are equipped to work with engineers to apply these
tools. e widespread use of systems-engineering tools will require determined eorts
Industrial Engineers in Public Health ◾ 269
on the part of health care providers, the engineering community, state and federal
governments, private insurers, large employers, and other stakeholders.
Researchers at the Centers for Disease Control and Prevention (CDC) have a history of col-
laborating with and hiring industrial engineers to solve some public health problems. is chapter
will discuss three historical projects
*
within the CDC utilizing industrial engineer skills. e stud-
ies are (1) a binary-integer linear program to schedule childhood vaccinations while minimizing
parental costs, (2) a public vaccine-need forecasting model, and (3) and a discrete-event computer
simulation of a sexually transmitted disease (STD) clinic to analyze its capacity to administer
more hepatitis B (HBV) vaccinations.
Linear Programming
e widespread use of vaccines has been one of the most successful public health interventions
of the twentieth century. Vaccination has resulted in the eradication of smallpox; elimination of
poliomyelitis in the Americas; and control of measles, rubella, tetanus, diphtheria, Haemophilus
inuenzae type b, and other infectious diseases in the United States and other parts of the world
(CDC 1999a).
Because of the biotechnology revolution, another problem developed as more childhood vac-
cines became available and the childhood vaccine schedule became more complex: determining
how to administer vaccines in the most economical manner, while limiting the number of injections
per visit. Jacobson et al. (1999) and Weniger et al. (1998) created a binary-integer linear program
model to determine the minimum cost of vaccinating a child against selective vaccine-preventable
diseases based upon a few constraints. Under the January to December 1998 Advisory Committee
on Immunization Practices (ACIP) Recommended Childhood Immunization Schedule for the
United States (Figure32.1), which was the current schedule at the time of the study, the possibil-
ity existed for a child to receive up to six vaccinations during one visit. Healthcare researchers and
practitioners had experienced diculty in determining the most economical clinical visit schedule
for parents and which vaccines to administer during a visit (Weniger et al. 1998).
e pilot project only included four vaccine manufacturers. Vaccines included in the model
were diphtheria-tetanus-acellular pertussis (DTaP), Haemophilus inuenzae type b (Hib), HBV,
and a combination vaccines of DTaP-Hib and Hib-HBV. Vaccines that were on the 1998 schedule
but not included in this pilot study were polio, measles-mumps-rubella, and varicella. e objec-
tive function of the model was to minimize the cost of administering the vaccinations, which only
included the cost of the vaccine, vaccine preparation, injection, and clinical visit. Constraints dur-
ing the pilot project were adhering to the 1998 ACIP Recommended Childhood Immunization
Schedule for the United States up to age ve, limiting injections per visit to no more than three,
and not permitting excess vaccinations (receiving more than was recommended).
Results from the pilot project found the minimum cost of vaccinating a child to be $490.32.
Results are shown in Table32.1 (Jacobson et al. 1999; Weniger et al. 1998). Other scenarios were
*
e historical examples presented in this chapter are intended for demonstration purposes only, in order to show
the utility of industrial engineering in public health. Interested readers are encouraged to visit http://www.cdc.
gov/hepatitis/B/index.htm and http://www.cdc.gov/vaccines/ for the most up-to-date information about hepa-
titis and vaccine-preventable diseases.
270 ◾ Michael L. Washington
considered: rst HBV vaccination given in the second month, second-lowest cost, maximum cost,
and all manufacturers represented in the vaccines being administered.
is project was a pilot study; however, the researchers conducted more research and devel-
oped a more detailed and complete working model on the Internet (http://vaccineselection.com).
e pilot project expanded to include other recommended vaccines. Although updates to this
website ended with the 2007 immunization schedule, it was the start of other industrial engineers
creating tools to assist in immunization scheduling, in which the tools used some form of linear
and nonlinear programming and the ACIP recommendations:
◾ Catch-Up Immunization schedulers (https://www.vacscheduler.org/) based on the 2013
Childhood and Adolescent Immunization Schedule
◾ Adolescent Immunization Scheduler (http://www.cdc.gov/vaccines/schedules/Schedulers/ado-
lescent-scheduler.html) based on the 2013 Recommended Adolescent Immunization Schedule
◾ Adult Immunization Scheduler (http://www.cdc.gov/vaccines/schedules/Schedulers/adult-
scheduler.html) based on the 2012 Recommended Adult Immunization Schedule
Forecasting
During the late 1990s and early 2000s, better vaccine forecasting was needed. Six Georgia Institute
of Technology industrial engineering students developed regression models for four immunization
programs to predict vaccine needs by using historical data and interviewing appropriate scien-
tists and administrators. Because this was a pilot project, the students only considered the most
Vaccine
Age
Birth
Range of Acceptable Age for Vaccination
Vaccines to Be Assessed and Administered if Necessary
1
Mo.
2
Mos.
4
Mos.
6
Mos.
12
Mos.
15
Mos.
18
Mos.
4–6
Yrs.
DTaP
or DTP
DTaP
or DTP
Hib
Polio Polio
Hib Hib
DTaP
or DTP
DTaP
or DTP
Polio
MMR
MMR
Var
11–12
Yrs.
14–16
Yrs.
Hepatitis B
†s
Hep B
5
Poliovirus
††
Diphtheria and
tetanus toxoids
and pertussis
1
Measles-mumps-
rubella
55
Varicella virus
11
Ha emophilus
influenzae
type b
**
Figure 32.1 ACIP Recommended Childhood Immunization Schedule, United States, January–
December. (From Centers for Disease Control and Prevention (CDC).) (1998). “Notice to Readers
Recommended Childhood Immunization Schedule: United States, 1998.” The Morbidity and
Mortality Weekly Report 47, no. 1 (1998): 8–12. Used by Weniger et al. in 1998 linear program-
ming paper.)
Industrial Engineers in Public Health ◾ 271
purchased childhood vaccines by doses for modeling. Vaccines selected for modeling were Hib,
HBV, measles-mumps-rubella (MMR), and DTaP. Each vaccine was modeled separately for each
program. Independent variables used in the models included modied estimates of numbers of
children less than one year of age, total vaccine purchased from the preceding year, and a deter-
ministic trend.
Compared with the forecasting methods used by the US CDC National Immunization
Program (now the National Center for Immunization and Respiratory Diseases) in 2000, the
regression models underestimated projects’ vaccine needs 20% less frequently, and overestimated
projects’ vaccine needs 26% less frequently (Happ et al. 2000). is project, based on limited data,
was a step in forecasting public vaccine needs.
Simulation
e CDC recommends those who seek evaluation of treatment for an STD be vaccinated against
HBV (CDC 2009). An STD clinic in California oered the rst of three HBV vaccines to those
who wanted it during the initial visit. Unfortunately, they had diculties getting clients to return
for their second and third doses. e clinic wanted to enhance a program to encourage these clients
to return and receive their nal two HBV vaccinations. As a pilot project, a study was conducted to
Table32.1 Linear Program Model Calculating the Minimum Cost of
Vaccinating a Child with a Hepatitis B Birth Dose
Visit
(months)
Vaccine
(manufacturer)
Total Vaccine
Price ($)
Injection
Cost ($)
Clinic Visit
Cost ($)
Total Cost
($)
0–1 HepB(B) 9.72 15.00 40.00 64.72
2 DTaP-Hib (A) 22.24 15.00 40.00 77.24
4 HepB (B),
DTaP-Hib (A)
31.96 30.00 40.00 101.96
6 DTaP-Hib 22.24 15.00 40.00 77.24
12–18 HepB (B),
DTaP-Hib(A)
31.96 30.00 40.00 101.96
60 DTaP (A) 12.20 15.00 40.00 67.20
Total Cost 130.32 120.00 240.00 490.32
Only 2 manufactures are represented (A and B).
Total vaccine prices used are from the US Federal contract purchase prices
(including excise taxes) effective as of September 4, 1997.
Source: Data from S. H. Jacobson, E. C. Sewell, R. Deuson, and G. B. Weniger, “An
Integer Programming Model for Vaccine Procurement and Delivery for
Childhood Immunization: A Pilot Study,” Health Care Management Science 2
(1999): 1–9; B. G. Weniger, R. T. Chen, S. H. Jacobson, E. C. Sewell, R. Deuson, J.
R. Livengood, and W. A. Orenstein, “Addressing the Challenges to
Immunization Practice with an Economic Algorithm for Vaccine Selection.”
Vaccine 16, no. 19 (1998): 1885–1897.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.