Computational Epidemiology From Disease Transmission
Modeling to Vaccination Decision Making
Visit the link below to download the full version of this book:
https://medipdf.com/product/computational-epidemiology-from-disease-transmission
-modeling-to-vaccination-decision-making/
Click Download Now
This book is dedicated to all the people
around the world, who came together to fight
against the novel coronavirus (COVID-19)
pandemic.
Jiming Liu and Shang Xia
Preface
What Can We Learn from COVID-19?
To see a World in a Grain of Sand
And a Heaven in a Wild Flower
Hold Infinity in the palm of your hand
And Eternity in an hour
William Blake (1757–1827) Auguries of Innocence
The title of this preface may look a bit unusual for a research monograph.
Nevertheless, this was indeed the kind of questions that came to our mind when
we were writing it.
This book was born in a special time. As it was being written and published, the
world was enduring one of its greatest challenges in decades, if not in centuries. The
novel coronavirus, known as COVID-19, had rapidly spread to around 200 countries
and territories in 6 continents (with only Antarctica untouched) within a few months,
resulting in more than 5 million people infected and over 300,000 deaths (as of
May 2020). All the people, no matter where they were and who they were, found
themselves caught right amid this most unprecedented global crisis, with devastating
casualties, country lockdowns, service/business shutdowns, and possible economic
meltdown.
The world is truly in a state of emergency, a time of great uncertainty and anxiety.
Yet, as in the history of human civilization, we all should be hopeful that
humankind will be able to learn and prevail in the end. There will be no exception
this time. One of the important lessons that we can probably learn from the
Mother Nature in this global fight against COVID-19 is that only by being
united as one, as humanity, working together to remove the barriers of races,
vii
viii Preface
nation-states, political ideologies, religions, and special interests, and coexisting
harmoniously in an increasingly interconnected and interdependent world, can
human beings be saved. This also calls upon scientists to rethink their roles and
social responsibilities, to rediscover the world, and to advance sciences beyond
the usual disciplinary boundaries. Under such a unique circumstance, the theme
of this book becomes particularly appropriate, as it attempts to show how disciplines
such as computer science, systems science, and epidemiology can converge and
address some of the most pressing, socially relevant issues in eradicating diseases.
The contents presented in this book reflect part of our ongoing initiatives at Hong
Kong Baptist University (HKBU), which are aimed to address several important
problems in infectious disease epidemiology and to solve them in a systematic
way through the developed computational models, methods, tools, and case studies.
Some examples of the problems are as follows:
• How has the field of epidemiology evolved (Chap. 1)? How can data-centric
technologies be incorporated? (Chaps. 1 and 7)
• How can the heterogeneous nature of disease transmission be modeled and
characterized? (Chap. 2)
• How can we strategically plan and achieve disease interventions (Chap. 3)?
• How can we take into consideration the human (individual and social) aspects of
decision-making in disease interventions? (Chaps. 4–6)
• How can the epidemiological challenges be best addressed from a systems
perspective? (Chap. 7)
• What promises does systems epidemiology hold? What is the best way to pursue
it? (Chap. 7)
Solutions to the above problems can help governments, public health policy-
makers, scientists, and front-line practitioners in seeing the current and future global
health challenges, such as COVID-19, from a systematic, data-driven computational
modeling perspective, and hence developing the corresponding effective interven-
tion strategies. For instance, the solutions provided in this book can help respond
to the following questions in the case of COVID-19: Once a coronavirus vaccine
becomes available, what will be the best (scientifically sound and yet practically
acceptable) way to administer the limited supplies? Who will have the priorities?
Will there be enough people to take the vaccine, so that the target coverage (herd
immunity) can be achieved? How will people make their vaccination decisions?
The book is intended to serve as a reference book for researchers and practition-
ers in the fields of computer science and epidemiology, who may read Chaps. 1 and 7
of the book first, to gain a holistic view of the domain, prior to reading Chaps. 2–6
for further studies on the specific problems and issues involved.
Together with the provided references for the key concepts, methods, and
examples being introduced, the book can readily be adopted as an introductory text
for undergraduate and graduate courses in computational epidemiology as well as
systems epidemiology and as training materials for practitioners and field workers,
Preface ix
who may study the book in the regular order of Chaps. 1–7 and then revisit Chaps. 2–
6 to extend some of the topics and problems.
Hong Kong Jiming Liu
Hong Kong Shang Xia
May 2020
Acknowledgements
Jiming Liu is extremely grateful for the rigorous foundation development in
Physics that he acquired from East China Normal University in Shanghai in the
late 1970s and early 1980s and for the mind-opening education and enriched
inquiries in Philosophy, Cybernetics, and Psychology that he gained through the
most inspirational teachings of David Mitchell, Gary Boyd, and Gordon Past, as
well as other thinkers and visionaries, from Concordia University in Montreal in
the mid-1980s. Both periods have profoundly impacted him throughout his career
and life. He would like to acknowledge the amazing collegiality and friendships that
he has enjoyed in more than three decades from many of his mentors, colleagues,
collaborators, and students in Montreal before 1994, in Windsor in 2006–2007, and
in Hong Kong since 1994, who have not only accompanied, but also enlightened,
him throughout his odyssey of intellectual discovery, exploration, and wonder. For
the past ten years, he has made special efforts in developing solutions to address
real-world problems, such as global health and infectious disease epidemiology
in particular, from the novel perspectives of complex systems, network science,
machine learning, and autonomy-oriented computing. For this and other rewarding
journeys, he would like to express his heartfelt gratitude to: Xiao-Nong Zhou (as
well as dedicated colleagues) of National Institute of Parasitic Diseases (NIPD) at
Chinese Center for Disease Control and Prevention (China CDC), with whom he
co-established the Joint Research Laboratory for Intelligent Disease Surveillance
and Control; his long-time colleagues as well as collaborators and supporters at
Hong Kong Baptist University (HKBU), William Cheung, Pong Chi Yuen, Yiu-
ming Cheung, Yang Liu, among so many others; his previous postdoctoral fellows
and research collaborators, Bo Yang, Zhiwen Yu, Xiaofeng Xie, Qing Cai, Zhanwei
Du, etc.; his earlier research students, Shang Xia, Benyun Shi, Chao Gao, Li Tao,
Xiaolong Jin, Hongbing Pei, Hechang Chen, Xiaofei Yang, Shiwu Zhang, Hongjun
Qiu, Jianbing Wu, Qi Tan, Jinfu Ren, and many more. Also, he would like to thank
HKBU as a whole for the trust and opportunities to shape and contribute to the
university environment in the capacities of Chair Professor in Computer Science,
Head of Computer Science Department, Associate Dean (Research) of Faculty
of Science, Dean of Faculty of Science, and Associate Vice-President (Research)
xi
xii Acknowledgements
of the university, to make it the most conducive place for scholarship. He would
like to thank Hong Kong Research Grants Council (RGC) for the funding support
over the years; a number of grants have been awarded to specifically support his
team’s research on understanding and solving epidemiological problems through the
exciting routes of computer science, machine learning, and artificial intelligence.
Last but the foremost, he would like to express his deepest thanks to his wife
M.L. and his daughters I.Y.Y. and B.Y.X. for their long-lasting love and the most
wonderful time.
Shang Xia would like to express his sincere gratitude to Prof. Jiming Liu for his
enlightening, patience, motivation, enthusiasm, and profound knowledge. Without
his encouragement and persistence, this book could not be accomplished. He would
like to express his sincere gratefulness to Computer Science Department at Hong
Kong Baptist University (HKBU), where he acquired his PhD degree, benefited a lot
from the most inspirational guidance, and enjoyed a fulfilling campus life. For this
rewarding journey in Hong Kong, he would sincerely express his heartfelt gratitude
to Dr. Benyun Shi, Dr. Li Tao, and Dr. Yang Liu, from whom he benefited their
collaboration and support. The sincere thanks also go to Prof. Xiao-Nong Zhou and
the National Institute of Parasitic Diseases at Chinese Center for Disease Control
and Prevention for the great support for his academic career and research. Last but
not least, he would like to thank his family: his wife Yao Q.Q. and his daughters
Yoyo and Xiuxiu for their caring, love, and support in this wonderful life.
Both authors wish to express their special thanks to Dr. Yang Liu for his great
efforts in proofreading the manuscript and offering excellent editorial suggestions
and help.
Hong Kong Jiming Liu
Hong Kong Shang Xia
May 2020
Contents
1 Paradigms in Epidemiology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Methodological Paradigms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Recent Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Infectious Diseases and Vaccination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 Objectives and Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.4.1 Modeling Infectious Disease Dynamics . . . . . . . . . . . . . . . . . . . . . . . 8
1.4.2 Modeling Vaccine Allocation Strategies. . . . . . . . . . . . . . . . . . . . . . . 8
1.4.3 Modeling Vaccination Decision-Making . . . . . . . . . . . . . . . . . . . . . . 9
1.4.4 Modeling Subjective Perception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2 Computational Modeling in a Nutshell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.1 Modeling Infectious Disease Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.1.1 Infectious Disease Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1.2 Age-Specific Disease Transmissions . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2 Modeling Contact Relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.2.1 Empirical Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.2.2 Computational Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3 Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.3.1 Hong Kong H1N1 Influenza Epidemic . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.2 Age-Specific Contact Matrices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3.3 Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.4 Further Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3 Strategizing Vaccine Allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.1 Vaccination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.1.1 Herd Immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.1.2 Vaccine Allocation Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Vaccination Priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.3 Age-Specific Intervention Priorities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.3.1 Modeling Prioritized Interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
xiii
xiv Contents
3.3.2 Effects of Vaccination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.3.3 Effects of Contact Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.3.4 Integrated Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.4 Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.4.1 Hong Kong HSI Vaccination Programme . . . . . . . . . . . . . . . . . . . . . 42
3.4.2 Effects of Prioritized Interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.5 Further Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4 Explaining Individuals’ Vaccination Decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.1 Costs and Benefits for Decision-Making. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
4.2 Game-Theoretic Modeling of Vaccination Decision-Making . . . . . . . . . 51
4.3 Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.3.1 Hong Kong HSI Vaccination Programme . . . . . . . . . . . . . . . . . . . . . 53
4.3.2 Vaccination Coverage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.4 Further Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5 Characterizing Socially Influenced Vaccination Decisions . . . . . . . . . . . . . . 57
5.1 Social Influences on Vaccination Decision-Making . . . . . . . . . . . . . . . . . . . 57
5.2 Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.2.1 Vaccination Coverage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.3 Further Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6 Understanding the Effect of Social Media. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6.1 Modeling Subjective Perception. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6.2 Subjective Perception in Vaccination Decision-Making . . . . . . . . . . . . . . 74
6.2.1 Dempster-Shafer Theory (DST) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
6.2.2 Spread of Social Awareness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
6.3 Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
6.3.1 Vaccination Decision-Making in an Online Community . . . . . 78
6.3.2 Interplay of Two Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
6.4 Further Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
7 Welcome to the Era of Systems Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
7.1 Systems Thinking in Epidemiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
7.2 Systems Modeling in Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
7.3 Systems Modeling in Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
7.4 Toward Systems Epidemiology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Abbreviations
ACIP Advisory Committee on Immunization Policy
AEFI Adverse events following immunization
BPA Basic probability assignment
CHP Centre for Health Protection
DST Dempster–Shafer theory
H1N1 Influenza A virus (H1N1)
H7N9 Avian influenza A virus (H7N9)
HSI Human swine influenza
HSIVP Human Swine Influenza Vaccination Programme
MMR Measles–mumps–rubella
NVAC National Vaccine Advisory Committee
SARS Severe acute respiratory syndrome
SEIR Susceptible–exposed–infectious–recovered
SIR Susceptible–infectious–recovered
SIS Susceptible–infectious–susceptible
SIT Social impact theory
STD Sexually transmitted disease
xv
Notation
S Population in susceptible compartment
I Population in infectious compartment
R Population in recovered compartment
N Overall population
Si Susceptible subpopulation i
Ii Infectious subpopulation i
Ri Recovered subpopulation i
Ni Overall subpopulation i
α Infectivity
β Susceptibility
λ Infection rate
μ Transmission rate
γ Recovery rate
cij Contact frequency between two subpopulations i and j
R0 Basic reproduction number
Rt Effective reproduction number
CH Contact matrix for household setting
CS Contact matrix for school setting
CW Contact matrix for workplace setting
CG Contact matrix for general community setting
C Contact matrix for overall social setting
Φ Social settings (H, S, W, G)
rH Household contact coefficient
rS School contact coefficient
rW Workplace contact coefficient
rG General community contact coefficient
K Disease reproduction matrix or next-generation matrix
A Infectivity matrix, diag (α1 , . . . , αN )
B Susceptibility matrix, diag (β1 , . . . , βN )
S Susceptible population matrix, diag (S1 , . . . , SN )
I Infectious population vector, [I1 . . . IN ]T
xvii
xviii Notation
ρ(K) Top eigenvalue of K
x1 Top left eigenvector of K
y1 Top right eigenvector of K
Nivac Number of vaccinated neighbors
Ninon Number of unvaccinated neighbors
wij Social closeness between two connected individuals i and j
λ̂i Perceived infection rate for subpopulation i
β̂ Perceived susceptibility
θ Herd immunity threshold
ζ Cost of disease infection
ξ Cost of vaccination
rc Cost ratio rc = ξ/ζ
σi Vaccination decision
σˆi Cost-minimized choice
σ˜i Social opinion from connected neighbors
ιvac
i Social influence for vaccination
ιnon
i Social influence against vaccination
ιi Influence discrepancy
ν Responsiveness to influence discrepancy in Fermi function
P (ιi ) Probability generated from Fermi function
rf Conformity rate
G Social network, G = V , L
V Network nodes (individuals)
L Network links (interactions)
Θ Universal set of vaccination decision, {Yes, No}
φ Empty set
2Θ Power set, {φ, {Y es}, {No}, Θ}
m(·) Basic probability assignment
m(Y es) Belief value of vaccination
m(No) Belief value of non-vaccination
m(Θ) Belief value of no decision (uncertainty)
mi Set of belief values
mi Updated belief values
mei Obtained awareness about negative events
medis Belief values generated by a severe disease infection
mevac Belief values generated by a vaccine adverse event
f Coefficient of awareness fading
Reporting rate of severe disease infections
κ Reporting rate of vaccine adverse effects
Chapter 1
Paradigms in Epidemiology
Epidemiology deals with “the study of the occurrence and distribution of health-
related states or events in specified populations, including the study of the deter-
minants influencing such states, and the application of this knowledge to control
the health problems” [1]. As defined by MacMahon et al. [2], epidemiology is
interdisciplinary by nature, concerning the sciences of etiology, genetics, biology,
pharmacy, geography, ecology, as well sociology and human behavior. Epidemio-
logical studies motivated by combating infectious diseases mainly focus on four
aspects of challenges, as follows: (1) pattern analysis, by investigating the spatio-
temporal distributions of the observed disease occurrences; (2) causal inference, by
identifying and evaluating associated impact factors; (3) forecasting and prediction,
by evaluating the dynamics of infectious diseases with reference to different sce-
narios; and (4) policy analytics, by exploring and conducting effective intervention
measures.
Toward these ends, the pioneers in epidemiology have provided much useful
knowledge to guide efforts in infectious disease control. As pointed out by Merrill
[3], epidemiology has evolved from supernatural practices to research based on
scientific foundations, from ad hoc reports to systematic investigations of public
health events and problems, from ignorance of the causes of diseases to a scientific
understanding of their hidden factors, determinants, and outcomes, and from lacking
feasible means for solving public health problems to having effective approaches to
disease intervention.
Developmental milestones in infectious disease epidemiology can be dated
back to the work of Hippocrates (460–377 BC), who examined the influence of
environments and attempted to explain how diseases transmit and cause infection
in a group of host individuals [3]. Other early studies include the work of John
Graunt (1620–1674), who described disease mortality rates by applying statistical
and census methods [4], and Thomas Sydenham (1624–1689), who studied disease
distribution patterns, moving from an observational to an analytical perspective [5].
In the nineteenth century, John Snow (1813–1858) traced the sources of disease
© Springer Nature Switzerland AG 2020 1
J. Liu, S. Xia, Computational Epidemiology, Health Information Science,
https://doi.org/10.1007/978-3-030-52109-7_1
2 1 Paradigms in Epidemiology
outbreaks (e.g., cholera in Soho, London, in 1854) and thereafter pointed out the
associations of disease outbreaks with social and natural environments [6]. To
more formally describe the dynamics of disease transmission, Ross (in 1911) and
MacDonald (in 1957) developed a set of mathematical equations and proposed
a threshold indicator, named the basic reproduction number, to quantitatively
characterize the extent of disease transmission [7].
1.1 Methodological Paradigms
Various methodologies have been developed to address a wide range of challenges
in infectious disease control and prevention, and these methods have been applied in
epidemiological studies in the past several decades. As stated by Zadoks [8], based
on the observation of disease occurrences, descriptive methods, such as clustering
and hot spot analysis, have been used to analyze the patterns of infectious diseases
in terms of their temporal, spatial, and demographic distributions in a population,
i.e., to answer the questions of when, where, and who. Statistical methods, such
as regression or Bayesian inference, can be used to further explore the causal
relationships between disease occurrence and the possible impact factors, i.e., to
answer the questions of why and how. Predictive methods, such as mathematical
modeling or computer-based simulation, have been developed to forecast the
dynamics of infectious diseases during an epidemic, and identify the most suitable
indicators for representing such a dynamic process. Based on these, prescriptive
methods, such as optimization, or scenario and sensitivity analyses, can be used by
public health authorities to decide how to implement the most effective intervention
strategies, such as the allocation of pharmaceutical resources (e.g., vaccines and
antivirals) and social distancing (e.g., segregation and school closures).
Infectious disease epidemiology has undergone a number of methodological
paradigm shifts throughout its development, as highlighted in Fig. 1.1. The typical
methods mentioned in the preceding paragraph, i.e., descriptive, predictive, and pre-
scriptive methods, correspond to three of those paradigms (the fourth is introduced
in the next subsection and discussed in detail in the final chapter of this book).
These three paradigms are (1) empirical investigation, (2) theoretical modeling, and
(3) computational modeling. Accordingly, we refer to the epidemiological method-
ologies based on these paradigms as (1) empirical epidemiology, (2) theoretical
epidemiology, and (3) computational epidemiology, respectively.
• Empirical Methods
The paradigm of empirical observation and investigation is well suited to the
early stage of epidemiological studies. As mentioned by Rothman et al. [9], it
typically involves (1) collecting observational data about disease transmission,
i.e., when, where, and who, and associated impact factors, e.g., the characteristics
of disease pathogens and host individuals at the microscopic scale, and of etiolog-
ical and meteorological environments at the macroscopic scale; (2) qualitatively
describing or quantitatively analyzing observational data to establish associative
